NSFH2/February 95 APPENDIX A - MEDICAL CONDITION CODES These codes are used with variables: Main: ME3, ME4, ME6, ME7, ME10, ME11, ME14, ME15, ME36, MI171 Parent: PE2, PE3, PE5, PE6, PE9, PE10, PE13, PE14 Proxy: ZZ12, ZZ32T01-ZZ32T03 Acute Conditions 101 Common cold 102 Other acute upper respiratory conditions: sore throat, laryngitis, "sinus trouble" (see also 206), sinus headache, U.R.I., vocal cord difficulties, pharynigitis, coryza, cough, mild throat infection. Strep throat, serious throat infections go in 111. 103 Pneumonia, pneumonitis, "lung infection" (acute). 104 Lower respiratory conditions: acute bronchitis, "chest cold", pleuritis, pleurisy (see also 209). "Flu" goes in 106. 105 Influenza with digestive manifestations (stomach "flu", acute enteritis), intestinal virus, 24 hour virus 106 Influenza ("the flu", grippe, "three-day virus"). 107 Virus, not codable in 101 to 106. 108 Mild infectious diseases common to children: chicken pox (varicella), German measles (roseola, rubella), measles (rubeola), mumps. 109 More serious infectious diseases common to children: whooping cough, croup, scarlet fever, rheumatic fever. 110 Mild infective and parasitic disease: herpes, thrush, fungus, mononucleosis, cold sore, swollen gland, lymphadenopathy, crabs, scabies, "sore in mouth", ulcers in throat, water blisters on hands, feet, etc., "mononucleosis hepatitis", ringworm. 111 More serious infections and parasitic diseases: ulcers in the mouth or throat, strep throat, shingles, herpes zoster, syphilis ("bad blood"), staph infection, trench mouth, "blood poisoning" (see also 217, 218), typhoid fever, more serious throat infections. 112 Severe infective and parasitic diseases: meningitis, viral meningitis, lock jaw (tetanus), histoplasmosis, infection in blood stream; (see also 212, 218). 113 Localized tissue infections: infected toe, boils, car-uncles, impetigo, erysipelas, felon, mastitis, abscesses, "sore on chest that wouldn't heal", "ingrown pimple", "acute inflammatory pressure". 114 Acute diseases of the eye: pink eye, conjunctivitis, cold in the eye, stye, hyphemia, hemorrhage in eye, blood clot in eye, clogged tear gland (duct). 115 Acute conditions of the ear: otitis media, ear infection, punctured ear drum, mastoiditis, myringitis, swimmer's ear, earache. 116 Tonsillitis and adenoiditis, trouble with tonsils. 117 Appendicitis. 118 Acute cholecystitis, gall bladder attack, pain in side relating to gall bladder. Gall bladder trouble, if not acute, goes in 232. 119 Acute digestive system disorders: functional and symptomatic digestive system conditions, stomachache, vomiting, nausea, diarrhea, food poisoning, nervous stomach (see also 231), gastroenteritis, peritonitis, colic, colicky babies (feeding problems), acute constipation, stomach cramps, abdominal pains, amoebic dysentery,. Diarrhea for babies under one year goes in 120. 120 Diarrhea for babies under a year old. 121 Headache. Migraine headache goes in 222. 122 Pelvic infection: acute "female trouble", gonorrhea (GC) and VD (unspecified) in females, salpingitis, acute vaginal bleeding, acute "female trouble", infected uterus (see also 233). 123 Acute genitourinary disorders: kidney infection, bladder infection, (see also 234), orchitis, cystitis, pyelitis, gonorrhea (GC) and VD (unspecified) in males, other urethritis, polyuresis, hypouresis, "kidney injury", inflamed testicle. 124 Acute, non-allergic or non-infectious conditions of the skin: ingrown toenail, corn, unspecified rashes, diaper rash, sunburn, hair loss, morbilliform rash, callosity, "low tide itch", rectal itch, itchy scalp, nervous skin rash, nervous dermatitis. 125 Burns. 126 Acute circulatory conditions: thrombosis, phlebitis (see also 217), blood clot, pulmonary embolism, berry aneurysm, "brain explosion", "blocked artery", langiagra. Broken blood vessel goes in 134. 127 Acute mild vertebral and lumbar disorder: lumbago, stiff neck-- no other information given, cold or pain or cramp in back, backache, shoulder pains. 128 Acute, more serious musculoskeletal disorders: fibrositis, myositis, neuritis, severe back pain, swelling of limb, back or joint, sciatica, slipped disc, knee drainage, pinched nerve, muscle spasm, back operation-- unspecified, water on the knee, tendinitis, torticollis, DeQuervain's disease, strained back, pulled/torn muscle in back, sacroiliac strain, ruptured disc, infection in joint, hemorrhage into ankle or knee joint, leg cramp in young person. Leg cramp in older person goes in 217. 129 Fractures and dislocations: bone chips, hairline fractures, dislocated vertebra. Any time the doctor put a cast on it should be coded here. 130 Connective tissue disorder: sprains and strains, whiplash, torn ligaments, knee injury-- unspecified, pulled muscle, hurt knee cap, "swollen finger", sore knee, "tore lining from ribs", "stubbed toe". 131 Open wounds and lacerations, mild: no mention of stitches or a few stitches required, cuts on face or genitals requiring no stitches. 132 Open wounds and lacerations, more serious: stitches on face or genitals, a large number of stitches required, blood transfusions, plastic surgery, or debridement of necrotic tissue required. 133 Penetrating injuries: stab wounds, gunshot wounds, stepped on nail, dog bite, puncture wound. 134 Superficial injuries: contusions, bruises, abrasions, nosebleed. 135 Complications of medical or surgical procedures, drug overdoses, poisonings: overdoses, both accidental and deliberate, drug reactions, penicillin reaction, over-sedated with tranquilizers, other kinds of poisonings (exposure to paint fumes, maced in demonstration), complications of medical procedures, BCP/IUD complications. Food poisoning goes in 119. 136 Surgical emergencies: intussusception, ruptured spleen, internal injuries from auto accidents, Mallory-Weiss syndrome, obstruction to the common iliac vein due to injury, emergencies involving digestive system, emergencies involving reproductive system (e.g., ruptured ovarian cyst), collapsed lung, acute pancreatitis, "ruptured cystic ovary." 137 Unconscious, concussion, fainted, passed out (no further description of condition which caused symptom). 138 Tiredness or pains from overwork, fatigue or tiredness that apparently only lasted a day or so (no further explanation or description of condition). 139 Hangover 198 Residual acute conditions: pain, apparently acute with no diagnosis, dehydration, sunstroke, febrile convulsions (see also 226), fever of unknown origin, choking, surgery on finger, "didn't have any appetite," "couldn't see," "fell on back," "near drowning," "the bends," "cramps from skin diving," "passing blood, find no cause." Chronic Conditions 201 Cancer: malignant neoplasm, leukemia, Hodgkins disease. 202 Unspecified tumor: "had a biopsy," "afraid I had cancer," lump, tumor, growth, skin tumor behind ear drum (not 238), "lump in breast." 203 Benign tumor, mild problem (not much discomfort): mole, sebaceous cyst, warts, hemangioma, birth mark, "cyst in roof of mouth," "cyst on nose," ganglion cyst, "polyp in eye," "small growth on the white of my eyeball," "skin lesions." If removed on outpatient basis and if no clear indication that it should be coded below in 204, then code here. Pilonidal cyst goes in 233. 204 Benign tumor, more serious (problem or worry producing): biopsies, no malignancy found, fibroids, fibrocystic disease of breast, polyps, "polyp in colon," cystic mastitis, ovarian cystic meningiomas, "breast tumor." 205 Asthma, bronchial asthma. Asthmatic bronchitis goes in 206. 206 Chronic sinusitis or bronchitis: bronchiectasis, sinus condition (see also 102), asthmatic bronchitis, "recurring pulmonary infection," deviated septum. 207 Tuberculosis: active and inactive, arrested, reactive TB test. 208 Other chronic infective diseases: trichinosis, worms, brucellosis, malaria, undulant fever, tularemia. 209 Chronic diseases of the respiratory tract not mentioned above: emphysema, shortness of breath, collapsed lung, cystic fibrosis, histoplasmosis, atelectasis, black lung, pleurisy (chronic), pleural effusion--not otherwise specified. Hold fluid in the lungs for medical consultant. Pulmonary edema is heart failure and goes in 212. Shortness of breath without other information goes in 298. 210 Heart attack, M.I. (myocardial infarction), "coronary." See other codes for additional conditions respondent may cite as resulting from this condition, e.g., coded 212. 211 Heart trouble, mild: patient thought there was something wrong with his heart but doctor found nothing wrong, functional heart condition, chest pain, functional heart murmurs, other very minor cardiac abnormalities, problems with cholesterol, arrhythmias, palpitations. 212 Heart trouble, more serious: serious structural abnormalities (aortic aneurysm, valve lesions, aortic stenosis, mitral insufficiency); cardiac symptoms after a myocardial infarction (heart muscle damage, coronary insufficiency); post-coronary thrombosis; cardiac symptoms after rheumatic fever, "chronic rheumatic heart"; angina--no mention of normal findings or no mention that it followed a heart attack; coronary spasm; heart infection; "heart trouble," "bad heart," "heart condition"--no further elaboration; cardiovascular disease--no further elaboration; enlarged heart' minor structural abnormalities (congenital heart disease, no further elaboration); heart failure (big heart and shortness of breath). 213 Stroke (cerebrovascular accident (CVA), thrombosis of cerebral artery), large intracerebral hematoma. See other codes for additional conditions respondent may cite as resulting form this condition. 214 High blood pressure (hypertensive disease, hypertension ), "blood pressure." 215 Varicose veins (varicocele). 216 Hemorrhoids ("piles"), rectal condition - not otherwise specified, fissurectomy, fistula. 217 Chronic diseases of the circulatory system: arteriosclerosis, "hardening of the arteries," "calcium deposits in the arteries," chronic phlebitis, (see also 126), other chronic diseases of the circulatory system, (fainting, dizziness, or leg cramps in older person). If fainting or dizziness in person over 45, see 137, 255, 298. If leg cramps in person under 45, see 128. Low blood pressure goes in 298. 218 Blood disorders: anemia, low blood count, thick blood, hemophilia, sickle cell anemia, borderline bleeder; see also 111, 112. Leukemia is cancer and goes in 201. "Bad blood" is syphilis and goes in 111. "Infection in blood stream" goes in 112. "Bruises easily," see 298. 219 Diseases of the thyroid gland: thyroid trouble, goiter, hypothyroid, thyroid adenoma, Graves' Disease, myxedema; see also 227. 220 Diabetes (diabetes mellitus, "sugar," hyperglycemia). Diabetes insipidus goes in 227. 221 Allergies: hayfever, poison ivy, allergic rhinitis, contact dermatitis, angioneurotic edema (angioneuredema), urticaria, hives. 222 Migraine headache (including allergic migraine). Headache goes in 121. 223 Chronic skin disease: eczema, keratosis, psoriasis, lupus-- not systemic, skin ulcer, acne, bed sores, pseudoxanthoma elasticum, alopecia, pilonidal sinus or cyst, seborrhea, "leg ulcer," "skin eruption"; see also 124, 203. 224 Arthritis and rheumatism, "congenital hips," (double code with 241). 225 Other chronic musculoskeletal disorder: bursitis, "back trouble," vertebral disc disease, bunion, hallux valgus, knock-kneed, "baby's feet need corrective shoes" (see also 241), flat feet, "feet turn in," bone spur, Osgood-Schlatter disease, Scheuermann's disease, chondromalacia patella, osteochondritis dissecans, chronic backache, "calcium buildup under food," "pain in bottom of foot," "nerve injuries in cervical spine caused from fall," chronic slipped disc, chronic pinched nerve. 226 Seizures: epilepsy, epileptic fits, "fits" or seizures of unknown origin. Include febrile convulsions either associated with another chronic condition or with evidence that they are a recurring condition. Febrile convulsions, no further elaboration, go in 198. 227 Other metabolic or endocrinologic diseases, glandular disturbances: gout, (uric acid in blood), porphyria, Addison's disease, diabetes insipidus; see also 219. Diabetes unspecified or mellitus goes in Code 220. 228 Hernia ("rupture"): femoral hernia, hydrocele in a child, Hydrocele in an adult goes in 234. Hiatus hernia goes in 229. 229 Peptic ulcer (ulcer of the stomach or duodenum). Hiatus hernia (ruptured esophagus) goes here. Skin ulcer goes in 223. 230 Other serious chronic diseases of the digestive system: pyloric stenosis, diverticulitis, yellow jaundice, hepatitis, cirrhosis of the liver, ulcerative colitis, bleeding colon, pancreatitis (infected pancreas), stomach operation--unspecified. All colitis in hospital goes here. 231 Mild or functional diseases of the digestive system: "nervous stomach," constipation, nervous diarrhea, intestinal disorders, incipient ulcer, colitis, intestinal blockage, "twisted colon", spastic colon, "peptic stomach"; see also 119. 233 Chronic gynecologic conditions: "female trouble"-- chronic "cramps", menopausal symptoms, hot flashes, hysterectomy -- cause unspecified, cervicitis, "suspicious" Pap test, trouble with periods, structural problems, enlarged ovary, vaginal discharge, vaginitis, painful intercourse. Notice that all pelvic infections will go into 112 even if they have been going on for a long time. Gonorrhea (GC) is females goes in 122. 234 Chronic diseases of the urinary system and male genital system: kidney or bladder stone, nephrosis, nephritis, uremia, prostate trouble, undescended testicle, cryptorchidism, cystocele, urinary problems secondary to chronic condition, "urethral carbuncle," "fallen bladder:, chronic pyelitis, hydrocele in adult; see also 123. Hydrocele in child goes in 228. 235 Blindness (one or both eyes). 236 Visual impairments, mild: refractive errors with no other eye problem mentioned (myopia, hyperopia, presbyopia, astigmatism). Place here also "normal eye exam" when no refractive error was found but no other eye problem was mentioned either in symptoms or diagnosis. Also, glasses broken, contact lens prescribed, check-up when doctor is obviously an ophthalmologist. 237 Visual impairments, more serious: glaucoma, "almost blind", eye trouble, weak eye muscles, cross eyed, cataracts, diabetic retinopathy, eye injuries. If blindness resulted double code with 235. 238 Chronic difficulties with ears or hearing: deaf, hearing impairments, hard of hearing, too much ear wax, surgery to construct a new ear drum. Meniere's syndrome, labyrinthitis--inner ear dizziness, fluid in middle ear, chronic otitis, condition requiring mastoidectomy, "inner ear ringing," "inner ear problem," chronic ear infection, vertigo. 239 Paralysis: paraplegic, hemiplegic, quadriplegic, paralyzed in accident. 240 Amputations. 241 Other impairments including congenital malformations: incontinent osteogenesis imperfecta, after effects of polio - not further specified, water head baby, cleft palate, after effects of brain surgery -- not further specified, kyphosis, microcephaly (double code with 253), "congenital hips" (double code with 224), "cyst on spine -- born with it." 242 Serious chronic neurological disorders: multiple sclerosis, muscular dystrophy, Parkinson's disease, cerebral palsy, Bell's palsy, tic douloureux, Huntington's chorea, neurological conditions of questionable etiology. 243 Weight problems: obesity, underweight, nutritional deficiencies. 244 Phimosis, other reasons for needing circumcision. Do not code new baby in hospital here or anywhere. 245 Edema, dropsy (fluid accumulating in body). Pulmonary edema goes in 212. If any mention of heart trouble put in 212. If any mention of kidney trouble, put in 234. 247 Osteomyelitis (inflammation of the hip socket). 248 Collagen diseases: systemic lupus erythematosis (SLE), polyarteritis nodosa, dermatomyositis, scleroderma, polymyositis, Raynaud's disease. 249 Pregnancy and termination of pregnancy, other complications of pregnancy, miscarriage, delivery. Chronic Mental Conditions (Condition Expected to Last More than 3 Months) 250 Minor psychological problems: nervousness given either as a symptom or a diagnosis, nervous exhaustion, nervous breakdown; insomnia, nightmares; problems of childhood (bedwetting (enuresis), hyperkinetic child, "holding breath"); no specific diagnosis or symptoms mentioned but indication that tranquilizers or sleeping pills have been prescribed (the most common tranquilizers are Librium, Valium, Miltown (Meprobamate); the most common sleeping pills are Seconal, Nembutal, Doriden); hyperventilation syndrome; depressed, unhappy; social problems such as "fight with boyfriend," accident prone, family problems, marital problems with no sexual difficulty mentioned. Do no code 250 if respondent mentions worry or anxiety over an illness unless it is clear that anxiety is the problem, rather than the illness. 251 Intermediate psychiatric problems: this category is difficult in some cases to distinguish from 250 but, in general, it is meant to represent more long standing conditions or more specific or sophisticated psychiatric diagnoses than are used in 250: anxiety, anxiety reaction, no further qualification; any neurosis mentioned by name (hysteria, obsessive-compulsive, mixed psychoneurosis, phobic anxiety neurosis, globus hystericus, psychasthenia, neurasthenia); psychosomatic conditions, sexual problems, underlying physical cause not given; neurotic, no further qualification; attempted suicide' endogenous depression, depressive neurosis; involutional reaction. 252 Mental illness: schizophrenia (acute schizophrenic reaction, other mention of schizophrenia, schizophrenia in remission, latent schizophrenia); psychosis other than schizophrenia (psychotic manic-depressive, borderline psychotic); neurotic reaction of such intensity as to require hospitalization (acute anxiety attack, panic, hyper- ventilation syndrome, hysterical, acute fugue state)--Note reaction or neurosis here; otherwise, it goes in 250 or 251; problems with alcohol, alcoholic; problems with drugs, drug addiction that requires hospitalization; senility, chronic or acute brain syndrome (in hospital); depressions serious enough to be hospitalized; "mental illness," "emotional disorder," other vague category attended by psychiatrist in hospital but no further diagnosis given. Acute Mental Conditions 253 Mental retardation, brain damage: microcephaly, (double code with 241), speech pathology and persevarative speech, hydrocephalic, neurological immaturity, "slow learner," "slow starter." 254 Senility (not in mental hospital). Hospitalized senility goes in 252. 255 Dizzy, (no diagnoses). See also 217. 298 Residual chronic condition: weakness, tiredness, fatigue, "low blood pressure," low blood sugar, hypoglycemia, aches and pains, "old age," "growing pains," "shortness of breath." NOTE: These symptoms account for the bulk of this category. Also included are a few miscellaneous cases: "not able to perspire," "breaks blood vessels easily," "a constant numbness in left arm," "he runs a fever and gets delirious... gradually growing out of it...," "I went for a check up and the doctor kept me 5 days...," "exhaustion and breathing problems," "cortisone shot," " checked for diabetes and blood pressure and blood test...," "hospitalized and they couldn't find anything," "nauseated, weak and sick...nothing showed up," "I wasn't feeling well," "I have so many things wrong," "went to get some water pills," " went for x-rays." 301 Check up in absence of symptoms (preventive care): routine check-up; gynecological check up; well baby care; pre-employment physical; physical for school, or camp; premarital exam; exam for insurance company, etc. if condition found as a result of check up, double code with, and prior to condition found. If verbatim comment of "check up," 302 Follow up care: check up in connection with another condition, post-operative check up, removal of stitches, prescription refill. Inclusion being followed up. Infer condition with "prescription renewed," type of answer; if an inference is not possible, code only 302. If a response implies that the respondent has had more than one visit for a specific condition. Code 302. If the number of different conditions being followed up exceed the space allowed, code the conditions to the exclusion of 302. Other Conditions 303 Test and shots (preventive care): vaccination, flu shot, German measles shot, Pap test, hearing test, blood sugar test. Code only when it appears the respondent was seen only for test or shots but not a general check up or specific diagnosis (see also 302 and 302 Eye exam goes in 236. 304 Fertility problems and advice: birth control advice and services; vasectomy, tubes tied, infertile, trying to become pregnant, sterility, BCP refill, IUD insertion/removal. Abortion goes in 249. 305 Dental problems: gingivitis, toothache, abscess, teething, bleeding gums, "tooth extraction--wouldn't heal." 306 Consultation with M.D. for reasons other than illness or preventive care: ear pierced, "to get to know doctor," other advice or consultation excluding medical treatment. 307 Auto accident 308 Other accident (i.e., drowning) 309 Suicide NOTE that a normal or forceps delivery is not counted as an operation; "put in traction" is not an operation. 401 Tonsillectomy, adenoidectomy. 402 Appendectomy. 403 Hemorrhoidectomy, fistulectomy. 404 Operations for hernia (hernioplasty, herniorrhaphy). 405 Hysterectomy (ovaries, tubes removed). 406 Dilation and curettage (d and c) for other than miscarriage or termination of pregnancy; see also 444 and 447. 407 Other operation of female genital organs or for "female trouble" (removed fibroid tumor, packed uterus for bleeding, insertion of radium seeds, hymenotomy). 408 Gall bladder operation, cholecystectomy. 409 Setting a fracture or dislocation (includes removal of pins, spikes, etc. afterwards; other orthopedic procedures--osteotomy, club foot surgery, meniscectomy, hip replacement). Bunionectomy goes in 437. 410 Operation for spinal ("disc") conditions (spinal fusion, back operation--unspecified, lumbar disc removal, laminectomy). 411 Circulatory system operation, including varicose veins (endarterectomy). Cardiac catherization goes in 445. 412 Minor eye surgery (adjustment of eye muscles, removal of pterygium). 413 Cataract operation, "ultrasonic probe on eye for cataract." 414 Operations on the eye other than 412 and 413 (detached retina, removal of fluid behind eyeball). 415 Operations on the ear or mastoids. 416 Amputation and repair of old amputation. 417 Operation for ulcers of stomach, duodenum or jejunum. 418 Other operations on the digestive system (liver operation, intussusception, colostomy, coloscopy, esophageal surgery). 419 Operations on the teeth or gums, including extraction, jaw operation. 420 Operation on the nose, throat, pharynx, nasopharynx, or tonsils, nasal polypectomy. Tonsillectomy and adenoidectomy go in 401. 421 Simple or radical mastectomy (remove breast), other breast surgery. Biopsies go in 430. 422 Operation on kidney. 423 Operation on bladder (urethral stricture, cystoscopy for other than diagnosis, meatotomy). 424 Operation on prostate. 425 Other operation on male genital organs. Circumcision goes in 431. Vasectomy goes in 448. 426 Operation on skin, including skin grafts (remove mole, cyst, wart), skin cancer, lymph nodes, extensive stitching requiring anesthesia, cauterizing debridement of burns, ganglions removed). 427 Operation on the endocrine system (thyroidectomy, removal of parotid gland tumor). 428 Operation on brain and skull. 429 Operation on nervous system, spinal cord. 430 Biopsy; bone marrow aspiration. 431 Circumcision of other than newborn. 432 Removed cancerous tumor (operation for cancer--site not specified). 433 Dilated esophagus (bougienage, esophagoscopy). 434 Removed benign tumor--site not specified, lipoma. 435 Removed unspecified tumor--site not specified. 436 Any operation on the heart (open heart surgery). 437 Bunionectomy, ingrown toenail surgery, other minor foot operations, remove Morton's neuroma 438 Exploratory laparotomy, splenectomy. 439 Plastic surgery. 440 Boil or ear drained, infected gland lanced or drained, tube put in ear, knee joint aspirated, inserted tube--no other information, drained abscess. 441 Operations on the respiratory system (sticking a tube into trachea, bronchia, lungs, thoracentesis, broncoscopy), operation on larynx. 442 Operation on muscles, tendons, ligaments, and nerves in extremities (sew together broken muscles inside leg, carpal tunnel syndrome, fix trigger thumbs). 443 Remove bullet--site unspecified, operation in connection with stub wounds--site unspecified. 444 D and C for miscarriage. Abortion, legal or illegal, goes in 447. 445 Cardiac catherization. 446 Exchange transfusion. 447 D and C or suction curettage for abortion, legal or illegal (any medical procedure for purpose of abortion). 448 Tubes tied, vasectomy (double code with 304 when operation not performed for health reasons). 449 Caesarian section. 450 Other operation connected with pregnancy (cyst removed, biopsy prior to delivery, hysterectomy to terminate early pregnancy, laparotomy for ruptured ovarian pregnancy). Chronic Conditions (appended) 500 Deaf 775 Old age 780 Surgery, not specified 800 AIDS 777 Other Condition 990 No specific condition given NSFH2/February 96 APPENDIX B - RELIGION CODES These codes are used with variables: Main: MI17, MI67, MI97, MI128, MI146, MN11 Child: FO205 0 No religion 1 Roman Catholic 2 Jewish 3 Baptist 4 Episcopalian 5 Lutheran 6 Methodist 7 Latter Day Saints (Mormon) 8 Presbyterian 9 United Church of Christ (Congregational) 10 Protestant (UNSPECIFIED DENOMINATION) 12 Assembly of God (or Assemblies of God) 13 Christian and Missionary Alliance 14 Christian Church (Disciples of Christ) (Christian Church -- any modifier such as First, Eastside, Community, etc.) (Christian - Disciples) (Christian - no including "just a Christian" or "Christian - no denomination") 15 Christian Congregation 16 Christian Reformed Church of North America (Christian Reformed) 17 Church of Christ, Scientist (Christian Scientist) 18 Church of God - Anderson, IN 19 Church of God - Cleveland, TN 20 Church of God (no affiliation specified) 21 Church of God in Christ 22 Church of the Brethren (If Brethren appears by itself, code in this category) 23 Church of the Nazarene (If Nazarene appears by itself, code in this category) 24 Church of Christ 25 Community Churches (Interdenominational) (Non-sectarian) 26 Evangelical Covenant Church 27 Evangelical Free Church 28 Full Gospel Fellowship 29 International Church of the Foursquare Gospel (Foursquare Gospel) 30 Jehovah's Witness 31 Mennonite Church 32 Orthodox Churches (ANY Eastern, Greek, Russian, Serbian, or Ukrainian Orthodox Churches INCLUDING Orthodox Church in America, American Orthodox Church) 33 Pentecostal (All Churches with Pentecostal in title) 34 Reformed Church (Reformed) 35 Reorganized Church of Jesus Christ of Latter-Day Saints (Reorganized Mormon) 36 Salvation Army 37 Seventh Day Adventists 38 Unitarian 39 Wesleyan 40 Non-Chalcedonian Orthodox Churches (Armenian, Assyrian, Syrian, Coptic, and Ethiopian) 41 All other western Roman Catholic Churches besides the Roman Catholic Church 42 All other Reformed-Presbyterian Churches 43 All other liberal churches 44 All other members of Pietist Family 45 All other members of Holiness Family 46 All other members of Pentecostal Family 47 All other members of European Free Church Family (Mennonites, Amish, Brethren, Quakers) 48 All other members of the Christian Church Churches of Christ subfamily - Restoration 49 All members of Independence Fundamentalist Family 50 All other members of Adventist Family 51 All other members of Latter-day Saint Family 52 All members of Communal Group 53 All New Thought Family 54 All Psychic Group 55 All Ritual Magick Groups 56 All Islamic Sub-family 57 All Hindu Sub-family 58 All Buddhist Sub-family 59 All Shinto and Taoism Families 60 All Miscellaneous Religious Bodies 61 "Christian" (no other description given. Could be "just a christian" or member of Christian Church. 62 Personal Churches (My own, Practice at home, Studying different Churches; personal bible study, believe in supreme being.) 63 "Born again Christian" 64 "Charismatic" 95 NOT CODEABLE, (a church was not mentioned or was incomplete) NSFH2/February 95 APPENDIX C - REASONS STOPPED LIVING WITH PARENTS CODES These codes are used with variables: Main: MF123, MF130, MF134, MF138, MF145, MF148 01 To attend school 02 To get married 03 To live with (boy/girl) friend 04 Found job in another area 05 Moved into own house/apartment/room 06 Travel 07 To enter the Armed Forces 09 Arguments, fights, family problems, to get away from family 10 Institutionalized 11 R wanted to be on own, to be independent 12 R migrated to U.S./emigration to other country 13 R reunited with husband/wife 14 R was pregnant/gave birth 15 R moved in with other relative/friends 16 R's parent(s) died 17 R's parents separated/divorced 18 Other change in parents' lives or living arrangements (e.g., parents moved; parents retired; parent remarried) 19 Financial reasons (Family couldn't afford to have R there, R left to help family financially, etc.) 20 R or child's health improved 97 Refused 98 Don't know 99 Inap/No Answer NSFH2/February 95 APPENDIX D - REASONS STARTED LIVING WITH PARENTS CODES These codes are used with variables: Main: MF125, MF128, MF132, MF136, MF140, MF143 01 Finished or left school 02 Got divorced or separated 03 Returned from Armed Forces 04 Couldn't afford to live on my own 05 Problems with roommates 06 Broke up with lover/partner 07 Husband/Wife was away in Armed Forces 09 De-institutionalized 13 R or Spouse lost job/returned home to look for work/found job in home town 14 Work related separation from spouse 15 Farm mentions 16 School related mentions 17 Health reasons (respondent)--include pregnancy 18 Take care of family/family members 19 R is homesick 20 To save money/share expenses/transitional between houses 21 Roommate/spouse/partner/lover died 22 R had no other choice/no where else to live 23 Mother or father died 24 Getting ready to join military 25 Getting ready to marry 26 Returned from church mission (returned from LDS mission) 27 Parents wanted R to move back (includes: parent asked, wanted, begged R to come home) 28 Resolved family differences (made up with parents) 29 Returned from vacation/travel 30 Never lived apart 97 Refused 99 Inap/No Answer NSFH2/August 96 APPENDIX E - CENSUS DIVISIONS OF THE UNITED STATES (REGIONS) NORTHEAST Maine (ME) New Hampshire (NH) Vermont (VT) Massachusetts (MA) Rhode Island (RI) Connecticut (CT) New York (NY) New Jersey (NJ) Pennsylvania (PA) MIDWEST (used to be called NORTH CENTRAL) Michigan (MI) Wisconsin (WI) Ohio (OH) Indiana (IN) Illinois (IL) Minnesota (MN) Iowa (IA) Missouri (MO) North Dakota (ND) South Dakota (SD) Nebraska (NE) Kansas (KS) SOUTH Delaware (DE) Maryland (MD) Virginia (VA) West Virginia (WV) North Carolina (NC) South Carolina (SC) Georgia (GA) Florida (FL) Kentucky (KY) Tennessee (TN) Alabama (AL) Mississippi (MS) Arkansas (AR) Oklahoma (OK) Louisiana (LA) Texas (TX) WEST Montana (MT) Wyoming (WY) Colorado (CO) New Mexico (NM) Idaho (ID) Utah (UT) Arizona (AZ) Nevada (NV) Washington (WA) Oregon (OR) California (CA) Alaska (AK) Hawaii (HI) NSFH2/February 95 APPENDIX F - OCCUPATION CODES (Same codes as 1990 Census of Population and Housing) These codes are used with variables: Main: MP10A, MP13A, MP85A, MP89A Child: FY33, FO221A, FO239A, FO253, FO256A Proxy: ZZ25A, ZZ51A 003 Legislators (111) 004 Chief executives and general administrators, public administration (112) 005 Administrators and officials, public administration (1132-1139) 006 Administrators, protective services (1131) 007 Financial managers (122) 008 Personnel and labor relations managers (123) 009 Purchasing managers (124) 013 Managers, marketing, advertising, and public relations (125) 014 Administrators, education and related fields (128) 015 Managers, medicine and health (131) 016 Postmasters and mail superintendents (1344) 017 Managers, food serving and lodging establishments (1351) 018 Managers, properties and real estate (1353) 019 Funeral directors (pt 1359) 021 Managers, service organizations, n.e.c. (127, 1352, 1354, pt 1359) 022 Managers and administrators, n.e.c. (121, 126, 132-1343, 136-139) 023 Accountants and auditors (1412) 024 Underwriters (1414) 025 Other financial officers (1415, 1419) 026 Management analysts (142) 027 Personnel, training, and labor relations specialists (143) 028 Purchasing agents and buyers, farm products (1443) 029 Buyers, wholesale and retail trade except farm products (1442) 033 Purchasing agents and buyers, n.e.c. (1449) 034 Business and promotion agents (145) 035 Construction inspectors (1472) 036 Inspectors and compliance officers, except construction (1473) 037 Management related occupations, n.e.c. (149) 043 Architects (161) 044 Aerospace (1622) 045 Metallurgical and materials (1623) 046 Mining (1624) 047 Petroleum (1625) 048 Chemical (1626) 049 Nuclear (1627) 053 Civil (1628) 054 Agricultural (1632) 055 Electrical and electronic (1633, 1636) 056 Industrial (1634) 057 Mechanical 1635) 058 Marine and naval architects (1637) 059 Engineers, n.e.c. (1639) 063 Surveyors and mapping scientists (164) 064 Computer systems analysts and scientists (171) 065 Operations and systems researchers and analysts (172) 066 Actuaries (1732) 067 Statisticians (1733) 068 Mathematical scientists, n.e.c. (1739) 069 Physicists and astronomers (1842, 1843) 073 Chemists, except biochemists (1845) 074 Atmospheric and space scientists (1846) 075 Geologists and geodesists (1847) 076 Physical scientists, n.e.c. (1849) 077 Agricultural and food scientists (1853) 078 Biological and life scientists (1854) 079 Forestry and conservation scientists (1852) 083 Medical scientists (1855) 084 Physicians (261) 085 Dentists (262) 086 Veterinarians (27) 087 Optometrists (281) 088 Podiatrists (283) 089 Health diagnosing practitioners, n.e.c. (289) 095 Registered nurses (29) 096 Pharmacists (301) 097 Dietitians (302) 098 Respiratory therapists (3031) 099 Occupational therapists (3032) 103 Physical therapists (3033) 104 Speech therapists (3034) 105 Therapists n.e.c. (3039) 106 Physicians' assistants (304) 113 Earth, environmental, and marine science teachers (2212) 114 Biological science teachers (2213) 115 Chemistry teachers (2214) 116 Physics teachers (2215) 117 Natural science teachers, n e.c. (2216) 118 Psychology teachers (2217) 119 Economics teachers (2218) 123 History teachers (2222) 124 Political science teachers (2223) 125 Sociology teachers (2224) 126 Social science teachers, n.e.c. (2225) 127 Engineering teachers (2226) 128 Mathematical science teachers (2227) 129 Computer science teachers (2228) 133 Medical science teachers (2231) 134 Health specialties teachers (2232) 135 Business, commerce, and marketing teachers (2233) 136 Agriculture and forestry teachers (2234) 137 Art, drama, and music teachers (2235) 138 Physical education teachers (2236) 139 Education teachers (2237) 143 English teachers (2238) 144 Foreign language teachers (2242) 145 Law teachers (2243) 146 Social work teachers (2244) 147 Theology teachers (2245) 148 Trade and industrial teachers (2246) 149 Home economics teachers (2247) 153 Teachers, postsecondary, n.e.c. (2249) 154 Postsecondary teachers, subject not specified 155 Teachers, prekindergarten and Kindergarten (231) 156 Teachers, elementary school (232) 157 Teachers, secondary school (233) 158 Teachers, special education (235) 159 Teachers, n.e.c. (236, 239) 163 Counselors, educational and vocational (24) 164 Librarians (251) 165 Archivists and curators (252) 166 Economists (1912) 167 Psychologists (1915) 168 Sociologists (1916) 169 Social scientisls, n.e.c. (1913, 1914, 1919) 173 Urban planners (192) 174 Social workers (2032) 175 Recreation workers (2033) 176 Clergy (2042) 177 Religious workers, n.e.c. (2049) 178 Lawyers (211) 179 Judges (212) 183 Authors (321) 184 Technical writers (398) 185 Designers (322) 186 Musicians and composers (323) 187 Actors and directors (324) 188 Painters, sculptors, craft-artists, and artist printmakers (325) 189 Photographers (326) 193 Dancers (327) 194 Artists, performers, and related workers, n.e.c. (328, 329) 195 Editors and reporters (331) 197 Public relations specialists (332) 198 Announcers (333) 199 Athletes (34) 203 Clinical laboratory technologists and technicians (362) 204 Dental hygienists (363) 205 Health record technologists and technicians (364) 206 Radiologic technicians (365) 207 Licensed practical nurses (366) 208 Health technologists and technicians, n.e.c. (369) 213 Electrical and electronic technicians (3711) 214 Industrial engineering technicians (3712) 215 Mechanical engineering technicians (3713) 216 Engineering technicians, n.e.c. (3719) 217 Drafting occupations (372) 218 Surveying and mapping technicians (373) 223 Biological technicians (382) 224 Chemical technicians (3831) 225 Science technicians, n.e.c. (3832, 3833, 384, 389) 226 Airplane pilots and navigators (825) 227 Air traffic controllers (392) 228 Broadcast equipment operators (393) 229 Computer programmers (3971, 3972) 233 Tool programmers, numerical control (3974) 234 Legal assistants (396) 235 Technicians, n.e.c. (399) 243 Supervisors and proprietors, sales occupations (40) 253 Insurance sales occupations (4122) 254 Real estate sales occupations (4123) 255 Securities and financial services sales occupations (4124) 256 Advertising and related sales occupations (4153) 257 Sales occupations, other business services (4152) 258 Sales engineers (421) 259 Sales representatives, mining, manufacturing, and wholesale (423, 424) 263 Sales workers, motor vehicles and boats (4342, 4344) 264 Sales workers, apparel (4346) 265 Sales workers, shoes (4351) 266 Sales workers, furniture and home furnishings (4348) 267 Sales workers; radio, TV, hi-fi, and appliances (4343, 4352) 268 Sales workers, hardware and building supplies (4353) 269 Sales workers, parts (4367) 274 Sales workers, other commodities (4345, 4347, 4354, 4356, 4359,4362, 4369) 275 Sales counter clerks (4363) 276 Cashiers (4364) 277 Street and door-to-door sales workers (4366) 278 News vendors (4365) 283 Demonstrators, promoters and models, sales (445) 284 Auctioneers (447) 285 Sales support occupations, n.e.c. (444, 446, 449) 303 Supervisors, general office (4511, 4513, 4514, 4516, 4519, 4529) 304 Supervisors, computer equipment operators (4512) 305 Supervisors, financial records processing (4521) 306 Chief communications operators (4523) 307 Supervisors; distribution, scheduling, and adjusting clerks (4522, 4524-4528) 308 Computer operators (4612) 309 Peripheral equipment operators (4613) 313 Secretaries (4622) 314 Stenographers (4623) 315 Typists (4624) 316 Interviewers (4642) 317 Hotel clerks (4643) 318 Transportation ticket and reservation agents (4644) 319 Receptionists (4645) 323 Information clerks, n.e.c. (4649) 325 Classified-ad clerks (4662) 326 Correspondence clerks (4663) 327 Order clerks (4664) 328 Personnel clerks, except payroll and timekeeping (4692) 329 Library clerks (4694) 335 File clerks (4696) 336 Records clerks (4699) 337 Bookkeepers, accounting, and auditing clerks (4712) 338 Payroll and timekeeping clerks (4713) 339 Billing clerks (4715) 343 Cost and rate clerks (4716) 344 Billing, posting, and calculating machine operators (4718) 345 Duplicating machine operators (4722) 346 Mail preparing and paper handling machine operators (4723) 347 Office machine operators, n.e.c. (4729) 348 Telephone operators (4732) 353 Communications equipment operators, n.e.c. (4733, 4739) 354 Postal clerks, exc. mail carriers (4742) 355 Mail carriers, postal service (4743) 356 Mail clerks, exc. postal service (4744) 357 Messengers (4745) 359 Dispatchers (4751) 363 Production coordinators (4752) 364 Traffic, shipping, and receiving clerks (4753) 365 Stock and inventory clerks (4754) 366 Meter readers (4755) 368 Weighers, measurers, checkers and samplers (4756, 4757) 373 Expediters (4758) 374 Material recording, scheduling, and distributing clerks n.e.c (4759) 375 Insurance adjusters, examiners, and investigators (4782) 376 Investigators and adjustors except insurance (4783) 377 Eligibility clerks, Social welfare (4784) 378 Bill and account collectors (4786) 379 General office clerks (463) 383 Bank tellers (4791) 384 Proofreaders (4792) 385 Data-entry keyers (4793) 386 Statistical clerks (4794) 387 Teachers' aides (4795) 389 Administrative support occupations n.e.c. (4787, 4799) 403 Launderers and ironers (503) 404 Cooks, private household (504) 405 Housekeepers and butlers (505) 406 Child care workers, private household (50S) 407 Private household cleaners and servants (502, 507, 509) 413 Supervisors, firefighting and fire prevention occupations (5111) 414 Supervisors, police and detectives (5112) 415 Supervisors, guards (5113) 416 Fire inspection and fire prevention occupations (5122) 417 Firefighting occupations (5123) 418 Police and detectives, public service (5132) 423 Sheriffs, bailiffs, and other law enforcement officers (5134) 424 Correctional institution officers (5133) 425 Crossing guards (5142) 426 Guards and police, exc public service (5144) 427 Protective service occupations, n.e.c (5149) 433 Supervisors, food preparation and service occupations (5211) 434 Bartenders (5212) 435 Waiters and waitresses (5213) 436 Cooks (5214, 5215) 438 Food counter, fountain and related occupations (5216) 439 Kitchen workers, food preparation (5217) 443 Waiters'/waitresses' assistants (5218) 444 Miscellaneous food preparation occupations (5219) 445 Dental assistants (5232) 446 Health aides, except nursing (5233) 447 Nursing aides, orderlies, and attendants (5236) 448 Supervisors, cleaning and building service workers (5241) 449 Maids and housemen (5242, 5249) 453 Janitors and cleaners (5244) 454 Elevator operators (5245) 455 Pest control occupations (5246) 456 Supervisors, personal service occupations (5251) 457 Barbers (5252) 458 Hairdressers and cosmetologists (5253) 459 Attendants, amusement and recreation facilities (5254) 461 Guides (5255) 462 Ushers (5256) 463 Public transportation attendants (5257) 464 Baggage porters and bellhops (5262) 465 Welfare service aides (5263) 466 Family child care providers (pt 5264) 467 Early childhood teacher's assistants (pt 5264) 468 Child care workers, n.e.c. (pt 5264) 469 Personal service occupations, n.e.c. (5258, 5269) 473 Farmers, except horticultural (5512-5514) 474 Horticultural specialty farmers (5515) 475 Managers, farms, except horticultural (5522-5524) 476 Managers, horticultural specialty farms (5525) 477 Supervisors, farm workers (5611) 479 Farm workers (5612-5617) 483 Marine life cultivation workers (5618) 484 Nursery workers (5619) 485 Supervisors, related agricultural occupations (5621) 486 Groundskeepers and gardeners,except farm (5622) 487 Animal caretakers, except farm (5624) 488 Graders and sorters, agricultural products (5625) 489 Inspectors, agricultural products (5627) 494 Supervisors, forestry, and logging workers (571) 495 Forestry workers, except logging (572) 496 Timber cutting and loggng occupations (573, 579) 497 Captains and other officers, fishing vessels (pt 8241) 498 Fishers (583) 499 Hunters and trappers (584) 503 Supervisors, mechanics and repairers (60) 505 Automobile mechanics (pt 6111) 506 Automobile mechanic apprentices (pt 6111) 507 Bus, truck, and stationary engine mechanics (6112) 508 Aircraft engine mechanics (6113) 509 Small engine repairers (6114) 514 Automobile body and related repairers (6115) 515 Aircraft mechanics, exc. engine (6116) 516 Heavy equipment mechanics (6117 517 Farm equipment mechanics (6118) 518 Industrial machinery repairers (613) 519 Machinery maintenance occupations (614) 523 Electronic repairers, communications and industrial equipment (6151, 6153, 6155) 525 Data processing equipment repairers (6154) 526 Household appliance and power tool repairers (6156) 527 Telephone line installers and repairers (6157) 529 Telephone installers and repairers (6158) 533 Miscellaneous electrical and electronic equipment repairers (6152, 6159) 534 Heating, air conditioning and refrigeration mechanics (616) 535 Camera, watch, and musical instrument repairers (6171,6172) 536 Locksmiths and safe repairers (6173) 538 Office machine repairers (6174) 539 Mechanical controls and valve repairers (6175) 543 Elevator installers and repairers (6176) 544 Millwrights (6178) 547 Specified mechanics and repairers, n.e.c. (6177, 6179) 549 Not specified mechanics and repairers 553 Supervisors; brickmasons, stonemasons, and tile setters (6312) 554 Supervisors, carpenters and related workers (6313) 555 Supervisors,electricians and power transmission installers (6314) 556 Supervisors; painters, paperhangers, and plasterers (6315) 557 Supervisors; plumbers, pipefitters, and steamfitters (6316) 558 Supervisors, construction n.e.c. (6311, 6318) 563 Brickmasons and stonemasons (pt 6412, pt 6413) 564 Brickmason and stonemason apprentices (pt 6412, pt 6413) 565 Tile setters, hard and soft (pt 6414, pt 6462) 566 Carpet installers (pt 6462) 567 Carpenters (pt 6422) 569 Carpenter apprentices (pt 6422) 573 Drywall installers (6424) 575 Electricians (pt 6432) 576 Electrician apprentices (pt 6432) 577 Electrical power installers and repairers (6433) 579 Painters, construction and maintenance (6442) 583 Paperhangers (6443) 584 Plasterers (6444) 585 Plumbers, pipefitters, and steamfitters (pt 645) 587 Plumber, pipefitter, and steamfitter apprentices (pt 645) 588 Concrete and terrazzo finishers (6463) 589 Glaziers (6464) 593 Insulation workers (6465) 594 Paving, surfacing, and tamping equipment operators (6466) 595 Roofers (6468) 596 Sheetmetal duct installers (6472) 597 Structural metal workers (6473) 598 Drillers, earth (6474) 599 Construction trades, n.e.c. (6467, 6475, 6476, 6479) 613 Supervisors, extrative occupations (632) 614 Drillers, oil well (652) 615 Explosives workers (653) 616 Mining machine operators (654) 617 Mining occupations, n.e.c. (656) 628 Supervisors, production occupations (67, 71) 634 Tool and die makers (pt 6811) 635 Tool and die maker apprentices (pt 6811) 636 Precision assemblers, metal (6812) 637 Machinists (pt 6813) 639 Machinist apprentices (pt 6813) 643 Boilermakers (6814) 644 Precision grinders, filers, and tool sharpeners (6816) 645 Patternmakers and model makers, metal (6817) 646 Lay out workers (6821) 647 Precious stones and metals workers (Jewelers) (6822, 6866) 649 Engravers, metal (6823) 653 Sheet metal workers (pt 6824) 654 Sheet metal worker apprentices (pt 6824) 655 Miscellaneous precision metal workers (6829) 656 Patternmakers and model makers, wood (6831) 657 Cabinet makers and bench carpenters (6832) 658 Furniture and wood finishers (6835) 659 Miscellaneous precision woodworkers (6839) 666 Dressmakers (pt 6852, pt 7752) 667 Tailors (pt 6852) 668 Upholsterers (6853) 669 Shoe repairers (6854) 674 Miscellaneous precision apparel and fabric workers (6856, 6859, pt 7752) 675 Hand molders and shapers, except jewelers (6861) 676 Patternmakers, lay-out workers, and cutters (6862) 677 Optical goods workers (6864, pt 7477, pt 7677) 678 Dental laboratory and medical appliance technicians (6865) 679 Bookbinders (6844) 683 Electrical and electronic equipment assemblers (6867) 684 Miscellaneous precision workers, n.e.c. (6869) 686 Butchers and meat cutters (6871) 687 Bakers (6872) 688 Food batchmakers (6873, 6879) 689 Inspectors, testers, and graders (6881, 828) 693 Adjusters and calibrators (6882) 694 Water and sewage treatment plant operators (691) 695 Power plant operators (pt 693) 696 Stationary engineers (pt 693, 7668) 699 Miscellaneous plant and system operators (692, 694, 695, 666) 703 Lathe and turning machine set-up operators (7312) 704 Lathe and turning machine operators (7512) 705 Milling and planing machine operators (7313, 7513) 706 Punching and stamping press machine operators (7314, 7317,7514, 7517) 707 Rolling machine operators (7316, 7516) 708 Drilling and boring machine operators (7318, 7518) 709 Grinding, abrading, buffing, and polishing machine operators (7322, 7324, 7522) 713 Forging machine operators (7319, 7519) 714 Numerical control machine operators (7326) 715 Micellaneous metal, plastic, stone, and glass working machine operators (7329, 7529) 717 Fabricating machine operators n.e.c. (7339, 7539) 719 Molding and casting machine operators (7315, 7342, 7515,7542) 723 Metal plating machine operators (7343, 7543) 724 Heat treating equipment operators (7344, 7544) 725 Miscellaneous metal and plastic processing machine operators (7349, 7549) 726 Wood lathe, routing and planning machine operators (7431,7432, 7631, 7632) 727 Sawing machine operators (7433, 7633) 728 Shaping and joining machine operators (7435, 7635) 729 Nailing and tacking machine operators (7636) 733 Miscellaneous woodworking machine operators (7434, 7439, 7634, 7639) 734 Printing press operators (7443, 7643) 735 Photoengravers and lithographers (6842, 7444, 7644) 736 Typesetters and compositors (6841, 7642) 737 Miscellaneous printing machine operators (6849, 7449, 7649) 738 Winding and twisting machine operators (7451, 7651) 739 Knitting, looping, taping, and weaving machine operators (7452, 7652) 743 Textile cutting machine operators (7654) 744 Textile sewing machine operators (7655) 745 Shoe machine operators (7656) 747 Pressing machine operators (7657) 748 Laundering and dry cleaning machine operators (6855, 7658) 749 Miscellaneous textile machine operators (7459, 7659) 753 Cementing and gluing machine operators (7661) 754 Packaging and filling machine operators (7462, 7662) 755 Extruding and forming machine operators (7463, 7663) 756 Mixing and blending machine operators (7664) 757 Separating, filtering, and clarifying machine operators (7476, 7666, 7676) 758 Compressing and compacting machine operators (7467, 7667) 759 Painting and paint spraying machine operators (7669) 763 Roasting and baking machine operators, food (7472, 7672) 764 Washing, cleaning, and pickling machine operators (7673) 765 Folding machine operators (7474, 7674) 766 Furnace, kiln, and oven operators, exc. food (7675) 768 Crushing and grinding machine operators (pt 7477, pt 7677) 769 Slicing and cutting machine operators (7478, 7678) 773 Motion picture projectionists (pt 7479) 774 Photographic process machine operators (6863, 6868, 7671) 777 Miscellaneous machine operators, n.e.c. (pt 7479, 7665, 7679) 779 Machine operators, not specified 783 Welders and cutters (7332, 7532, 7714) 784 Solderers and brazers (7333, 7533, 7717) 785 Assemblers (772, 774) 786 Hand cutting and trimming occupations (7753) 787 Hand molding, casting, and forming occupations (7754, 7755) 789 Hand painting, coating, and decorating occupations (7756) 793 Hand engraving and printing occupations (7757) 795 Miscellaneous hand working occupations (7758, 7759) 796 Production inspectors, checkers, and entertainers (782, 787) 797 Production testers (783) 798 Production samplers and weighers (784) 799 Graders and sorters, exc. agricultural (785) 803 Supervisors, motor vehicle operators (8111) 804 Truck drivers (8212-8214) 806 Driver-sales workers (8218) 808 Bus drivers (8215) 809 Taxicab drivers and chauffeurs (8216) 813 Parking lot attendants (874) 814 Motor transportation occupations, n.e.c. (8219) 823 Railroad conductors and yardmasters (8113) 824 Locomotive operating occupations (8232) 825 Railroad brake, signal, and switch operators (8233) 826 Rail vehicle operators, n c.c. (8239) 828 Ship captains and mates, except fishing boats (pt 8241, 8242) 829 Sailors and deckhands (8243) 833 Marine engineers (8244) 834 Bridge, lock, and lighthouse tenders (8245) 843 Supervisors, material moving equipment operators (812) 844 Operating engineers (8312) 845 Longshore equipment operators (8313) 848 Hoist and winch operators (8314) 849 Crane and tower operators (8315) 853 Excavating and loading machine operators (8316) 855 Grader, dozer, and scraper operators (8317) 856 Industrial truck and tractor equipment operators (8318) 859 Miscellaneous material moving equipment operators (8319) 864 Supervisors, handlers, equipment cleaners, and laborers, n.e.c. (85) 865 Helpers, mechanics and repairers (863) 866 Helpers, construction trades (8641-8645, 8648) 867 Helpers, surveyor (8646) 868 Helpers, extractive occupations (865) 869 Construction laborers (871) 874 Production helpers (861, 862) 875 Garbage collectors (8722) 876 Stevedores (8723) 877 Stock handlers and baggers (8724) 878 Machine feeders and offbearers (8725) 883 Freight, stock, and material handlers, n.e.c. (8726) 885 Garage and service station related occupations (873) 887 Vehicle washers and equipment cleaners (875) 888 Hand packers and packagers (8761) 889 Laborers, except construction (8769) 903 Commissioned Officers and Warrant Officers 904 Non-commissioned Officers and Other Enlisted Personnel 905 Military occupation, rank not specified 909 Last worked 1984 or earlier NSFH2/February 95 APPENDIX I - INDUSTRY CODES (Same codes as 1990 Census of Population and Housing) These codes are used with variables: Main: MP11, MP86 Child: FO240, FO257 010 Agricultural production, crops (01) 011 Agricultural production, livestock (02) 012 Veterinary services (074) 020 Landscape and horticultural services (078) 030 Agricultural services, n.e.c (071, 072, 075, 076) 031 Forestry (08) 032 Fishing, hunting, and trapping (09) 040 Metal mining (10) 041 Coal mining (12) 042 Oil and gas extraction (13) 050 Nonmetallic mining and quarrying, except fuels (14) 060 CONSTRUCTION (15, 16, 17) 100 Meat products (201) 101 Dairy products (202) 102 Canned, frozen, and preserved fruits and vegetables (203) 110 Grain mill products (204) 111 Bakery products (205) 112 Sugar and confectionery products (206) 120 Beverage industries (208) 121 Miscellaneous food preparations and kindred products (207,209) 122 Not specified food industries 130 Tobacco manufactures (21) 132 Knitting mills (225) 140 Dyeing and finishing textiles, except wool and knit goods (226) 141 Carpets and rugs (227) 142 Yarn, thread, and fabric mills (221-224, 228) 150 Miscellaneous textile mill products (229) 151 Apparel and accessories, except knit (231-238) 152 Miscellaneous fabricated textile products (239) 160 Pulp, paper, and paperboard mills (261-263) 161 Miscellaneous paper and pulp products (267) 162 Paperboard containers and boxes (265) 171 Newspaper publishing and printing (271) 172 Printing, publishing, and allied industries, except newspapers (272-279) 180 Plastics, synthetics, and resins (282) 181 Drugs (283) 182 Soaps and cosmetics (284) 190 Paints, varnishes, and related products (285) 191 Agricultural chemicals (287) 192 Industrial and miscellaneous chemicals (281, 286, 289) 200 Petroleum refining (29l) 201 Miscellaneous petroleum and coal products (295, 299) 210 Tires and inner tubes (301) 211 Other rubber products, and plastics footwear and belting (302-306) 212 Miscellaneous plastics products (308) 220 Leather tanning and finishing (311) 221 Footwear, except rubber and plastic (313, 314) 222 Leather products, except footwear (315-317, 319) 230 Logging (241) 231 Sawmills, planing mills, and millwork (242, 243) 232 Wood buildings and mobile homes (245) 241 Miscellaneous wood products (244, 249) 242 Furniture and fixtures (25) 250 Glass and glass products (321-323) 251 Cement, concrete, gypsum, and plaster products (324,327) 252 Structural clay products (325) 261 Pottery and related products (326) 262 Miscellaneous nonmetallic mineral and stone products (328, 329) 270 Blast furnaces, steelworks, rolling and finishing mills (331) 271 Iron and steel foundries (332) 272 Primary aluminum industries (3334, part 334, 3353-3355,3363,3365) 280 Other primary metal industries (3331, 3339, part 334, 3351,3356, 3357, 3364, 3366, 3369, 339) 281 Cutlery, handtools, and general hardware (342) 282 Fabricated structural metal products (344) 290 Screw machine products (345) 291 Metal forgings and stampings (346) 292 Ordnance (348) 300 Miscellaneous fabricated metal products (341, 343, 347, 349) 301 Not specified metal industries 310 Engines and turbines (351) 311 Farm machinery and equipment (352) 312 Construction and material handling machines (353) 320 Metalworking machinery (354) 321 Office and accounting machines (3578, 3579) 322 Computers and related equipment (3571-3577) 331 Machinery, except electrical n.e.c. (355, 356, 358, 359) 332 Not specified machinery 340 Household appliances (363) 341 Radio, TV, and communication equipment (365, 366) 342 Electrical machinery, equipment, and supplies, n.e.c. (361,363,364, 367, 369) 350 Not specified electrical machinery, equipment, and supplies 351 Motor vehicles and motor vehicle equipment (371) 352 Aircraft and parts (372) 360 Ship and boat building and repairing (373) 361 Railroad locomotives and equipment (374) 362 Guided missiles, space vehicles, and parts (376) 370 Cycles and miscellaneous transportation equipment (375, 379) 371 Scientific and controlling instruments (381, 382 exc. 3827) 372 Medical, dental, and optical instruments and supplies (3827,384, 385) 380 Photographic equipment and supplies (386) 381 Watches, clocks, and clockwork operated devices (387) 390 Toys, amusement, and sporting goods (394) 391 Miscellaneous manufacturing industries (39 exc. 394) 392 Not specified manufacturing industries 400 Railroads (40) 401 Bus service and urban transit (41, except 412) 402 Taxicab service (412) 410 Trucking service (421, 423) 411 Warehousing and storage (422) 412 U.S. Postal Service (43) 420 Water transportation (44) 421 Air transportation (45) 422 Pipe lines, except natural gas (46) 432 Services incidental to transportation (47) 440 Radio and television broadcasting and cablc (483, 484) 441 Telephone communications (481) 442 Telegraph and miscellancous communications services (482, 489) 450 Electric light and power (491) 451 Gas and steam supply systems (492, 496) 452 Electric and gas, and other combinations (493) 470 Water supply and irrigation (494, 497) 471 Sanitary services (495) 472 Not specified utilities 500 Motor vehicles and equipment (501) 501 Furniture and home furnishings (502) 502 Lumber and construction materials (503) 510 Professional and commercial equipment and supplies (504) 511 Metals and minerals, except petroleum (505) 512 Electrical goods (506) 521 Hardware, plumbing and heating supplies (507) 530 Machinery, equipment, and supplies (508) 531 Scrap and waste materials (5093) 532 Miscellaneous wholesale, durable goods (509 exc. 5093) 540 Paper and paper products (511) 541 Drugs, chemicals and allied products (512, 516) 542 Apparel, fabrics, and notions (513) 550 Groceries and related products (514) 551 Farm-product raw materials (515) 552 Petroleum products (517) 560 Alcoholic beverages (518) 561 Farm supplies (5191) 562 Miscellaneous wholesale, nondurable goods (5192-5199) 571 Not specified wholesale trade 580 Lumber and building material retailing (521, 523) 581 Hardware stores (525) 582 Retail nurseries and garden stores (526) 590 Mobile home dealers (527) 591 Department stores (531) 592 Variety stores (533) 600 Miscellaneous general merchandise stores (539) 601 Grocery stores (541) 602 Dairy products stores (545) 610 Retail bakeries (546) 611 Food stores,n.e.c. (542, 543, 544, 549) 612 Motor vehicle dealers (551, 552) 620 Auto and home supply stores (553) 621 Gasoline service stations (554) 622 Miscellaneous vehicle dealers (555, 556, 557, 559) 623 Apparel and accessory stores, except shoe (56, except 566) 630 Shoe stores (566) 631 Furniture and home furnishings stores (571) 632 Household appliance stores (572) 633 Radio, TV, and computer stores (5731, 5734) 640 Music stores (5735, 5736) 641 Eating and drinking places (58) 642 Drug stores (591) 650 Liquor stores (592) 651 Sporting goods, bicycles, and hobby stores (5941, 5945, 5946) 652 Book and stationery stores (5942, 5943) 660 Jewelry stores (5944) 661 Gift, novelty, and souvenir shops (5947) 662 Sewing, needlework and piece goods stores (5949) 663 Catalog and mail order houses (5961) 670 Vending machine operators (5962) 671 Direct selling establishments (5963) 672 Fuel dealers (598) 681 Retail florists (5992) 682 Miscellaneous retail stores (593, 5948, 5993-5995, 5999) 691 Not specified retail trade 700 Banking (60 exc. 603 and 606) 701 Savings institutions, including credit unions (603, 606) 702 Credit agencies, n.e.c. (61) 710 Security, commodity brokerage, and investment companies (62, 67) 711 Insurance (63, 64) 712 Real estate, including real estate-insurance offices (65) 721 Advertising (731) 722 Services to dwellings and other buildings (734) 731 Personnel supply services (736) 732 Computer and data processing services (737) 740 Detective and protective services (7381, 7382) 741 Business services, n.e.c. (732, 733, 735, 7383-7389) 742 Automotive rental and leasing, without drivers (751) 750 Automobile parking and car washes (752, 7542) 751 Automotive repair and related services (753, 7549) 752 Electrical repair shops (762, 7694) 760 Miscellaneous repair services (763, 764, 7692, 7699) 761 Private households (88) 762 Hotels and motels (701) 770 Lodging places, except hotels and motels (702, 703, 704) 771 Laundry, cleaning, and garment services (721 exc part 7219) 772 Beauty shops (723) 780 Barber shops (724) 781 Funeral service and crematories (726) 782 Shoe repair shops (725) 790 Dressmaking shops (part 7219) 791 Miscellaneous personal services (722, 729) 800 Theaters and motion pictures (781-783, 792) 801 Video tape rental (784) 802 Bowling centers (793) 810 Miscellaneous entertainment and recreation services (791, 794, 799) 812 Offices and clinics of physicians (801, 803) 820 Offices and clinics of dentists (802) 82l Offices and clinics of chiropractors (8041) 822 Offices and clinics of optometrists (8042) 830 Offices and clinics of health practitioners, n.e.c. (8043, 8049) 831 Hospitals (806) 832 Nursing and personal care facilities (805) 840 Health services,n.e.c. (807, 808, 809) 841 Legal services (81) 842 Elementary and secondary schooL (821) 850 Colleges and universities (822) 851 Vocational schooLs (824) 852 Libraries (823) 860 Educational services, n.e.c. (829) 861 Job training and vocational rehabilitation services (833) 862 Child day care services (part 835) 863 Family child care homes (part 835) 870 Residential care facilities, without nursing (836) 871 Social services, n.e.c. (832, 839) 872 Museums, art galleries, and zoos (84) 873 Labor unions (863) 880 Religious organizations (866) 881 Membership organizations, n.e.c. (861, 862, 864, 865, 869) 882 Engineering, architecturaL and surveying services (871) 890 Accounting, auditing, and bookkeeping services (872) 891 Research, development, and testing services (873) 892 Management and public relations services (874) 893 Miscellaneous professional and related services (899) 900 Executive and legislative offices (911-913) 901 General government, n.e.c. (919) 910 Justice, public order, and safety (92) 921 Public finance, taxation, and monetary policy (93) 922 Administration of human resources programs(94) 930 Administration of environmental quality and housing programs (95) 931 Administration of economic programs (96) 932 National security and international affairs (97) 940 Army 941 Air Force 942 Navy 950 Marines 951 Coast Guard 952 Armed Forces, Branch not specified 960 Military Reserves or National Guard 992 Last worked 1984 or earlier NSFH2/MAY 96 APPENDIX J - INCOME VARIABLES: CATEGORICAL AND CONSTRUCTED CATEGORICAL: INCOME INTERVALS WHEN RESPONDENT DOESN'T KNOW OR REFUSES DOLLAR AMOUNT In the income series, respondents who initially answer "don't know" or "refused" are asked a series of follow-up questions designed to bracket the income. The brackets are different for different income sources. Wage and Salary income and self-employment income intervals are: a I4a Less than $20,000 b I5a Less than $10,000 c I6a Less than $5,000 d I7a Less than $1,000 e I8a More than $30,000 f I9a More than $40,000 g I10a More than $50,000 (Note that each path through the series gives a unique amount, so that if someone doesn't like the allocation, they can change it.) Wage and Salary and Self-employment income variables: MQ2C (I4a), MQ4C (I12a), MQ6P01C (I4b), MQ7P01C (I12b), MQ6P02C (I4c) and MQ7P02C (I12c) Set incomenew = whatever it is (99999) a. If I4a = 1, Income = 10000 goto b If I4a = 2, Income = 40000 goto e If I4a = d, Income = 99998 If I4a = r, Income = 99997 goto end b. If I5a = 1, Income = 5000 goto c If I5a = 2, Income = 15000 goto end If I5a = d, r, goto end c. If I6a = 1, Income = 2500 goto d If I6a = 2, Income = 7500 goto end If I6a = d, r, goto end d. If I7a = 1, Income = 500 goto end If I7a = 2, Income = 3000 goto end If I7a = d, r, goto end e. If I8a = 1, Income = 50000 goto f If I8a = 2, Income = 30000 goto end If I8a = d, r goto end f. If I9a = 1 Income = 60000 goto g If I9a = 2 Income = 35000 goto end If I9a = d, r goto end g. If I10a = 1, Income = 70000 goto end If I10a = 2, Income = 45000 goto end If I10a = d, r goto end end For the following sources (Pensions, Social Security, Public Assistance, Other Govt Programs, Interest or Dividends, and Other Sources, the intervals are: a. (Less than $15,000) If a = 1; income = 8000 goto b If a = 2; income = 30000 goto d If a = d, r goto end b. (Less than $10,000) If b = 1, income = 5000 goto c If b = 2 income = 12500 goto end if b = d, r goto end c. (Less than $5,000) if c = 1, income = 2500 goto end if c = 2, income = 7500 goto end if c = d, r goto end d. (More than $20,000) if d = 1, income = 40000 goto end if d = 2, income = 17500 goto end if d= d, r goto end end CONSTRUCTED INCOME VARIABLES There are two types of constructed income files: 1) The "main" income data has other constructed variables appended to the end of it and was created from it's respective type or original file (R, C, X) exclusively. 2) The "best measures" file was created from a comparison of the main R and current spouse constructed income data. The following income data were collected in NSFH2: If R is a householder or spouse/partner of housholder: reports of all household members by source If R is neither a householder nor cohabiting: reports of R's income by source for himself/herself only In both cases, the following sources of income are distinguished: MQ2/6 wage/salary income MQ4/7 self employment income MQ12PA Social Security and SSI MQ12BP Other pensions MQ12CP Public Assistance MQ12DP Other government benefits MQ12EP Child Support/Alimony MQ12FP Other Interval measures were used in NSFH2 to get an approximate income report when the respondent initially reports "refused" or "don't know". For instance, if asked what the respondent's spouse earns from self employment, and the respondent answer's "don't know", then the interviewer would ask, "Was it less than $20,000", and if so, "Was it less than $15,000", and so on. -- Flags -- A flag was created for each variable indicating whether the income value is a direct response (flag=0), from an interval approximation (flag=1), or the final answer was don't know or refused (flag = 2). If there is not income reported for a given household member, that income and the respective flag will be = 0. Construction of the "Main" income data The task with the "main" constructed income files was to designate income receivers with an ordinal number (as opposed to simply the hh person number) so that their incomes could be readily accessed and used for creating aggregate measures. To do this, a matrix was created with income variables in fixed fields. The household members would be referenced by their ordinal position in the household, which is the order in which they were listed by the respondent. Household Relation Self-Emp. Member# To R Wages Income _________________________________________________ R wmemr semer ... 1 hhpn1 rel1 wmem1 seme1 ... 2 hhpn2 rel2 wmem2 seme2 ... 3 hhpn3 rel3 wmem3 seme3 ... . . . . . . . . . . . . . . . 16 hhpn16 rel16 wmem16 seme16 ... S pns rels wmems semes ... General format for creating household members' income variables: This example is for wage and salary income. There is some variation in the code that constructs wage/salary and self employed incomes versus all other sources. This is because the interview did not distinguish between respondent and everyone else for sources other than wages/salary and self employed. However, this variation does not change the basic format of the code that constructs income by ordinal position in the household. Example: Wage and Salary income loop through all household member numbers (1-16) if there is someone receiving income from the source (MQ5NUM>0) and the current person number in the loop 1-16 = the person number of the household member receiving the wage/sal income (MQ5P01-MQP02) and if the income is not refused and not DK (MQ6P01 ne 999997 or 999998) member n wage/sal income (constructed) = wage/sal income (MQ6PO1) flag = 0 else if the income is refused or DK and interval is dollar value member n wage/sal income (constructed) = wage/sal interval (MQ6P01C) flag = 1 else if the income and income interval is refused and/or DK member n 1-16 wage/sal income (constructed) = 0 flag = 2 in all other cases if there is a household member and no income matched to them member n income (constructed) = 0 flag = 0 if there is no household member, or the respondent is not a householder or cohabitor member n income (constructed) = 999999 flag = 9 -- Under what conditions will others' incomes be reported -- If the respondent is a householder or cohabitor, there will be real amounts in the fields of variables that represent people in the household. If the respondent is not a householder or cohabitor, all fields for other household members will be filled with 9's, including aggregate incomes. If there is not at least one person related to the Respondent, then the family total, other family members total, earnings, property, public assistance, transfer payments, and other income totals will be 9'd. Jointly Received Incomes These are not aggregated, but pulled directly from the original data. If the respondent is a householder or cohabitor, the following incomes will be valid. (Invalid variables = 999999). Income received by respondent and spouse jointly: - this includes joint SS, joint stock/assets and joint other Income received by all household members jointly: - this includes Public Assistance Income received in some other combination: - this includes all income types that individuals may receive The construction of Joint incomes is similar to the construction of individual incomes. If the person number of the person that received the income is 66, then the respondent and spouse receive it; if it is 77, then it is received in some other combination of the household members. Wage/salary and self-employed income cannot be received jointly. Example: Social Security received jointly by respondent and spouse Joint incomes received if the person number of the household member(s) that receives the income is 66 (MQ9AP01 = 66) and if the income is not refused and not DK (MQ7P01 ne 999997 or 999998) R/S joint SS income (constructed) = social security income (MQ12AP1) flag = 0 else if the income is refused or DK and interval is dollar value R/S joint SS income (constructed) = social security interval (MQ12AP1C) flag = 1 else if the income and income interval is refused and/or DK R/S joint SS income (constructed) = 0 flag = 2 in all other cases there is someone in the household with which income could be jointly received R/S joint = 0 flag = 0 if there are no other household members with which to receive income jointly R/S joint = 999999 flag = 9 Aggregated Incomes These incomes are constructed. If the respondents is a householder or a cohabitor, the following will be valid. (Invalid variables = 999999). Household Income The sum of all individual and jointly received incomes. Family Income The sum all all related individuals incomes. This includes income jointly received by respondent and spouse, but not the "other combinations" incomes (because they may include non-relatives). Family Income excluding respondent and spouse The same as the previous income minus the respondent and the spouse incomes. Non-Family member total income The total of all non-family member incomes (this does not include jointly received incomes). Total Jointly Received income. The sum of the income recieved jointly by the respondent and spouse Total income received in some other combination The total of incomes received by more than person, but the combination is neither the respondent and spouse alone, or the entire household. -- Family Member Totals For a case to be a "family" case, there must be at least one household member related to the respondent (has a "related to respondent" code < 19). Total Earnings the sum of all family members' wage and self-employed incomes Total Property the sum of all family members' interests and dividends. Total Public Assistance the sum of all family members' public assistance incomes. Total Transfer Payments the sum of all family members' tranfer payments- social security, pensions, other government programs. Total Other Incomes the sum of all family members' child support/alimony, and anything in the "other" category. Best Measures Construction To create the best measure variables, R1 and R2 data were compared and criteria established for choosing the better of the two's reports of household member's incomes. The best report is a record that more accurately reflects the income of the household, family, and their members. A flag variable was created which would indicate which decision rule was invoked in choosing the best measure. This flag is present for all values except the best measure aggregate incomes. What follows is a detailed decription of how the best income measures were contructed: Criteria for selecting: Decision rules for choosing the "best" measures Key: R = Respondent report S = Spouse report f = flag W = wage/salary income SE = self employment income job2 = respondent has 2 jobs NOTE: R refers to either the primary respondent or the current spouse/partner- R is the self-reported income and S is what the R's spouse reported for R's income. PRIMARY RESPONDENT/CURRENT SPOUSE WAGE/SALARY Summary: The default income is always that reported by the respondent. The spouse's report of a wage salary income is preferred only when: 1) R reports an interval and S reports a dollar value 2) R reports nothing/dk/refused for their own W and they do not have a second job and S reports dollar value 3) R reports nothing/dk/refused for their own W and R does have a second job. S accounts for BOTH incomes. 4) There is no respondent interview and there is a spouse interview. * Assigning flag values take R's when Default: R>0 and S>0 flag = 0 R=S and R>0 and Rf=0 and Sf=0 flag = 1 R=S and R>0 and Rf=1 and Sf=1 flag = 3 R>0 and Rf=0 and Sf=1 flag = 4 Rf=1 and Sf= 1 and R ne S flag = 5 R>0 and Rf=0/1 and S=0 and Sf=0 flag = 6 R>0 and Rf=0/1 and Sf=2 flag = 7 R>0 and there is no spouse interview flag = 8 take S's when Rf=1 and Sf=0 and S>0 flag = 4 R=0 and Rf=0 and S>0 and Sf=0/1 flag = 6 R=0 and Rf=0 and S>0 and Sf=0/1 and SSE>0 Job2 flag = 6 R=0 and Rf=2 and S>0 and Sf=0/1 and RSE>0 Job2 flag = 7 R=0 and Rf=2 and S>0 and Sf=0/1 and RSE=0 flag = 7 S>0 and there is no R interview flag = 8 RESPONDENT/SPOUSE SELF-EMPLOYMENT Summary: The default income is always that reported by the respondent. The spouse's report of a wage salary income is preferred only when: 1) R reports interval and S reports dollar value 2) R reports nothing/dk/refused and S reports dollar value/interval 3) If R reports nothing for W and dk/refused for SE and Spouse reports nothing for SE and a dollar value/interval for R's W, then RSE gets SW. * Assigning flag values take R's when Default: R>0 and S>0 flag = 0 R=S and R>0 and Rf=0 and Sf=0 flag = 1 R=S and R>0 and Rf=1 and Sf=1 flag = 3 R>0 and Rf=0 and Sf=1 flag = 4 Rf=1 and Sf= 1 and R ne S flag = 5 R>0 and Rf=0/1 and S=0 and Sf=0 flag = 6 R>0 and Rf=0/1 and Sf=2 flag = 7 R>0 and there is no spouse interview flag = 8 take S's when Rf=1 S>0 Sf=0 flag = 4 R=0 Rf=0 S>0 Sf=0 caser=1 flag = 6 R=0 Rf=0 R=0 S>0 Sf=1 flag = 6 Rf=2 S>0 Sf=0/1 flag = 7 Best R gets the spouse's report of R's wage/salary when: RW=0 RWf=0 Rf=2 S=0 Sf=0 SW>0 SWF=0/1 flag = 7 RESPONDENT/SPOUSE ALL OTHER INCOMES Summary: The default income is always that reported by the respondent. The spouse's report of a wage salary income is preferred only when: 1) R reports interval and S reports dollar value 2) R reports nothing/dk/refused and S reports dollar value/interval * Assigning flag values take R's when Default: R>0 and S>0 flag = 0 R=S and R>0 and Rf=0 and Sf=0 flag = 1 R=S and R>0 and Rf=1 and Sf=1 flag = 3 R>0 and Rf=0 and Sf=1 flag = 4 Rf=1 and Sf= 1 and R ne S flag = 5 R>0 and Rf=0/1 and S=0 and Sf=0 flag = 6 R>0 and Rf=0/1 and Sf=2 flag = 7 R>0 and there is no spouse interview flag = 8 take S's when Rf=1 S>0 Sf=0 flag = 4 R=0 Rf=0 S>0 Sf=0 caser=1 flag = 6 R=0 Rf=0 S>0 Sf=1 flag = 6 Rf=2 S>0 Sf=0/1 flag = 7 INCOMES OF HOUSEHOLD MEMEBERS OTHER THAN R AND S The decision rules for which of the R and S reports of other household member incomes is preferred are the same as the decision rules for the Joint/Other incomes above. take R's when: R and S report the same dollar value flag = 1 R and S report the same interval flag = 3 take average of R's and S's: R and S report different dollar values flag = 2 R and S report different intervals flag = 5 take the other's report when: one reports interval and other reports dollar value flag = 4 one reports zero and other reports a dollar value/int flag = 6 one reports dk/refused and other reports dollar value/int flag = 7 one interview was not done and the other was flag = 8 JOINT INCOMES/OTHER COMBINATIONS NOTE: The terms "one" and "other" in this section and the following refer to either the R or S. If one = R, then other = S. take R's when: R and S report the same dollar values flag = 1 R and S report the same intervals flag = 3 take average of R and S when: R and S report different dollar values flag = 2 R and S report different intervals flag = 5 take the other's report when: one reports interval and other reports dollar value flag = 4 one reports zero and other reports a dollar value/int flag = 6 one reports dk/refused and other reports dollar value/int flag = 7 one interview was not done and the other was flag = 8 The best aggregate measures are constructed in the same way as the regular constructed aggregate incomes, using the individuals' best measures incomes. In cases where there is a discrepancy between the reports of the respondent and spouse, the discrepant individual(s) are included in the best measures aggregate incomes. NSFH2/February 95 APPENDIX K - CAUSE OF DEATH CODES These codes are used with variables: Main: MI166 Proxy: ZZ8 01 Heart disease, heart attack 02 Cancer (malignancy, leukemia) 03 Stroke 04 Accident 06 Pneumonia/influenza 07 Homicide/murder 08 Suicide 09 Diabetes 10 Cirrhosis of liver 11 Emphysema, chronic bronchitis, or asthma 12 Surgery 13 Hepatitis 14 Perforated esophagus 15 Multiple Sclerosis 16 Alzheimer's 17 Infections 18 Liver failure 19 Kidney failure 20 Food poisoning 21 Blood clot 22 Aneurysm 23 Multi-system failure 24 AIDS 25 Liver disease 26 Old age 27 Aplastic anemia 28 Peritonitis 29 Parkinson's disease 30 Cerebral hemorrhage 31 Natural causes 32 Stomach problems, intestinal blockage 33 Breathing problems 34 Blood poisoning, bleeding disorder, blood disorder 35 Brain damage 36 Muscular Dystrophy 37 Sudden death syndrome 38 Dehydration, malnutrition 77 Some other cause (SPECIFY) 97 Refused 98 Don't know 99 No answer NSFH2/February 95 APPENDIX L - RELATIONSHIP CODES IN PROXY INTERVIEW These codes are used with the following variables in the Proxy Interview: ZZ6, ZZ16, ZZ19, ZZ31P01-ZZ31P08, ZZ41, ZZ46, ZZ55. Relationship to Primary Respondent 01 Spouse/Partner 02 Son/Stepson 03 Daughter/Stepdaughter 04 Son-in-law 05 Daughter-in-law 06 Father 07 Mother 08 Father-in-law 09 Mother-in-law 10 Brother 11 Sister 12 Brother-in-law 13 Sister-in-law 14 Grandson 15 Granddaughter 16 Male friend 17 Female 18 Grandchild/children (sex unspecified) 19 Great grandchildren (sex unspecified) 20 Companion 21 Neighbor 22 Lawyer 23 Niece 24 Nurse 25 Nursing Home Staff 26 Nephew 27 Housekeeper 28 Caregiver 29 Church 30 Friend (sex unspecified) 31 Child/children (sex unspecified) 32 Home Health Care Aid 33 Grandmother 34 Female cousin 35 Aunt 36 Veterans hospital staff 37 Uncle 38 Co-administrator of estate 39 Godson 40 Pastor 77 Other (SPECIFY) 99 No answer NSFH/February 95 APPENDIX M SHOWCARDS USED WITH MAIN INTERVIEW CARD 1 Question: MB8 Relationship Husband or wife Lover/partner Biological child Step-child Adopted child Foster child Child of lover/partner Son- or daughter-in-law Mother or father Step-parent Mother- or father-in-law or partner's parent Grandparent Brother or sister Step-brother or step-sister Half-brother or half-sister Brother- or sister-in-law Grandchild Other relative Roommate Friend Other non-relative CARD 2 Question: MB19 Work-related separation You or your spouse were in a nursing home You or your spouse were in another medical care facility School-related separation You or your spouse were in the Armed Forces You or your spouse were in a prison or correctional institution Marital problems or conflicts Some other reason - PLEASE TELL THE INTERVIEWER THIS OTHER REASON CARD 3 Questions: MC10, MC11, MC30, MC31, MF26, MF28, MF36, MF38, MF40, MF49, MF50 MF57, MF58, MH7, MH8, MH13, MH14, MM35, MM44, MM49, MM50, MM51, MM53, MM126, MM128, MM129, MM130, MM132 Not at all About once a year Several times a year One to three times a month About once a week More than once a week Some other frequency - PLEASE TELL THE INTERVIEWER THE OTHER FREQUENCY CARD 4 Questions: MC20, MC40 WITH WHOM IS CHILD NOW LIVING With other parent With a grandparent With other relatives In a foster home Away at college In an institution or group home, including hospitals and correctional institutions With partner or lover (boyfriend/girlfriend) With a friend In own household, alone or with roommates Other reason--PLEASE TELL THE INTERVIEWER THE OTHER REASON CARD 5 Questions: ME1, ME5, ME8, ME12 LONG-TERM MEDICAL PROBLEMS Diabetes Mental retardation Heart disease Mental illness Cancer, Alzheimer's disease, other than skin senility Respiratory disease Stomach or intestinal emphysema, asthma problems, ulcers, etc. Liver condition Joint disease, arthritis High blood pressure, Loss of sight or hearing hypertension CARD 6 Questions: ME32, ME41 RELATIONSHIP Son Daughter Son-in-law Daughter-in-law Father Mother Father-in-law Mother-in-law Brother Sister Brother-in-law Sister-in-law Grandson Granddaughter Male friend Female friend Other--PLEASE SPECIFY WHO THE OTHER PERSON WAS CARD 7 Questions: MF18, MF22, MF30, MF33, MF42, MF44, MF53, ML87, ML109, ML173 HOW DESCRIBE RELATIONSHIP Really Absolutely Bad Perfect |___|___|___|___|___|___|___|___|___|___| 0 1 2 3 4 5 6 7 8 9 10 CARD 8 Questions: MF63, MF68 REASONS PARENT STARTED LIVING WITH YOU Parent ill/disabled You were ill/disabled Parent had financial problem You had financial problem Parent was providing other assistance to you Other reason--PLEASE TELL THE INTERVIEWER THE OTHER REASON CARD 9 Questions: MF65, MF70 WHERE DID PARENT LIVE Parent died Parent went to own household Parent moved to a nursing home Parent moved in with your brother or sister or other relative Parent moved to some other place--PLEASE TELL THE INTERVIEWER THAT OTHER PLACE CARD 10 Question: MH27 Black White - not of Hispanic origin Mexican American, Chicano, Mexicano Puerto Rican Cuban American Indian Asian Other - PLEASE TELL THE INTERVIEWER WHAT OTHER GROUP BEST DESCRIBES YOU CARD 11 Questions: MI17, MI67, MI97, MI128, MI146, MN11 RELIGION No religion Roman Catholic Jewish Baptist Episcopalian Lutheran Methodist Latter Day Saints (Mormon) Presbyterian United Church of Christ - Congregational Protestant Other - PLEASE TELL THE INTERVIEWER YOUR RELIGIOUS PREFERENCE CARD 12 Questions: MI160, ML114, ML115, ML116, ML117, ML118, ML119 Strongly agree Agree Disagree Strongly disagree CARD 13 Question: MI166 CAUSE OF DEATH Heart disease, heart attack Cancer - malignancy, leukemia Stroke Accident Pneumonia or influenza Homicide or murder Suicide Diabetes Cirrhosis of the liver Emphysema, chronic bronchitis, or asthma Some other cause - PLEASE TELL THE INTERVIEWER WHAT THE OTHER CAUSE OF DEATH WAS CARD 14 Questions: MI176, MI182 RELATIONSHIP Son Son-in-law Daughter Daughter-in-law Brother Brother-in-law Sister Sister-in-law Some other relative - PLEASE TELL THE INTERVIEWER WHO THIS OTHER RELATIVE WAS Neighbor Friend Paid help Some other person - PLEASE TELL THE INTERVIEWER WHO THIS OTHER PERSON WAS CARD 15 Question: ML8 One of the best Above the middle In the middle Below the middle One of the worst CARD 16 Question: ML12 LEVEL OF EDUCATION Not finish high school Graduate from high school Graduate from vocational or trade school Graduate from a two-year junior or community college Complete one to three years of college Graduate from a four-year college or university Complete a master's or doctorate degree Other --PLEASE TELL THE INTERVIEWER ANY OTHER LEVEL CARD 17 Questions: ML19, ML20, ML21, ML22, ML29, ML30, ML31, ML32 Never Less than half the time About half the time More than half the time Always CARD 18 Questions: ML23, ML24 Almost every day Several times a week About once a week Two or three times Once Never CARD 19 Questions: ML66, ML67 None Less than half About half More than half All CARD 20 Questions: ML72, ML74, ML77, ML78, ML79, ML80, ML81, ML82, ML83, ML84, ML85 ML86, ML95, ML96, ML97, ML98, ML99, ML100, ML101, ML102, Ml103 ML104, ML105 Never or rarely Once a month or less Several times a month About once a week Several times a week Almost every day CARD 21 Questions: ML93, ML94, ML107, ML108 Almost every day Several times a week or more About once a week One to three times a month Once a month or less Never CARD 22 Questions: ML110, ML111, ML112, ML113 Never Seldom Sometimes Often Always CARD 23 Questions: ML120, ML121, ML122, ML123, ML124 Extremely unhappy Somewhat unhappy Somewhat happy Extremely happy CARD 24 Questions: ML127, ML128, ML129, Ml130, Ml131 Definately would not do it Probably would not do it Fifty-fifty chance Probably would do it Definately would do it CARD 25 Question: ML132 HOW DESCRIBE RELATIONSHIP Really Absolutely Bad Wonderful |___|___|___|___|___|___|___|___|___|___| 0 1 2 3 4 5 6 7 8 9 10 CARD 26 Questions: ML149, ML150, ML151, ML154, ML155, ML156, ML157, ML158, ML159, ML160 Almost never Less than half the time About half the time More than half the time Almost always CARD 27 Questions: MM33, MM118 ANNUAL INCOME AMOUNTS MONTHLY INCOME AMOUNTS 0--Did not have any income 0--Did not have any income 1--Less than $5,000 per year 1--Less than $417 per month 2--$5,000 - $9,999 per year 2--$417 - $832 per month 3--$10,000 - $19,999 per year 3--$833 - $1667 per month 4--$20,000 - $29,999 per year 4--$1668 - $2500 per month 5--$30,000 - $39,999 per year 5--$2501 - $3332 per month 6--$40,000 - $49,999 per year 6--$3333 - $4166 per month 7--$50,000 or more per year 7--$4167 or more per month CARD 28 Questions: MM41, MM42, MM43, MM124, MM125 Never Seldom Sometimes Very often Always CARD 29 Questions: MM54, MM55, MM56, MM57, MM58, MM59, MM60, MM61, MM62, MM131, MM133, MM134, MM135, MM136, MM137, MM138, MM139 None A little Some Pretty much A great deal CARD 30 Questions: MM63, MM64, MM65, MM140, MM141, MM142 Very satisfied Somewhat satisfied Somewhat dissatisfied Very dissatisfied CARD 31 Questions: MM66, MM143 Very Very Unhappy Happy |___|___|___|___|___|___|___|___|___|___| 0 1 2 3 4 5 6 7 8 9 10 CARD 32 Questions: MN20, MN21M, MN22 TYPE OF SCHOOL Vocational, technical, or trade school Junior or community college Four-year college or university Professional or graduate school requiring prior college work for entrance, such as medical, dental, or law school, or seminary Business college or secretarial or nursing school CARD 33 Question: MN27 DEGREE Associate's Degree (2-year) Bachelor's Degree Master's Degree Doctorate (Ph.D., M.D., LL.D., etc.) Other--PLEASE TELL THE INTERVIEWER ANY OTHER DEGREE OR CERTIFICATES CARD 34 Question: MP98 Your spouse or partner The child's grandparent Other relatives, including brothers or sisters Neighbor or babysitter - NOT a relative Day care center, nursery school or kindergarten You care for your child while working Other - PLEASE TELL THE INTERVIEWER WHO THIS OTHER PERSON IS CARD 35 Questions: MQ46, MQ56, MQ61, MQ65, MQ69, MQ73, MQ77, MQ81 RELATIONSHIP Your parent(s) Mother- or father-in-law, or partner's parent Your brother or sister Your spouse's brother or sister Your grandparent Your spouse's grandparent Your child Your spouse's child Grandchild Other relative Non-relative Other - PLEASE TELL THE INTERVIEWER WHO ELSE HELPED CARD 36 Questions: MQ93, MQ94 TOTAL VALUE 0-- None 1-- $1 to $1,499 2-- $1,500 to $2,999 3-- $3,000 to $4,999 4-- $5,000 to $9,999 5-- $10,000 to $19,999 6-- $20,000 to $49,999 7-- $50,000 to $99,999 8-- $100,000 or more NSFH2/February 96 APPENDIX N - ISR FIELD REPORT INSTITUTE FOR SURVEY RESEARCH TEMPLE UNIVERSITY -Of The Commonwealth System of Higher Education- 1601 North Broad Street Philadelphia, PA 19122 NATIONAL SURVEY OF FAMILIES AND HOUSEHOLDS Study #40-1591-402 1992-1994 FIELD REPORT Prepared by: Elaine Trull Lisa Famularo (Chapter VII) TABLE OF CONTENTS I. Overview of the Project A. The Original Survey, 1987-1988 B. NSFH-II, the First Follow-Up C. Organization of NSFH-II D. Planning for the First CAPI Project 1. Selecting Equipment 2. Planning for Training 3. Developing Materials 4. Pretesting the Instruments 5. Planning for Interviewer Payment E. The NSFH-II Samples 1. The CAPI Sample 2. The CATI Samples 3. The Proxy Interviews 4. The Spanish Interviews II. Role of Regional Field Coordinators A. Role and Responsibilities of Coordinators B. Selection Criteria for Coordinators C. Coordinator Training III. The Coordinator Database A. Objectives of the Coordinator Database B. Using the Mac Database 1. Data Entered By Coordinators 2. Use of the Mac in the Office C. Relationship of the Mac Database to Unify D. Transmitting Information To and From the Office E. Problems Using the Database F. Usefulness of the Coordinator Database IV. Interviewer Staffing for NSFH-II A. Initial Staffing Efforts B. Supplementary Staffing by Conference C. Interviewer Attrition D. Staffing the Cleanup Conference V. Interviewer Training Procedures A. Training Schedule B. Pilot Training C. Interviewer Training D. Adjustments to Training Procedures 1. Evening Practice Sessions 2. Adjustments to the Agenda 3. Development of the Mock 3 Training Module E. Problems Unique to CAPI Training F. Evaluation of Trainees VI. Field Procedures A. Contacting Respondents B. Submitting Completed Work C. Refusal Conversion Efforts D. Tracing the NSFH-II Samples E. Validation of Interviews F. Case Transfers G. Other Field Problems H. Payment of Incentives I. Managing the Equipment Inventory J. Repair of Machines K. Using the Hotline L. Payment for Interviews VII. CATI Procedures A. Timing of CATI Interviews With Parents & Children B. Timing of CATI Interviews With Main Rs & Spouses C. Overview of the CATI Study D. CATI Staffing 1. Selection of CATI Supervisors 2. Recruiting & Selecting CATI Interviewers E. CATI Training 1. Training on the Parent & Child Questionnaire 2. Training on the Main Questionnaire 3. Proxy Training 4. CATI Interviewer Trainings F. Preparing Parent and Child Sample 1. Contact Information Sheet (CIS) 2. Call Report Form (CRF) 3. Parent/Child Address Form 4. Parental Consent Form for Child's Participation G. Preparing Main Respondent & Spouse/Partner Sample H. Data Collection 1. Assigning Cases 2. Handling Cases With Non-Interview Results 3. Converting Parent and Child Refusals 4. Problems Caused by Data Entry Errors 5. Tracing Parent and Child Respondents 6. Parent & Child Respondents Without Telephone 7. Spanish Language Data Collection I. Monitoring of Telephone Interviews J. Telephone Interviewer Payment and Bonus K. Differences Between CATI and CAPI 1. Materials 2. Focus of Training 3. Procedures L. Coordination of Work Between Field and Phone M. Final Results N. Problems and Their Solutions 1. Dissemination of Information 2. Motivating Interviewers VIII. The Cleanup Phase of the Study A. Changes Made During Cleanup B. The Big Push--January through March, 1994 C. Payment for Cleanup D. Cleanup Problems and Their Solutions I. OVERVIEW OF THE PROJECT The Institute for Survey Research conducted the first follow-up to the 1987-88 National Survey of Families and Households for the University of Wisconsin. The field interviews began with a pilot study in July 1992 and continued in three waves plus a cleanup stage through May 1994. Telephone interviewers at ISR began working in December 1992 and finished interviews with parents and children in August 1994. The National Institutes of Health, specifically the National Institute on Child Health and Human Development and the National Institute on Aging, provided the funding. Principal investigators are Dr. James Sweet and Dr. Larry Bumpass of the University of Wisconsin at Madison. Laptop computers were used for the first time by ISR to simplify the interviewing process and eliminate some of the time-consuming data entry tasks. A. The Original Survey, 1987-1988 The National Survey of Families and Households (NSFH-I) began with the screening of randomly-selected households in the forty eight coterminous states in 1987-1988. Interviews were conducted in person with one eligible respondent age nineteen or older in each main sample household. In addition, respondents were also selected from an oversample, so that a double sample of blacks, Puerto Ricans, Mexican Americans, single-parent families, families with step-children, cohabiting couples and recently married persons were interviewed. Self-administered questionnaires, shorter than the main respondent interviews, were collected from a current spouse or partner if there was one and, in some cases, from another adult relative, the "householder." A total of 13,017 main respondents were interviewed. Data collection for the original survey began in March 1987 and continued through May 1988. The study was designed by a University of Wisconsin team of eight researchers with closely related interests in American family life. The resulting study examined many aspects of family life, from current behavior and retrospective family history to relationships, interactions, attitudes and feelings. Key variables were measured in unusual detail in NSFH-1, and both spouse/partners were interviewed to facilitate the understanding of relation- ships and decision processes. Another priority of the study was to provide a data resource for the research community at large. The public use data file and complete document- ation were made available within six months of the end of the field period. Researchers from over 150 colleges and universities are currently analyzing the data provided by Wave 1. B. NSFH-II, the First Follow-Up The National Survey of Families and Households-Wave II (NSFH-II) provides detailed longitudinal data on 13,014 Wave I households. (Three households were deleted from the sample at the request of the principal investigator because of the respondents' fears of family violence if inter- viewed again.) The content of the interview was again broad, including an update of family history since the first interview, quality of relationships between spouse/partners, parenting practices and parent-child relationships, relationships between parents and their absent minor children, relationships of parents with young adult children, health and well-being, kinship and social support, attitudes and opinions about family life, and economic situation. The number of interviews was expanded to include face-to-face inter- views with the main respondent's current spouse/partner. (In Wave 1, spouses had been given a self-administered questionnaire that was shorter than the main respondent interview.) When a relationship had been terminated since the first interview, we conducted a personal interview with the ex-spouse/ partner also. New spouses or partners currently living with the main respondent became respondents also. The "focal children" aged 10 - 24 from NSFH-1 were interviewed by telephone from ISR's telephone facility. (Detailed information about these children was collected from the main respondent and spouse/partner as part of Wave I.) One randomly selected parent of the main respondent was interviewed by phone as well. C. Organization of NSFH-II Since interviews were to be conducted with laptop computers, planning and implementation of NSFH-II was managed by a team of personnel from several ISR departments. The original Project Team consisted of Allan Walters, Marcia Murphy, Dolores Williamson, Christine Schuler, Tony DeFusco, and Elaine Trull. Christine returned to school in the fall of 1992, and three others--Allan, Dolores, and Tony---left ISR in the summer of 1993. Additions to the project for the remainder of the time included Karl Landis, Ellin Spector, and Carolyn Rahe. The parts played by these and other study personnel in the completion of the National Survey of Families and Households-Wave II are described below: Allan Walters. Study Director of the 1987-1988 NSFH and NSFH-II. Responsible for initial planning and management, helped design the Coordinator database. Ellin Spector. Interim Study Director, managed project from July 1993 to September 1993. Remained a part of the staff until study ended. Karl Landis. Study Director beginning in October 1993. Managed the cleanup phase and completion of the project. Marcia Murphy. Liaison with the principal investigator on the questionnaire program, developed or oversaw the development of all in-house computer programs and utilities, wrote program documentation, major role in planning and implementing training. Christine Schuler. Helped with initial staffing, debugging the questionnaire, and training. Tony DeFusco. Responsible for translation of materials into Spanish, helped develop forms, Coordinator for Region 11. Dolores Williamson. Made travel and conference arrangements, supervised in-house clerical staff, in charge of initial staffing, supervised Co- ordinators and interviewers for northern and western half of the country. Elaine Trull. Wrote most of the training materials, responsible for staffing in eastern and southern half of the country, supervised Coordinators and interviewers in one half of the country, major role in training. Carolyn Rahe. Supervised in-house staff during cleanup stage, Field Administrator for last months of the study. Yvonne Shands. Maintained the Coordinator database in-house, major role in overseeing and doing case transfers, major role in tracing in-house, trained and supervised proxy interviewers and telephone tracers. Lisa Famularo. Manager of telephone study, major role in training, in- house trouble shooting, supervised Hotline. David Tucker. Responsible for data transmission, prepared weekly management reports, helped develop utility programs for use by interviewers, created assignments, responsible for repair of equipment, handled transmission problems referred from the Hotline. Marian Anderson. Supervised check-in, check coding, editing, data entry for Self Enumerated booklets and handled problems with respondent incentives. Initially sent out all assignments and computers. Cynthia Geesey. Provided one-on-one help with tracing to Coordinators and interviewers during cleanup stage of the study. Ellen Greenberg. Helped develop and supervised Case Transfer system until July 1993. Telephone supervisors were Tanya Williams, Amanda George, Saskia Dowell, Ellen Greenberg, and Lakeisha Bey. Substitutes were Kristin Stanley and Linda Shanahan. D. Planning for the First CAPI Project While the Institute had been conducting in-house telephone interviews using Computer Assisted Telephone Interviewing (CATI) technology for a number of years, NSFH-II was ISR's first study to implement Computer Assisted Personal Interviewing (CAPI) technology with laptop computers. Weekly staff meetings of the Project Team began in November 1991. Prior to that, extensive work had been done on the questionnaire and the Co- ordinator database, prospective Coordinators had been contacted, and initial planning had begun for the management of the project in the field. 1. Selecting Equipment The pretests provided an excellent opportunity to try out different laptop computers before the final selection of a brand and model for the project was necessary. Pretest interviewers took turns using the machines, so that each had experience with all the machines to report at the debrief- ings. The laptop selected was the Digital model 320P, which seemed desirable because of its longer cord and a screen that was low enough to allow inter- viewers to have more eye contact with respondents than was possible with other laptops that were considered. The machine weighed 6.7 pounds, measured 12 3/16 x 10 x 1 13/16 inches, and had a nickel-cadmium rechargeable battery with a charge life of approximately two and one half hours. The laptop had a 386 processor running at 20 MHz with 2MB of RAM. An external modem, the Codex 3220 Plus, was used instead of the 2400 baud internal modem available with the machines. The Codex's transmission speed, at 9600 baud, was dramatically faster than the internal modem's. Fifty extra batteries were ordered to replace batteries that failed. To keep procedures simple, the decision was made to give interviewers their assignments on floppy disks, so that they could load them into the machine, but retain the floppy disk in case of machine failure. The Macintosh computer was selected for the Coordinator database because it was easy to learn to use and user-friendly. The model chosen was the Macintosh Classic II. 2. Planning for Training The project team felt that quality training was essential to the success of the project, so a great deal of discussion and thought were given to how the training should be structured and conducted. The following decisions were made: Three day conferences were held on Thursday through Saturday. Each day's training ran from 8:30 a.m. to 5:00 p.m. The size of the group of trainees should be limited to 20 to 25 for the first trainings, with 30 the maximum number for the later trainings, in order to accurately monitor each interviewer's progress. The training would be centered around mock interviews, so that trainees received as much actual practice with the laptops as possible. We decided to conduct mocks in one large group, rather than dividing the trainees into several small groups, so that the training would be more consistent. Coordinators, who normally conduct small group mocks, would not have had enough experience with the instrument and laptop by the time they would have had to begin conducting small groups on their own. The room would always be arranged in a U shape, with adequate room for trainers to walk behind the interviewers to observe their performances. Three trainers were scheduled for each conference, with two represent- atives from the office and one Coordinator handling the responsibilities. We decided to use a Home Study with an accompanying quiz to cover some of the field procedure training items usually covered at a conference, so that more time could be devoted to actual practice with the laptops and to the CAPI specific training needed (transmitting, programs, etc.) The Home Study would cover contacting households, avoiding bias, probing correctly, and questionnaire conventions. Laptops would be shipped to interviewers one week before training, allowing them time to try out the machines, and they would bring their laptops with them to the training conferences. Because we shipped to their homes, they would have the box and packing materials needed to return the machines when they finished their fieldwork. Most importantly, these plans would alleviate some of our concerns about the safety of machines stored at hotels. We planned to build into the training efforts to boost interviewer confidence. Knowing that many of the interviewers lacked experience with CAPI, computers, and sometimes even typing, we included confidence-building material in the interviewer's manual. At training conferences, we would emphasize our beliefs that trainees were doing well and were quite capable of learning the new method of interviewing. To alleviate anxiety and provide computer assistance to interviewers in the field, we planned to establish a Hotline. In addition to daytime hours, the telephone supervisors would be available to take calls in the late after- noons, nights, and on weekends, so that the Hotline would be covered as many hours as possible. Those who answered the Hotline would be trained to guide interviewers in correcting their own errors with the questionnaire, as well as to help with transmitting interviews or loading (and using) programs. We decided to have interviewers do the tutorial for the CAPI program at the beginning of the conference where we could help them, rather than risk confusing and discouraging the inexperienced interviewers by having them do the tutorial at home. In lieu of the tutorial, we devised a very simple set of instructions to help interviewers become familiar with the laptop. Each interviewer would satisfactorily complete two practice interviews with his/her Coordinator by telephone before beginning work on an actual assignment. We decided to begin slowly with a pilot conference to test procedures and the questionnaire in July 1992 before starting a full schedule of training two months later. After that, two training conferences a month were scheduled for most of the training period. Three conferences were planned for February 1993. 3. Developing Materials The most important of the materials was the CAPI instrument, developed by Jim Sweet with help from others at the University of Wisconsin. Many of the Wave I questions were used. The questionnaire was written as a series of modules, then linked together. Marcia Murphy at ISR polished the instrument, correcting the program, and Marcia and Jim together resolved the more difficult questions. The Self Enumerated booklet was also completed by University of Wisconsin personnel, with few changes from the 1987 version. The booklet, since it was to be self-administered, was the one part of the interview that was completed on paper. We determined that a separate Computer Guide was needed so that inter- viewers could have a thin manual to carry with them in their computer bags. This manual would provide the documentation for the programs we used and be a first source of help for computer problems, before calling the Hotline. We decided to incorporate the explanations, guidelines, and reminders that are normally included in a written manual as question-by-question instructions, or QxQs, into the questions on the computer screen. We felt that interviewers would have immediate access to guidelines this way, and that they would be more likely to use them than if they had to look them up in a separate manual. We also doubted that they would actually carry the QxQs with them, in addition to the Computer Guide and laptop. When we decided to have interviewers do the tutorial at the training conferences rather than at home, we then needed to have some information to send them with the laptop. The resulting introduction to the laptop consisted of instructions for such basics as adjusting the screen and a series of steps to familiarize them with the machine. The Interviewer's Manual we developed was much like a typical manual, without the QxQs. The Coordinator's Manual was divided into two major sections: a detailed description of Coordinator duties and instruction on using the Mac database. To facilitate training, the field portions of the conference were scripted as a Trainer's Guide. We developed a tracing form as an integral part of the Screening/Call Report Form for the study. The tracing form allowed interviewers to list their tracing steps on the screening form instead of on a separate sheet. The mocks for training were an outgrowth of the pretest interviews, with some changes to allow for a wider range of family experiences in the mocks. The mock interviews were entered into the computer, then both questions and answers were printed out, so that trainers had an exact copy of the questions that should be displayed on interviewers' computer screens. Interviewers were instructed to interview main respondents first. To allow them more flexibility when the main respondent wasn't home, we devised a Current Spouse Worksheet to learn about the main respondent's current marital status. Interviewers could use the form to determine the correct version of the questionnaire to administer and the correct Screening Form to use if a spouse respondent answered the door and the main respondent was not available. The other materials developed for the study were similar to those used on non-CAPI studies. A complete list and description of study materials is included in the Appendix. 4. Pretesting the Instruments Four pretests were used to test the main respondent and spouse inter- views. Questions were evaluated for clarity and ease of understanding. Because skip patterns were determined by the program, finding errors in the skip logic was more difficult than usual. Finding and correcting minor errors in the programming was a major part of the pretest debriefings. The separate modules of the questionnaire were first linked together to form one questionnaire in January 1992. Before that, ISR staff tried out the separate modules in December 1991. Pre-testers and staff interviewed them- selves as part of their preparation for the project. To pretest, interviewers first had to conduct a brief five minute interview by telephone to determine the respondent's situation in 1987. This information was loaded into the computer as case files, so that pretest interviews would be comparable to actual interviews containing data from Time 1. Pretest 1 took place between February 16 and February 27, 1992. Three experienced ISR CATI interviewers, Yvonne Shands, Justine Ganatra, and Lisa Famularo, were the pretest interviewers. For this initial trial of the questionnaire, the interviewers were instructed to find respondents who had eventful lives, with changes in such things as marital status, employment, and number of children in the previous five years. All were pretested as main respondents. Each interviewer conducted three to four interviews on each pretest. All had problems with the program on Pretest 1, reaching points in an inter- view where the program went in a loop or they were otherwise not able to continue the interview. The debriefing discussion revealed a variety of problems. The same interviewers worked on Pretest 2, from March 29, 1992, to April 1. A number of minor problems with the questionnaire continued to surface. In this and all of the debriefings, much of the time was spent isolating and correcting programming errors and typos. Because many of the same questions were used on the original NSFH survey, considering question content was not a major part of the debriefings. For Pretest 3, held May 6 - 14, the principal investigator suggested types of respondents that would be useful to pretest. We looked for younger respondents who had moved in or out of the parental household and older respondents with several grown children. As a result of the debriefing, other problems with the skip patterns and the wording of the questionnaire were resolved. We wanted to use interviewers from the local area, who did not have ISR CATI experience, to conduct Pretest 3. Mary Jane DiFresca, Barbara Jonas, and Grace Pasco participated. Although Grace had worked as a CATI inter- viewer at ISR, she was thought to be an average field interviewer and of the age of many of the interviewers who would work on the study. The final pretest of the main respondent and spouse interviews was conducted June 4 - June 12, 1992. Interviewers were Yvonne Shands, Lisa Famularo, and Mary Jane DiFresca. A special effort was made to include respondents who had divorced and remarried since Time 1, and respondents whose children lived elsewhere with the other parent, because those sections of the questionnaire had not been adequately pretested. Training matters were emphasized during the discussion at the debriefing, and most of the earlier problems with the questionnaire seemed to have been resolved. On average, the pretests indicated that the length of the interview was slightly longer than the 75 minutes (main respondent), 60 minutes (1987 spouse), or 45 minutes (new spouse/partner) estimated in the study proposal. The first pretest of the three telephone interviews--for parents, older children, and younger children--was conducted in September 1992. The inter- viewers were ISR experienced CATI interviewers--Lisa Famularo, Yvonne Shands, and Amanda George. Interviewers completed seven interviews, three with parents, and two each with younger and older children. For the second pretest of the telephone component, also in September 1992, Yvonne Shands, Lisa Famularo, Ellen Greenberg, and Amanda George conducted interviews. The parent interview was so similar to the main respondent interview that few problems were found during the pretests. The pretests of the older and younger childrens' interviews, however, revealed the need for additional or "other" categories to accommodate actual answers that children gave. Some of the questions were clarified as a result of the comments that children made. The Spanish materials were given a brief trial by ISR bilingual staff members. 5. Planning for Interviewer Payment In an attempt to standardize and control the charges for interviewers traveling over 50 miles roundtrip from home, we established a payment system for interviews that provided increased flat rate compensation for the time spent traveling to cases farther away. The roundtrip distance from the interviewer's home to the respondent's home would have to be entered into the computer during each interview and also recorded on the payment record. The roundtrip mileage figure was the basis for payment. (A complete listing of the payment scale is in the Interviewer Availability Notice included in the Appendix.) E. The NSFH-II Samples The second wave of the National Survey of Families and Households included as many as five interviews relating to one main respondent. 1. The CAPI Sample The CAPI sample included main respondents, 1987-88 spouse/partners, and new spouses or partners of our main respondents. The main respondents were the 13,014 respondents who were interviewed in person during the original survey. Of the 1987-88 spouse/partners, 4,508 were still living with and/or married to the NSFH-I main respondents. Another 791 were ex-spouses or partners, and 390 were deceased. All 1987-88 spouse/ partners remained part of the sample, although not all of them had actually completed a self-administered questionnaire during Wave 1. The sample included 1131 new spouses or partners of main respondents. By definition, all new spouse/partners had to be currently living with main respondents. Spouse/partners acquired by main respondents since the first wave of the study, but no longer living with them, were not included in the sample. 2. The CATI Samples There was also a main respondent (R), 1987-88 spouse/partner (S), and new spouse/partner (N) component of the CATI study. The study was planned to allow telephone interviews of main respondent and spouse cases when these individuals lived overseas or more than 75 miles away from a field inter- viewer. When telephone interviewers were first trained for these cases in March 1993, R and S cases were assigned to the telephone facility. As the study progressed, other field cases were assigned to telephone interviewers for attempts at refusal conversion and to increase production at the end of the study. During the 1987-88 interview, both main respondents and their spouse/ partners were asked a series of questions about one randomly selected child in the family. These focal children were followed in NSFH-II by again asking main respondents and spouses a series of questions about them. They were also interviewed themselves by telephone if they were between 10 and 24 years old. The 1,436 children aged 10 - 17 were identified as "Younger Children", while the 1,152 18 - 24 year olds comprised the "Older Children" sample. To get the viewpoint of yet another generation, one parent of each main respondent was randomly selected by computer to be the parent respondent. If the main respondent refused to allow us to contact the selected parent for the interview, the questionnaire program prompted the interviewer to ask for permission to interview the other parent. The telephone interview with the parents was somewhat shorter than, but very similar in content to, the main respondent interview. A total of 4,257 parents were eligible to be inter- viewed. There were a few cases where parents were both parent and main or 1987 spouse respondents. This situation occurred when parents and their adult children both lived within the same Time 1 listing area, for example. We tried to interview them for both samples, and most, but not all, cooperated. Similarly, we identified 17 situations where a selected parent had two child- ren who were main respondents in the study, and we asked for their coopera- tion as parent respondents of both children. 3. The Proxy Interviews The final type of interview was the short 10 - 15 minute proxy interview conducted with a family member when the main respondent was either deceased or too ill to be interviewed. A total of 1039 main respondents were ident- ified as either deceased (763) or too ill to be interviewed (276). Proxy interviews were included in the total number of interviews conducted with main respondents. 4. The Spanish Interviews Like the Time l interview, this survey could also be conducted in either English or Spanish. Spanish interviews were done by 25 bilingual field and telephone interviewers. The main respondent, 1987 spouse, new spouse and parent interviews were all translated into Spanish, but interviews in Spanish were not needed for children. II. ROLE OF REGIONAL FIELD COORDINATORS Coordinators were envisioned as an especially important link between the office and interviewers in ISR's first CAPI project. We depended upon them to help alleviate the interviewers' computer fears, provide encouragement and support, and to keep careful track of the large amount of sample in their areas. A. Role and Responsibilities of Coordinators The role of Coordinators in the National Survey of Families and House- holds was that of liaison between the office and interviewers. We expected interviewers to turn first to their Coordinators for assistance or encourage- ment. Coordinators' primary responsibilities included report-taking, assist- ance with the training of interviewers, and giving information, advice, and encouragement. Before each training conference, Coordinators were to call each inter- viewer who was scheduled to attend, to make sure he/she had received plane tickets and laptop computer, as appropriate. The calls allowed Coordinators to introduce themselves before the training, gave interviewers opportunity to ask questions, and helped ensure that the interviewer and the computer arrived at training as scheduled. At training conferences, Coordinators were expected to get to know their interviewers and schedule times for the practice interviews and reports. They usually acted as respondents for one or two of the mocks at training and helped with extra practice sessions in the evenings. They circulated in the training room to help interviewers, and called the regular trainers over when they weren't sure of a procedure. In the week after each training, Coordinators were expected to do two practice interviews over the telephone with each of their newly trained inter- viewers. In these practices, Coordinators acted as respondents by following a script, and helped interviewers only when they could not figure out the solution themselves. When a Coordinator had a over 10 interviewers at one training, another Coordinator usually helped with the mock interviews so that all practices could be completed within the week. When taking weekly reports, Coordinators were expected to enter each contact made during the week into their computers. The only contacts not recorded were telephone contacts that yielded only busy signals, no answers or answering machines. Coordinators were required to record respondents' telephone numbers and birthdates, and changes of addresses and names. We expected them to enter all reporting information directly into the MAC, rather than recording contacts on paper and risking mistakes or omissions by trans- ferring data later. Coordinators reported weekly to their Field Administrator, to discuss problems and production in their areas. They also telephoned as needed with questions and problems. We expected Coordinators to do at least five actual interviews on the project for practice, but they were welcome to take a regular assignment as well. Seven Coordinators accepted interviewing assignments, four did not. B. Selection Criteria for Coordinators We began the selection of Coordinators by making two lists: the criteria we felt were important, and the individuals who had worked for ISR that we felt should be considered as meeting those criteria. The Field Administrators contacted all those listed as potential Coordinators in October and November 1991 to find out both their availability for the extended time the study would be in the field, and to learn more details about their CAPI experience, if any. All of the following criteria were considered most important in selecting Coordinators for the study: * experience as a CAPI supervisor * experience with CAPI as an interviewer * availability for the 15 month period * proximity to the Region to be supervised * ability to help with training * ability to motivate and get the work done * availability of enough time each week to devote to the study * ability and desire to keep careful records * ability to correctly answer questions about study and procedures. The Project Team selected the final group of eleven Coordinators after hearing reports about the initial contacts with 22 potential candidates. Both those selected and not selected were notified by the Field Administrators by telephone. Coordinator Availability Memos were sent to the selected Co- ordinators in February 1992. TABLE 1: COORDINATORS AND THEIR REGIONS Region 1 Joan Hopner New England, Upstate NY, Florida Region 2 Marlene Blomberg Wisconsin, OH, MI Region 3 Sandy Lewis Kansas, MO, OK, So. IL Region 4 Linda DeWild Iowa, No. Ilinois, IN Region 5 Evelyn Massaroni * Baltimore, DC, VA, WV, KY, Rhona Collins Pittsburgh Region 6 Sandy McGuire ** North Carolina, SC, GA, Elaine Trull part of TN Region 7 Barbara Sukman Louisiana, TN, AL, AK Region 8 Rosa Dwyer Texas, Arizona Region 9 Maria DiGregorio No. California, OR, WA, ID Region 10 Marilyn Gunn So. California, UT, WY Region 11 Tony DeFusco *** Pennsylvania, NJ, Lisa Famularo + New York City * Dropped out in Sept. 1992 because of health problems ** Abandoned project after training *** Coordinator until July 1993 + Coordinator from January 1993 until August 1993 Field Administrator Dolores Williamson supervised Regions 1, 2, 4, 9, 10, and 11. Field Administrator Elaine Trull supervised Regions 3, 5, 6, 7, and 8. C. Coordinator Training Allan Walters and Elaine Trull trained Coordinators in a two day session in conjunction with the pilot training for interviewers and Coordinators in July 1992, with additional training taking place in two evening sessions later in the training week. We sent each Coordinator his/her Coordinator's Manual ahead of time, along with the Interviewer's Home Study. We expected them to at least read the manual carefully before the conference. However, only about one third seemed to have prepared well for training, which was disappointing, consider- ing the new techniques they had to master. Coordinators spent most of their time learning and practicing with a test database loaded into their MACs. Each Coordinator, working with his/her own MAC, completed the tutorial that came with the machine. Sessions on Co- ordinator duties and field procedures were interspersed with teaching and practice sessions on the MAC. We encouraged Coordinators to call with questions as they began using the Mac database. They were to call Allan Walters directly. III. THE COORDINATOR DATABASE We selected the Macintosh system, known to be user-friendly and easy for novices to learn, for use with a database maintained by Coordinators to manage the project in the field. We supplied the Coordinators with a Mac- intosh Classic II and installed a larger capacity IIi Mac Classic in the office to handle the consolidated central office database. Using a computer system gave Coordinators easy access to the large amount of sample they were handling, but having a separate database from Unify meant that Coordinators were not able to directly access or change office records. The database was set up using the Omnis 7 data management program. A. Objectives of the Coordinator Database Our primary objectives for using the Coordinator database were the following: * To allow Coordinators to manage and easily access information about the large number of cases in their coordinating assignments * To replicate the detailed record of respondent contacts from the Call Report Form to the Coordinator database, so that Coordinators could use this information in guiding interviewers * To enable both Coordinators and the central office to track in- formation on the Call Report Form more quickly than relying on paper copies sent to the office * To allow Coordinators to generate their own on-screen reports as needed * By using the Macintosh, to provide a system that was easy to learn and easy to use for data management. B. Using the Mac Database The Macintosh database was set up to be used both by Coordinators in the field and by the office staff. A password system limited Coordinator access to certain parts of the program, while allowing more extensive use by those in the office. The Coordinators knew they were not able to access menu items that were grayed out, rather than black. 1. Data Entered by Coordinators Coordinators worked from a Home screen and six basic screens labeled Co- ordinators, Interviewers, Respondents, Contacts, Reports, and Help. They utilized the Contacts screen for most of the work, especially when taking reports. To take a report, Coordinators began by going to the Contact screen, selecting the "find interviewer" option, and bringing up first an interviewer, then an alphabetical list of that interviewer's respondents. For each case the interviewer had worked on that week, he/she reported each contact made. Exceptions were telephone attempts that yielded no results other than busy signals, no answers, answering machines, and multiple visits to a household that resulted in an interview within that week. Only the final contact resulting in the interview was recorded if the interviewer had made several contacts within the same week. Coordinators entered the date, time of day, and result code for each contact while on the telephone with interviewers. They asked for updates on addresses and changes in respondents' names. They confirmed or changed birth- dates. When notes were needed, they were entered along with the contact. Coordinators could delete or change anything they had entered into the Mac, but could not delete respondent names. Many interviewers complained about the time weekly reports took (from thirty minutes to as long as an hour, depending on how much the interviewer had worked). The increased time over the typical report time on other studies (5-15 minutes) was caused by the additional information taken and by the fact that all contacts were reported, not just those that were final for that in- terviewer. To avoid keeping interviewers on the telephone while they did the necessary paperwork, Coordinators went back to the Mac later to record names, addresses, and case numbers for the case transfers onto the Case Transfer Forms and to initiate the transfers in their machines. 2. Use of the Mac in the Office All information about interviewers and Coordinators was originally entered into the Mac database in the office by Yvonne Shands before individ- ualized Availability Memos were sent out. Each record contained name, ID number, address, telephone number, Region number, whether the interviewer was active on the study, and the Coordinator's name and ID number. Records were updated in the central office database and in the field department's inter- viewer files when new information about an interviewer was received from the field. All assignments were entered into the central office database by Yvonne Shands also. Assignments could be made by individual case, by zip codes, or by a mass transfer of all cases in one interviewer's name (especially useful when someone dropped out before training). To make assignments, the Field Administrators prepared lists of zip codes assigned to each new interviewer, and for efficiency, all assignments for interviewers scheduled for several conferences were made at one time. Although Coordinators initiated case transfers by entering transfer codes into their Macs, all case transfers were actually made by office clerical staff, who selected the new interviewer from the interviewer list in the central database and made the assignment. The case transfer process is ex- plained in detail in Chapter VI. Records of the case transfers were collected in a transaction file, copied to a floppy disk, and given to David Tucker to input into Unify on a regular basis. He also used the file to generate the setup files (information from the 1987 interview needed for each Time 2 interview) on disks for the new cases assigned to interviewers. C. Relationship of Mac Database to Unify The Mac Database was set up as a separate entity from our Unify database at ISR. Information needed to transfer cases from one interviewer to another was fed to Unify on a regular basis via a file of case transfers. Coordinators entered address changes and corrected names for respondents into their Macs, but these name and address changes also had to be keyed into Unify by office staff when a screener or completed interview was returned to the office. Case transfers were initiated by Coordinators in their Macs, but had to be actually processed in the central Mac database in the office for the trans- fer to be made. Only the office had the complete database, and therefore the capacity to select an interviewer anywhere in the country for the new assign- ment. After Allan Walters initially loaded respondent names and addresses into the central office database from the University of Wisconsin's data tape of the sample, no information was transferred from Unify to the central Mac data- base. Contrary to our usual procedures, the name and the ID number of the interviewer assigned to a case could not be changed at check-in. When mistakes were made in the Mac assignments, cases remained assigned to the wrong interviewer in Unify until they were changed in the Mac and exported (given) to Unify. D. Transmitting Information To and From the Office Allan Walters handled both the uploads (information taken from the Co- ordinators' machines) and downloads (information transferred to the Coordi- nators' machines) during the project. Coordinators turned on their modems and computers at the appointed time, but they played no active part in the trans- mission of information. Early in the study, uploads and downloads took place on an "as needed" basis, with machines uploaded more frequently for those Coordinators who had larger numbers of interviewers working. Information uploaded from the Coordinators' Macs included the records of all contacts with respondents since the last time the Mac was uploaded and the records of non-final cases that needed to be transferred to the office or to another interviewer. At first the downloads consisted mostly of adding new assignments to the Coordinators databases, but after about May 1, 1993, the primary downloads were case transfers. The schedule changed from uploads about every six weeks, to every two weeks, then finally once a week at the end of the study. Coordinators were taught to make a backup copy of their data on a floppy disk each week after taking reports, as a precaution against loss of in- formation due to power or machine failure. However, the size of the files quickly became too large to fit onto a single disk. Coordinators were not taught how to compress the file for storage on a floppy disk, and subsequent backups were done by Allan Walters through his remote connection. E. Problems Using the Database Although we thought we had devoted adequate time at the interviewer training conferences to giving reports, Coordinators still had to retrain most interviewers individually to give a report in the right order and to include all relevant information. The Macs are large and heavy enough to present a problem for Coordinators who were traveling. As a result, many of the Coordinators simply left the machines at home when they traveled, and took reports on paper or not at all. (Some who did this had worked for other companies that gave them laptops for report- taking.) When there was an extended period of time between downloads, interviewers often received cases and completed interviews before the cases were actually loaded into the Coordinator's computer. When that happened, all Coordinators were able to do was to take a report on paper, and transfer the information to the Mac whenever the case was downloaded. As the study progressed, we added additional field result codes that were used by the office and Coordinators: 98 to indicate final refusals and 93 for traced out cases. Although Coordinators were able to enter these codes into their own MACs, this information did not transmit to the central office database. (These codes had not been included in the original list of transfer codes.) F. Usefulness of the Coordinator Database Most of the Coordinators enjoyed working with the Mac and grew more proficient in its use as the study continued. They liked having more detailed information about the sample than was usually available at their fingertips. One of the most helpful features of the database allowed Coordinators to look at the times and days interviewers contacted respondents, summarized by respondent. This knowledge allowed them to alert interviewers to problems in the ways they were using their time. On most studies, Coordinators must wait until they receive the screening form back from the interviewer to see a re- cord of the days and times the respondent has been contacted. With the Macs, the information was instantly available, and especially useful while Co- ordinators were speaking to interviewers on the telephone. Coordinators were able to create their own instant on-screen reports with the MAC, sorting information by result code, by interviewer, or by region. They looked at individual interviewer's reports to see the status of their assignments as a whole. IV. INTERVIEWER STAFFING FOR NSFH-II Over 400 interviewers were needed for the large NSFH-II sample. Our usual sample points (PSUs) required 3-6 interviewers each, and there were enough respondents in other locations to warrant staffing interviewers in some places where we do not usually staff. Because the interview was conducted in both Spanish and English, bilingual interviewers were needed in some locations. Twenty five bilingual interviewers worked on the study. A. Initial Staffing Efforts Staffing for NSFH-II began in January 1992 with a thorough search of ISR's Rolodex files for interviewers who had worked successfully for us in the past. We paid particular attention to interviewers who had performed well on two recently-completed studies--the New Beneficiary Follow-Up Survey and Current Health Issues. We looked for ISR's best interviewers. Although we considered our per- ception of an interviewer's ability to learn CAPI, successful completion of previous work for us took a higher priority. We felt that interviewers who had served us well in the past deserved a chance at the new technology, even when we had some qualms about their ability to handle computers. To determine how many interviewers to hire in an area, we used a zip code printout of the sample that reported the number of respondents in each zip code. We sent out an initial mailing of Interviewer Availability Memos to approximately 350 interviewers in March 1992. We entered the names and addresses of those interviewers into the Mac database, and used a mail/merge feature to address each cover letter individually. So that all selected interviewers would know that they had been chosen, we sent Availability Memos to those who would not be trained until 1993, as well as those scheduled to be trained in fall 1992. Although the cover letter said that there was a possibility that some selected interviewers would not be used because of changes in the distribution of the sample, we expected to use those interviewers to whom we sent the Availability Memo. Some of the interviewers scheduled for June 1993 thought that they were supposed to be trained in June 1992 because of the long advance notice. B. Supplementary Staffing by Conference Once the initial mailing was sent, we concentrated our supplementary staffing efforts by conferences, so that we always put most of our effort into staffing the two or three conferences that were coming up next. We could not overstaff the conferences, expecting some dropouts, because we were work- ing with a limit to the number that could be trained at one time. Dropouts continued throughout the training period, so that it was not possible to keep a conference staffed ahead of time. We replaced interviewers as fast as they dropped out up until the time of each conference. Staffing was done by Dolores Williamson and Elaine Trull, with each one staffing the section of the country she was supervising. In looking for new interviewers outside of ISR, we put high priority on CAPI experience and on recommendations from trusted sources. We asked about any experience using computers, if applicants had no CAPI experience. We also asked whether or not they could type. Our search for new interviewers began with the new application file in the office. Then we called supervisors for other companies, asked some of our interviewers who supervise for other companies, and contacted people suggested by these sources. In some instances, we contacted college placement offices and local Employment Security Commissions. For some of the applicants, we had written evaluations on file, sent in response to our routine request to references listed on the application. However, some applicants gave the name of an employer out of the research field, even though they had interviewing experience. For these, we sometimes called our contacts with other research companies to locate a supervisor who had worked with the interviewer. To obtain references when time was short, we asked prospective interviewers for the names of their supervisors at other companies and called those supervisors for references. When we asked for a show of hands at each conference, six to eight interviewers usually indicated they had had CAPI experience. Of the 414 interviewers who were trained for the study, 264, or 64%, had previously worked for ISR. C. Interviewer Attrition In this first CAPI project, we had a higher than usual number of inter- viewers who performed inadequately at training. Each conference had 6 to 10 people out of 22 to 30 trainees who needed extra help. Although our extra sessions in the evenings helped most interviewers to improve enough to be allowed to work on the project, we had to release one or two interviewers without assignments from most of the later conferences. A total of 414 field interviewers were trained to work on the CAPI study. We released 10 interviewers after training without giving them assignments, and 16 dropped out without completing any actual cases after training. An additional 27 interviewers did only token work (6 or fewer interviews) on the project. TABLE 2: NSFH-II FIELD INTERVIEWER STAFFING BY TRAINING CONFERENCE Conf. Date City Inter- Given no Dropped 6 int # viewers Assign- Out or less Trained Trained -------------------------------------------------------------------------- 1 July Philadelphia 23 1 1 4** Pilot 1992 *22 -------------------------------------------------------------------------- 2 Sept. Philadelphia 25 1 1 1 1992 *14 -------------------------------------------------------------------------- 3 Sept. Philadelphia 30 0 2 2 1992 *19 -------------------------------------------------------------------------- 4 Oct. Los Angeles 27 0 1 4 1992 *12 --------------------------------------------------------------------------- 5 Oct. Chicago 24 0 0 0 1992 *22 --------------------------------------------------------------------------- 6 Nov. New 31 0 0 1 1992 Orleans *16 ----------------------------------------------------------------------- 7 Nov. Detroit 26 0 0 0 1992 *15 ------------------------------------------------------------------------ 8 Jan. Dallas 25 0 0 1 1993 *17 ------------------------------------------------------------------------- 9 Jan. Atlanta 26 0 1 1 1993 *18 ------------------------------------------------------------------------- 10 Feb. White Plains 24 1 1 1 1993 NY *18 ------------------------------------------------------------------------- 11 July Philadelphia 30 2 1 0 1993 *17 -------------------------------------------------------------------------- 12 Feb. San 23 1 2 1 1993 Francisco *14 -------------------------------------------------------------------------- 13 Apr. St. Louis 30 2 2 3 1993 *24 -------------------------------------------------------------------------- 14 June Chicago 29 1 3 0 1993 *16 --------------------------------------------------------------------------- 15 June Atlanta 30 1 0 1 1993 *17 --------------------------------------------------------------------------- 16 Jan. Philadelphia 14 0 1 7 1994 *5 --------------------------------------------------------------------------- Totals: 417 10 16 27 *266 --------------------------------------------------------------------------- * Number of interviewers with ISR experience at each training conference ** All four were Coordinators, who for various reasons could not complete the required five interviews D. Staffing the Cleanup Conference We planned a final training conference for January 1994 to provide additional help for the most difficult end-of-study cleanup. In addition to the criteria considered previously, we looked for field staff with experience as cleanup interviewers. They had to be available to travel extensively until May 1994, and to be willing to go into the largest cities and to work in undesirable neighborhoods. To locate these interviewers, we combed the new application file, looking especially for experienced people. Since these interviewers were hired to travel, they did not have to live within a PSU, and we were able to consider applicants from locations that we usually ignored. We described the kind of interviewers we were looking for to our personal contacts in other organiza- tions and asked for their recommendations. At that stage, we spoke to the promising candidates by telephone to establish interest in the study. We tried to prepare them for the extreme difficulty of the remaining work. They were told that the work consisted of refusals to convert and tracing problems. They came to the training confer- ence knowing that they would be working in either Philadelphia or New York City immediately after the training, and that we expected to take most of them on a similar trip to Los Angeles in a few weeks. We repeatedly said, "It's difficult work, it won't be easy," but from later results, it's clear that many of the interviewers trained at Conference 16 had not experienced such difficult cleanup previously. Fourteen additional interviewers were trained at the conference. A new Availability Memo was adapted from the original to specify the requirements for those interviewers. V. INTERVIEWER TRAINING PROCEDURES Although interviewer training was carefully planned for this first CAPI project, we found many adjustments and changes that needed to be made. Mod- ifications continued through several conferences in fall 1992, until we felt that we had created an effective training agenda and process. After that point, changes were minor to preserve consistency among trainings. A. Training Schedule Field interviewing on NSFH-II was conducted in three waves, to correspond as much as possible with the pattern of interviewing in the 1987-88 study. If most of the interviews in an area were completed early in the 1987 study, then we scheduled that PSU for training in the first wave. A pilot training and 14 regular training conferences were scheduled. We decided to add one additional training in January 1994 to enhance our inter- viewing staff for cleanup. No interviewers were field trained on this pro- ject. Three conferences were moved from their originally scheduled times. We had hoped to have the second conference in August 1992, then the third con- ference in September 1992, so that the project would begin gradually. We needed time to analyze the results of the pilot study, so the August confer- ence became one of two in September. Also, we had scheduled three conferences for February 1993, then realized that we would not have enough laptops re- turned from interviewers who had finished assignments to supply a third con- ference in February, so one of the February conferences was moved to July. The one conference scheduled for March 1993 was moved to April 1993 for the same reason. TABLE 3: NSFH-II FIELD INTERVIEWER TRAINING SCHEDULE Conf. # Date City Trainers -------------------------------------------------------------------- 1 July 22-25, Philadelphia Allan Walters (Pilot) 1992 Marcia Murphy Elaine Trull Christine Schuler -------------------------------------------------------------------- 2 Sept. 10-12, Philadelphia Allan Walters 1992 Marcia Murphy Elaine Trull Tony DeFusco -------------------------------------------------------------------- 3 Sept. 24-26, Philadelphia Allan Walters 1992 Marcia Murphy Elaine Trull Lisa Famularo Tony DeFusco -------------------------------------------------------------------- 4 Oct. 8-10, Los Angeles Marcia Murphy 1992 Lisa Famularo Marlene Blomberg Marilyn Gunn -------------------------------------------------------------------- 5 Oct. 22-24, Chicago Lisa Famularo 1992 Linda DeWild Marlene Blomberg Sandy Lewis -------------------------------------------------------------------- 6 Nov. 5-7, New Orleans Marcia Murphy 1992 Elaine Trull Lisa Famularo Barbara Sukman -------------------------------------------------------------------- 7 Nov. 19-21, Detroit Elaine Trull 1992 Lisa Famularo Linda DeWild Marlene Blomberg -------------------------------------------------------------------- 8 Jan. 7-9, Dallas Elaine Trull 1993 Lisa Famularo Linda DeWild Rosa Dwyer -------------------------------------------------------------------- 9 Jan. 21-23, Atlanta Elaine Trull 1993 Lisa Famularo Sandy Lewis Christine Schuler -------------------------------------------------------------------- 10 Feb. 4-6, White Plains, Lisa Famularo 1993 NY Tony DeFusco Linda DeWild Marlene Blomberg -------------------------------------------------------------------- 11* July 15-17, Philadelphia Elaine Trull 1993 Lisa Famularo Marlene Blomberg Sandy Lewis -------------------------------------------------------------------- 12 Feb. 18-20, San Francisco Christine Schuler 1993 Lisa Famularo Linda Dewild -------------------------------------------------------------------- 13** April 15-17, St. Louis Elaine Trull 1993 Lisa Famularo Sandy Lewis Marlene Blomberg -------------------------------------------------------------------- 14 June 12-14, Chicago Christine Schuler 1993 Lisa Famularo Linda DeWild Marlene Blomberg -------------------------------------------------------------------- 15 June 25-27, Atlanta Elaine Trull 1993 Christine Schuler Linda DeWild Sandy Lewis Barbara Sukman -------------------------------------------------------------------- 16 Jan. 27-29 Philadelphia Marcia Murphy 1994 Elaine Trull Lisa Famularo -------------------------------------------------------------------- * Conference 11 was moved from February to July, 1993 ** Conference 13 was originally scheduled for March 1993 B. Pilot Training To test our training procedures, the questionnaire, and field procedures, we started the field work gradually. Before we entered the intensive fall 1992 training schedule, we needed time to correct any problems uncovered from the first training. We used the term, "pilot training", although in actuality we simply began cautiously. We selected one interviewer from each of the study's eleven regions to attend the pilot training. All were dependable ISR interviewers, but not necessarily those who would learn the quickest. This was to be a typical training, from the standpoint of interviewer ability. For the pilot study, 100 cases for analysis were to be completed rapidly. We instructed interviewers and Coordinators to complete five cases as soon as possible, then stop work until they received further instructions. Problems in the field were reported to us quickly. Interviewers used overnight mailers we had provided and sent Problem Report Forms to detail any difficulties in their first interviews. We were pleased to find only minor problems in those first cases. However, our projected two week work stoppage for review of the pilot cases stretched to about five weeks because of delays in printing Self Enumerated booklets. Although the delay frustrated inter- viewers and Coordinators, it allowed office staff to straighten out some problems with the questionnaire. C. Interviewer Training Each three day training was held on Thursday, Friday, and Saturday, with Wednesday night allotted to the preconference meeting of new interviewers. Sessions each day ran from 8:30 a.m. to 5:00 p.m. Group lunches were served each day to cut down on time away from the conference. Interviewers who flew to the conference returned home on Sunday mornings. Air fares were much cheaper with a Saturday night stayover, and we did not have the disruption of people leaving the conference early to catch their flights. ISR trainers for NSFH-II were Marcia Murphy, Allan Walters, Elaine Trull, Lisa Famularo, Christine Schuler, and Tony DeFusco. In addition, Coordinators Linda DeWild, Marlene Blomberg, and Sandy Lewis received extra instruction in early October 1992 and assisted with training at a number of conferences. Interviewers were sent their Home Study packages five weeks before the training conferences. They were to read the Interviewer's Manual carefully and complete a multiple choice quiz. Most interviewers did well on the Home Study Quiz, which was intended to reinforce important points about field procedures. Quizzes were mailed back to the office before the conference, and returned to interviewers and discussed during the training. Along with the Quiz, interviewers returned a short questionnaire about their lives five years previously. The information was used to create a case file for them which was loaded into their computers, so that they could comp- lete a self-interview at the training conference. Since the self-interview was about their own lives, interviewers could see quickly when they had made a mistake, because inappropriate questions would appear on the screen. Conferences began with a Wednesday night session for interviewers new to ISR, whether they were experienced interviewers or not. The purpose of the two hour meeting was to present basic information about ISR's procedures that would not be covered in any detail in the training sessions to follow. During this session, interviewers were introduced to ISR, encouraged to ask questions about the project, and learned about questionnaire conventions, contacting respondents, non-directive probing, avoiding bias, and avoiding refusals. The pre-conference training began with Conference 3 because the trainees from the earlier conferences were primarily ISR experienced interviewers. Day 1 began with a general explanation of the CAPI program, followed by a tutorial to give interviewers practice using the commands. After the morn- ing break, we explained the Screening Form, emphasizing the three types of CAPI respondents. Then by quickly going through Mock Interview 1, interviewers gained an overview of the instrument and a higher comfort level with the lap- top computer. The second mock interview was designed to be the primary teaching demon- stration and took most of the afternoon of Day 1. An introduction to data transmission and a demonstration of transmitting interviews completed the day. Both Day 2 and Day 3 began with question and answer periods. Most of the morning of Day 2 was spent teaching trainees to move around in the question- naire, with Mock Interview 3 used for demonstration and practice. One of the University of Wisconsin principal investigators presented the objectives of the project and an overview of the structure of the questionnaire in the afternoon. Field procedures were discussed, and the day ended with inter- viewers doing self-interviews on their own in the conference room. By Day 3, most of the interviewers showed increased confidence and skill in handling the instrument. Two more mock interviews were used to give the interviewers more experience and to demonstrate the new spouse and 1987 spouse versions of the questionnaire. In presentations about field procedures, we stressed tracing skills and reports to the Coordinators, and explained payment procedures. In the final presentation, we demonstrated the other computer programs listed in the Computer Guide that would be used in the field occas- ionally. Common interviewer errors were demonstrated, and interviewers were shown how they were to be corrected. Interviewers were told that they would be sent updated versions of the Questionnaire and Communications disk when needed, and they were taught how to load the disk at training. Each day of the training, we made different seating assignments, using interviewer packets, self-interview rosters, and quizzes to mark each place. We found the procedure extremely helpful in avoiding too much interviewer socializing during training and in placing those having problems where train- ers could assist them more easily. It was necessary to have the training room locked by hotel personnel during lunch each day to protect the computers from theft. All interviewers were required to satisfactorily complete two practice mock interviews with their Coordinator immediately after the training confer- ence. Except in unusual circumstances, these practice interviews were comp- leted within one week of training. Most interviewers did well with the prac- tice mock interviews, and the exercise accomplished our purpose of helping them to gain confidence and experience before they tackled actual interviews in the field. Most of these interviews took place by telephone after inter- viewers returned home, but some Coordinators and interviewers arranged prac- tice time at the hotels on Saturday nights or Sunday mornings after training. To standardize the procedure, Coordinators acted as respondents, using scripts we provided. On the last day of the training, interviewers were told that if they wished, trainers would reset the mock interviews used at the conference so that the interviewers would be able to practice them again at home. Inter- viewers who wanted the extra practice brought their laptops to be reset to one of the trainers. These interviewers also took the mock screening forms home from training with them so that they could do the extra interviews. The majority of the interviewers took advantage of this opportunity for extra practice, and usually practiced with the mock interviews before they did the required two interviews with their Coordinator. D. Adjustments to Training Procedures 1. Evening Practice Sessions During the first three conferences held in Philadelphia, Lisa Famularo gave extra help to four or five interviewers who needed additional practice. We realized that a more formal procedure for providing extra assistance was needed, however, and began scheduling evening sessions on the first and second nights of training, as needed. Most of the conferences required at least one evening session. We found that if we did not have an extra session the first night, the weaker interviewers performed poorly the second day. Interviewer reaction to the extra sessions was mixed. Several insisted they did not need the practice, although the trainers had seen very clearly that they did. There seemed to be a psychological factor involved that made it difficult for seasoned interviewers to handle being beginners again for CAPI training. The extra practice sessions, although tiring for both trainees and trainers, accomplished the desired purpose of giving more individualized attention to those who needed it. Trainers had more time to discover in- dividual problems and thus, better evaluate interviewers' CAPI performance. 2. Adjustments to the Agenda We continued to adjust the conference agenda until we had the timing right. During the process we added a more thorough review of the content of the instruments by either Jim Sweet or Larry Bumpass, and a detailed demon- stration and practice exercise for Mock 3. The Day 1 schedule was too crowded, so we worked part of that day's agenda into the following two days. After several months, we realized that interviewers needed actual exper- ience making the connections for the modem. At subsequent conferences, at breaks and before and after training, we supervised four or five interviewers at a time while they practiced making the connections. To reinforce solutions to some of the problems interviewers were having in the field, we added three short quizzes to the training. After explaining that these exercises were to help them learn essential skills for successful CAPI interviewing, the trainer asked questions and interviewers wrote answers on tablets. The trainer reviewed the answers and their explanations with the group immediately. We found this to be a very effective way of reinforcing such things as the correct way to start an interview or to change answers. 3. Development of the Mock 3 Training Module The design of the roster section of the questionnaire posed problems in the execution of the questionnaire as well as in the training. In the roster, interviewers could not go backward to change answers to questions without risking damage to the answers already entered. In the first few trainings, we spent an inordinate amount of time trying to straighten out problems caused by interviewers backing up in the roster. We developed Mock 3 to teach interviewers an acceptable way to change answers within the roster, and to identify points in the questionnaire where they could jump back to change common errors that caused problems later in the interview. We taught them to change four common errors, demonstrating, then having them correct their own machines, one error at a time. (The version of this mock loaded in their laptops had a partial interview, with errors ready for them to correct.) E. Problems Unique to CAPI Training The following problems occurred often during training and are unique to CAPI training: * Interviewers new to computers would put in an answer and press enter, then if the next question didn't come up as quickly as they thought it should, they would reenter the answer and press enter again. In effect, they answered two questions without asking the second, and had no idea that they were doing so. * If interviewers couldn't type or hadn't typed recently, they had to hunt for the proper keys on the keyboard, which would often cause them to fall behind the group. * Interviewers would enter an answer hurriedly, in an attempt to catch up with the group, even if they were not sure of the answer that had been given by the mock respondent. This would sometimes result in the computer bringing up a series of entirely inappropriate questions, and a trainer would have to go back and find and correct the mistake. * New trainees would sometimes hold down a key too long, causing the key to repeat, so that 1 appeared as 11, for example. * We had trouble training interviewers to look at the laptop screen to see what they had done wrong. They tended to immediately try something else, or repeat the same mistake, without looking closely at the screen to see why the computer was rejecting their entry. * In every conference group, there was a mix of abilities caused by such variables as CAPI experience, computer experience, knowledge of typing, experience as an interviewer, and comfort level with computers. This un- evenness of abilities often caused some training problems. The computer know- ledgeable were often bored with the last mocks and the non-typists often felt they were struggling to catch up with the group. F. Evaluation of Trainees Throughout the conference period, trainers found it difficult to evaluate the work of trainees. In particular, it was hard to determine who had enough motivation to spend extra time at home to bring their CAPI skills up to par. We could tell that nervousness was a factor in the group situation for a number of interviewers. We tended to be more lenient at the first trainings, then more stringent at the later ones. However, we usually found ourselves struggling with the final decision of whether or not to give work to one or two of the inter- viewers on the last day. Any one trainer could determine that an interviewer needed the night practice sessions, but the decision to release an interviewer from the study without an assignment was made by a consensus of the trainers. VI. FIELD PROCEDURES Several important factors guided the field design and influenced field procedures for NSFH-II: * The original interviews took place over a 14 month period. Because the study was intended to be a five year follow-up, it was desirable to follow a pattern of interviewing similar to 1987-1988. * Considering the amount of the budget that would be used for the purchase of laptop computers, it seemed desirable to buy only half as many laptops (235) as the projected number of interviewers who would work on the study (450) and to reuse laptops several times. * The study was planned as a series of three waves of interviewing, so that training was held for a number of PSUs, all interviewers for those areas were trained, and work in those areas was completed within four months. Then the study would move to another set of PSUs, and then a third. Cont- inuity would be provided by the ISR management team and by the Coordinators, who would supervise different groups of interviewers at different times. * Wave 1 trained between September and November 1992. The PSUs involved were those in which the majority of the work in the original study was completed first. * Wave 2 trained from January to March, 1993. These PSUs were finish- ed in the middle of NSFH-I. * Wave 3 involved a smaller number of trainings and interviewers. These were the areas that were the last ones to be completed on NSFH-I. * Although the pattern of interviewing would follow the 1987-1988 study, provision would be made to start the work in trouble spots like New York City, Chicago, and Los Angeles with the first wave of interviewing. * To allow flexibility in interviewing respondents who were found in an area after the four months of field work in the PSU had ended, we left one laptop computer in each PSU with a reliable interviewer who could complete additional cases as they were found. * Cleanup work was anticipated, but expected to be on a smaller scale than actually happened. Respondents who had moved were expected to be located and reassigned to another interviewer in time for the interview to be comp- leted within the four month period. Because the case transfer process got off to a slow start, some interviewers who could have done other cases had to turn in their computers before they received the transfers. * The CATI portion of the study was planned to begin about two months after the field interviewing, so that enough cases had accumulated to keep the telephone interviewers busy. The telephone interviewers were expected to keep pace with the field work, so that the parents and children were interviewed on a timely basis also, three to four weeks after the main respondent. Many NSFH-II field procedures were similar to those used on other ISR studies, but new procedures had to be determined and to be set up for aspects of the field operation that related to the CAPI technology, such as trans- mitting interviews via modems and setting up a Computer Hotline. Original field procedures are detailed here. Changes made during the cleanup phase are explained in Chapter VIII, The Cleanup Phase of the Study. A. Contacting Respondents ISR sent respondent introductory letters to all main respondents and 1987 spouse/partner respondents prior to the initial contact with a household. Those letters were generated by David Tucker, along with assignment disks, screening forms, and Records of Work whenever new assignments were made. Data Processing personnel mailed the introductory letters. Most respondents received the introductory letter four to eight weeks before they were contacted by an interviewer, but for some respondents the lag time was longer. We generated assignments, and thus introductory letters, for several conferences at a time during the second and third phases of the study. Also, if an interviewer dropped out or was not allowed to work the assigned cases, his/her respondents were not contacted until another inter- viewer was trained at a later conference. At the beginning of the study, we gave interviewers the option of making initial contacts with respondents by telephone, rather than in person, al- though we stressed the effectiveness of personal contact. Because the refusal rate from some interviewers making telephone contacts seemed high, we changed the emphasis and trained interviewers to make personal contacts unless the respondent was an isolated case far from that interviewer's home. Interviewers secured parental permission for the telephone interview with a child under 18 as an integral part of the main respondent and spouse interviews. The questionnaire gave an explanation about why we needed the child's interview, then asked for parental permission. Interviewers read the question, then handed the parent the written permission form to sign. B. Submitting Completed Work We required interviewers to mail completed Self Enumerated booklets as soon as they were finished with the interview or the following day. The Self Enumerated booklet was sent to the office with the screening form, parental permission for a child's interview, participation form, and any Problem Report Form or Current Spouse Worksheet inside the front cover. We provided postage- paid envelopes and encouraged interviewers to mail only one Self Enumerated booklet per envelope. Interviewers transmitted their completed interviews to ISR electron- ically, using a modem connected to their computers. Interviews were to be transmitted on the day they were completed or the next morning, if they re- turned home late. Interviews could be transmitted 24 hours a day, seven days a week, except for the occasional times when ISR's computer system was down. Interviewers had to type a one word command to start the transmission program, insert and remove the backup disk when prompted, and type in their login (last name and id number). Even though the process was simple, there were numerous transmission problems. A number of transmission problems were caused by interviewers making improper connections between modem, laptop, and phone line. In May 1993, David Tucker, the project programmer, revised the program to give error messages to interviewers who did such things as type in capital letters when they should have used lower case ones. To make sure that both parts of an interview--the Self Enumerated booklet mailed by the interviewer and the computer file of the interview transmitted by modem--reached the office, lists of missing parts were generated period- ically. Office clerks sent routine memos to tell interviewers Self Enumerated booklets, screening forms, or transmitted interviews were missing. On weekends, Amanda George, one of the telephone supervisors, called in- terviewers during her Saturday and Sunday shifts. She told the interviewers they had untransmitted interviews and helped them transmit, or asked them to send in their backup disk if the cases did not transmit with her help. When laptops were returned to the office as interviewers finished their work, a backup of all cases that the interviewer had ever entered was made in ISR's computer system. Then Data Processing personnel compared a list of case numbers of the interviewer's transmitted interviews to the list of cases for which a Self Enumerated booklet had been returned. Any cases that hadn't transmitted were entered into the Unify database by data processing personnel. At the end of the study, there were only 14 cases for which we were not able to retrieve data. Although an interview without a Self Enumerated booklet was not complete, it was still a usable interview for the study. There were 35 interviews with missing Self Enumerated booklets, either because the respondent refused to complete the booklet or because the booklet never reached the office. In the spring of 1993 a problem developed with backup disks that couldn't be read by laptops, causing interviewers to get a "disk error" message. The problem was solved by instructing interviewers to copy a small file from the hard drive to the floppy disk, or by sending them a new backup disk without the problem. We suspected that the machine that formatted disks in quantity became slightly out of adjustment to cause the problem, which affected most of the interviewers trained at Conference 13 and some thereafter. Once one file was copied to the diskette, the floppy disk would accept the backup of cases when the interviewer tried to transmit. C. Refusal Conversion Efforts To avoid refusals, interviewers were provided with a great deal of in- formation about the study during training, so that they would be well-prepared to discuss the study with respondents. At trainings, we emphasized the im- portance of personal contacts, rather than phone calls. We expected interviewers to visit the respondent in person if they re- ceived a telephone refusal. When interviewers had received one firm refusal in person, whether on their first or tenth visit, we instructed them to return the case to the office to be reassigned for refusal conversion. However, at least one Coordinator did not adhere to this policy, requiring her inter- viewers to go back three or four times, resulting in repeated refusals and sabotaging the chance of another interviewer to convert the refusals. We be- lieve that another interviewer has the best chance of converting the refusal when the respondent has firmly refused the first interviewer. Respondent refusal letters were sent from the office as part of the pro- cess of reassigning refusal cases. Once the new interviewer received his/her disk and screening form, he/she was free to visit the respondent in person to attempt the refusal conversion. We taught interviewers that all cases that did not result in interviews should be discussed with the Coordinator before the screening forms were sent back to the office. D. Tracing the NSFH-II Samples University of Wisconsin study staff expected ISR to be able to locate 95% of the NSFH sample. However, tracing respondents proved to be the most diff- icult problem of the study. Despite the difficulty, our final results showed only 1285 respondents, or 5.8% not located, not quite one percent more than was expected. NSFH researchers had expected to secure up-to-date addresses for approx- imately half of the fifty percent of respondents they estimated had moved. They had received a large number of letters and messages from respondents giv- ing updated address information, and reference information from 1987 was available. The number of post office returns from their annual mailings to respondents was small, so the other respondents were assumed to be at the last given address. Of the main respondents, 90% gave the name and address of at least one reference in 1987-88, and 71 percent gave two or three references at that time. As we used the reference information, however, we learned that the con- tact information was often missing critical pieces, such as the actual street address within a city or the first name of the reference. ISR's usual policy is to do extensive institutional tracing before a sample is sent to the field. Instead, we submitted the best address that the University of Wisconsin had for our respondents to MetroMail, an address up- dating service, for confirmation. Based on the positive experience that one of the Wisconsin researchers had with another sample he had sent to MetroMail, we hoped to provide the most up-to-date addresses possible to the field. Names and last addresses of approximately 2600 respondents known to be tracing problems were processed by Metromail in December 1992. Tracing problems were compounded by discrepancies in the names of a large number of respondents. In some cases maiden or middle names appeared as last names. When we had respondents and spouses with different names and different addresses, it was hard to determine in the office and in the field if the couple had separated, the woman chose to use her maiden name, or there was some other problem. A tracing report was developed by Marcia Murphy, with input from the other members of the project team. From the 1987 interview, the items of in- formation that would be most useful in tracing respondents were selected. In addition to the names, addresses, and phone numbers of the people given by the main respondent as references in 1987, the report included employment and edu- cation information, marital status, and whether or not the respondent had been in the military or received welfare benefits. In an effort to avoid repeated contacts with references, the tracing in- formation was not released to interviewers at the beginning of the study. In- terviewers were expected to complete the four tracing steps outlined in train- ing, and, if unsuccessful, to submit the case to the office for contact with references and further tracing. The tracing steps required of interviewers were simple ones, expected to turn up a good number of respondents. Interviewers were taught to call dir- ectory assistance for the respondent and others of the same last name, to telephone others with the same last name, to contact the post office for an address update, and to speak to at least two people in the respondent's neigh- borhood. Many of our best interviewers did more than the required steps, and used ingenuity and perseverance to find more respondents. However, a number of the interviewers seemed to assume that if respondents had moved, they would never find them, and put forth little effort to do even the basic steps correctly. In the fall of 1992, only Yvonne Shands worked on tracing problems, tele- phoning references and checking directory assistance when she had time avail- able from her other duties. When the report came back from MetroMail, add- resses different from those listed by the University of Wisconsin as the best address were sent to the appropriate interviewers who were working. New MetroMail addresses for respondents in areas where interviewers had not yet been trained were input into Unify so that they would appear as the current address when those cases were assigned. Fourteen phone tracers were trained by Yvonne Shands to check directory assistance and call references in March, 1993. They worked from the tracing reports of all cases that had been returned to the office as Code 36s, all leads exhausted in the field. Two additional tracers were trained in July 1993. When new addresses for respondents were found in the office, the approp- riate Field Administrator or Coordinator was asked to designate the new inter- viewer, and the case was transferred as usual. E. Validation of Interviews A validation letter was sent to each respondent (R,S, and N) along with the incentive check for participating in the study. Marian Anderson, a super- visor in the Data Reduction Department, monitored the clerks who performed the task. The letter asked the respondent to answer nine questions about the in- terview, including whether they were interviewed in person, if the interviewer used a computer, and if they filled out answers for themselves in a Self Enumerated booklet. Respondents were also asked for the length of the inter- view and for their comments about the interview or the interviewer. In the beginning, Dolores Williamson reviewed the forms and recontacted respondents when necessary. After she left, the forms were reviewed by a validation clerk, with copies of forms indicating potential problems given to Elaine Trull for review. Carolyn Rahe began supervising the validation pro- cedures in January 1994. A validation screen was set up in Unify in August 1993, and the answers to the validation questions entered for comparison with actual questionnaire answers. To evaluate the validation forms, the clerk considered as potential prob- lems all cases in which the respondent reported that he/she was not inter- viewed in person or not interviewed at all, that a computer was not used, or that the time of the interview was less than 40 minutes. Validation clerks were also instructed to show the Field Administrator any pertinent comments. If the respondent reported that he/she was interviewed by telephone, the clerk compared the interviewer ID number with the list of telephone inter- viewers. Cases not completed by the telephone room were followed up by a phone call to the respondent. If the respondent said the interview took 40 minutes or less, the clerk compared the time recorded in the interview by the internal timers to the respondent's reported time. A program was designed to check for discrepancies in the respondent's and interviewer's reports about whether the respondent filled out Self Enumerated booklets on his/her own. We followed up by telephoning respondents, which resulted in clearing up most discrepancies. Marcia Murphy wrote a program that calculated the average length of each type of interview. The program eliminated times of less than 10 minutes to avoid counting the interviews that were reentered for corrections. The ave- rage time for the main respondent interview was 89 minutes; for current spouses, 81 minutes; and for ex-spouses, 86 minutes. F. Case Transfers The simple steps involved in the case transfer process belie the comp- lexity of the process. First, interviewers reported respondents who had moved beyond 75 miles of their homes, had refused, or had a language barrier to their Coordinator, who confirmed that the cases should be sent to the office for transfer. Coordinators completed a three part NCR form that specified the case number, respondent's original name and address, respondent's new name and/or address and telephone number, the transfer code and reason for the transfer. They identified the original interviewer and specified the inter- viewer to whom the case should be assigned, if the respondent was within their Region. Interviewers mailed the Screening Forms directly to the office; Co- ordinators mailed the Case Transfer Forms to the office and entered a code in their Mac databases to show that the case needed to be transferred. Upon receipt in the office, the Case Transfer Forms and Screening Forms were filed in folders by Coordinator, and clerks regularly checked for matches. Theoretically, both pieces of paper should have reached the office about the same time, been matched up, and processed as a transfer. In act- uality, some Coordinators were slow about sending in the Case Transfer Forms or made errors in the forms and some interviewers delayed sending the Screen- ing Forms. Such a process needs careful monitoring by a Field Administrator to make sure that cases continue to move through the case transfer cycle promptly. Case transfers were logged into the central office database. At the next export, the name of the assigned interviewer was changed in the Unify database and a disk containing the case setup file and Record of Work were generated for the new interviewer. In the beginning of the study, exports were sched- uled as needed, so the time between the report of a case needing transfer and its receipt by a new interviewer was often two months or more. To further compound the problem, sometimes there were errors that kept transfers from going through, so that cases did not transfer until the next export. The first case transfers were made in December 1992. G. Other Field Problems Some interviewers discovered that they were missing a 1987 spouse screen- ing form only when they met the spouse who was ready to be interviewed. They substituted the new spouse screening form and used the new spouse case that was available in the computer for that main respondent. In doing so, the in- terviewer had to ask some inappropriate questions of the 1987 spouse respond- ent and probably had some difficulty in going through the interview. There were approximately 35 of these cases. In the office, the 1987 spouse cases done incorrectly as new spouses caused recordkeeping problems. Until each situation was recognized and cor- rectly coded in Unify, our records made it appear that the 1987 spouse still needed to be interviewed. Although the front of the Screening/Call Report Form doesn't indicate that institutionalized respondents were to be interviewed, the instructions we gave at training were slightly different. Interviewers were instructed to try to find out if a nursing home resident was capable of doing the interview; if so, they conducted the interview. If the nursing home was too far away from the interviewer's home, the case was transferred to the telephone room. Since interviewers were supposed to explain cases to their Coordinators before returning the screeners, we also learned of some respondents in prisons that we were able to interview. A few prison or jail inmates were inter- viewed, on all parts of the study, by both CAPI and CATI interviewers. The fact that the interviewers were using computers caused no additional problems in completing interviews in institutions; the problems stemmed from the usual time and paperwork necessary to secure permission to interview prison inmates. H. Payment of Incentives Main respondents and spouses were each paid $20 for completing the NSFH- II interview. Interviewers turned in a participation form signed by the re- spondent along with the Self Enumerated booklet and other materials for each case. Respondents were sent their checks along with the validation letter. Marian Anderson, supervisor in the Data Reduction Department, handled re- spondents' phone calls or complaints about checks not received. Usually a check had been delayed for one of four causes: the Screening/Call Report Form had not been received from the interviewer; the mailing address was incorrect; information was keyed into our system incorrectly, or the respondent hadn't cashed the check promptly and it had expired. We instructed interviewers to tell respondents that checks would be re- ceived in four to six weeks, to avoid hard feelings if the checks did not come quickly. Interviewers were also told that they could not pay incentives in cash. When a few respondents refused the interview in the early months of the study, saying they were never paid their incentive for the first interview, the decision was made not to try to rectify any inaccuracies now for the 1987 interview. A memo was sent to interviewers on December 10, 1992, instructing them to tell respondents that we no longer had records of payments on the first wave, and that we were not able to go back and pay them for time 1. However, at the end of the study, six to ten respondents with this complaint were paid an incentive for the original study to secure their participation in Wave 2. I. Managing the Equipment Inventory Interviewers were required to fill out Equipment Release Forms, listing the serial number of their computer, at the training conferences. The modems were shipped from the office to the interviewers, along with their initial supplies, and the modem serial numbers were recorded in the office before shipment. When laptops were lost or stolen, as five were during the project, inter- viewers were required to make a police report and to obtain a copy of the re- port. The police reports were forwarded to the University of Wisconsin so that the insurance claims could be filed by them. One laptop was recovered by the Atlanta police, who called the University of Wisconsin in Milwaukee be- cause of the University of Wisconsin label on the bottom of the machine. J. Repair of Machines We did not anticipate the large number of repairs that would be needed to keep the laptops in working order. Shipping was one factor that caused a number of breakages. The most frequent repairs needed were to repair the floppy disk drive and to replace broken screens. We also had some problems with the power cord connections to the computer and some hard drive failures. When an interviewer reported a problem to the Coordinator or to the Computer Hotline, we tried to troubleshoot with simple solutions the inter- viewer could try. For most repairs, David Tucker also spoke to the inter- viewer by phone to make sure there was nothing else to be done before the machine was sent back to the office. When broken laptops came in, David checked them and sent the ones needing repair on to Digital. Machines were usually returned by Digital within one week, and the repairs were covered under warranty until August 1993. After that, we no longer repaired the broken machines and had accumulated about 35 broken machines by the study's end. At the beginning of the project, we ordered approximately fifty extra batteries to be sent to interviewers as replacements as needed. Because in- terviewers seldom conducted interviews by battery, only five or six inter- viewers needed battery replacements. K. Using the Hotline Three 800 lines were set up for study use. One served as the toll-free number for transmitting interviews, one was used in Room 504 to receive routine calls, such as inquiries about payment records and assignments, and the third was the Computer Hotline. The Hotline phone remained in Room 504 during the daytime, then was moved to the telephone supervisor's desk for evenings and weekends. The Computer Hotline was staffed for extensive hours during the busiest part of the study, and for fewer hours in the beginning and very end of the survey. Marcia Murphy and David Tucker trained the telephone supervisors to guide interviewers through problems with the roster, to answer other questions that might come up during an interview, and to handle transmission problems. Three of the supervisors also worked as clerks on the project during the days and answered the Hotline in Room 504. The Hotline was very effective for a number of interviewers who called from respondents' homes with problems and needed immediate help. L. Payment for Interviews Interviewers submitted Interviewer Payment Records weekly to receive compensation for their work. The original plan to automate the payment records, so that payment was generated when interviews were transmitted, was thwarted by the difficulty in getting university approval in time to be ben- eficial to the study. As usual, we checked records in Unify to make sure that the interview had been received before we made payment. Payment was made when either the Self Enumerated booklet or the transmitted interview was received, and was not held up because of transmission problems or mail delays. Clerical staff handled the pre-processing of payment records, asking Field Administrators about any questionable charges. The complete payment schedule is included in the Interviewer Availability Memo, which is part of the Appendix. VII. CATI PROCEDURES All main respondent, 1987-88 spouse/partner and new spouse/partner inter- views that were completed in-house, and all parent, older child and younger child interviews, were administered by Computer Assisted Telephone Inter- viewing (CATI). Paper copies of the questionnaires were not available. The average length of a parent interview was 59 minutes. Older child interviews also averaged 59 minutes. Interviews with younger children were shorter; these interviews averaged 34 minutes. A. Timing of CATI Interviews With Parents and Children In planning for the study, we decided to train all telephone interviewers on the parent and child questionnaires first, then train the most skilled in- terviewers from the group to conduct main respondent and spouse/partner inter- views by telephone. We expected focal children and parents to be contacted approximately three to four weeks after the main respondent and spouse interviews were completed. However, field interviewers having problems transmitting comp- leted main respondent and spouse interviews caused delays in the release of some parent and child cases. This resulted in a greater lag time between these main respondent interviews and parent or child interviews. The first group of parent and child cases, generated from the pilot study and interviews conducted in September 1992, were contacted a few months after the main respondent and spouse interviews were completed. These cases were held until we had accumulated enough cases to keep a staff of telephone inter- viewers busy. The first group of telephone interviewers was trained to con- duct parent and child interviews in November 1992 . A second group was train- ed in February 1993. B. Timing of CATI Interviews With Main Respondents, Spouses and Partners The study was planned to allow telephone interviews of main respondent and spouse/partner cases when these respondents lived overseas or more than 75 miles away from a field interviewer. As field assignments were made across the country, Field Administrators designated zip codes that were to be assign- ed to the telephone staff because of their isolated locations. In March 1993, all of those assignments were made and approximately 600 cases were assigned to the telephone staff. The first group of telephone interviewers was trained to conduct interviews with main respondents, 1987-88 spouse/partners and new spouse/partners in March 1993. C. Overview of the CATI Study Interviewing of parent and child respondents began immediately following the initial training in November 1992. Interviewers worked under supervision in ISR's Center For Telephone Interviewing in room 505 of the University Services Building. Due to the small amount of parent and child sample avail- able to the telephone staff at that time, shifts were only held every Monday and Wednesday, from 5:00 p.m. to 10:30 p.m., and Sunday from 10:00 a.m. to 9:30 p.m. As the number of completed main respondent interviews grew, the size of the parent and child sample grew. In March 1993, enough parent and child cases had accumulated to warrant expanding the telephone interviewing shifts. New hours were Monday through Thursday, from 5:00 p.m. to 10:30 p.m., Saturday from 10:00 a.m. to 4:00 p.m., and Sunday from 10:00 a.m. to 10:30 p.m. Every interviewer was required to work at least two shifts per week. In the beginning of the study, some interviewers occasionally worked on weekday mornings or afternoons to cover daytime appointments and reach re- spondents who were difficult to reach in the evening or on weekends. As the field period continued, it was clear that many of the parent respondents were easier to reach in the morning and many younger child respondents were easier to reach in the afternoon. As a result, two interviewers, who worked well independently, were scheduled to work daytime hours on a regular basis. To increase the number of interviewers per shift in the evenings and on weekends, groups of interviewers were trained on the parent and child quest- ionnaires in April 1993 and again in May 1993. The most skilled interviewers from these two groups were trained in October 1993 on the main questionnaire. In November of 1993, we added a Friday evening shift, from 5:00 p.m. to 10:30 p.m., and began interviewing respondents who lived overseas. Until March 1994, the focus of telephone interviewing was on completing parent and child interviews. At the end of February 1994, it was decided that the telephone staff should concentrate on completing more main respondent and spouse/partner interviews. At this time, all telephone interviewers who had previously been trained on the main questionnaire were told to work only on main respondent and spouse/partner sample. The most productive interviewers were given the opportunity to work day hours. Between eight and ten inter- viewers were interested in working during the day. We began supervised day shifts, ranging from five to eight interviewers per shift, late in April. To further increase the number of interviewers working on main respondent and spouse/partner cases, more interviewers were trained. Two groups of in- terviewers were trained on the parent and child questionnaires in March 1994. The most skilled interviewers from these two groups were trained on the main questionnaire in April 1994. Diane Hansen, a representative from the Univer- sity of Wisconsin, observed the April training. In May 1994, all main respondent and spouse/partner cases in the field were transferred to the telephone staff for a final clean-up effort. The goal of the telephone staff at this time was to complete as many main respondent and spouse/partner interviews as possible by the end of June. The telephone staff was not large enough to handle all of the cases being transferred in. We lost interviewers who graduated from college or left for the summer. The size of the staff was shrinking, so more interviewers were trained. Tele- phone interviewers who had recently finished working on another ISR telephone study were sent Availability Memos. Sixteen interviewers, all with ISR tele- phone interviewing experience, were trained on the main questionnaire mid-May. On May 16, 1994, we began interviewing shifts of between 20 and 30 inter- viewers. Shifts this large required us to work in both telephone interviewing rooms 505 and 509, so we scheduled two supervisors per shift, one for each room. Newly trained interviewers worked in room 505 and experienced inter- viewers worked in room 509. Individual interviewing stations were chosen by interviewers, although supervisors often exercised their right to reassign seats. When the focus changed to completing main respondent and spouse/partner interviews, all telephone interviewers who had previously been trained on the main questionnaires were assigned only main respondent and spouse/partner sample. Only two interviewers continued to work on parent and child cases. By July 1994, the interviewing of main respondents was complete and the focus of telephone interviewing shifted back to completing parent and child interviews. At this time, we had an 85% response rate for younger child in- terviews. All younger child sample was pulled out of the telephone room. Interviewers concentrated on completing parent and older child interviews. Interviewing of parent and child respondents ended on August 31, 1994. D. CATI Staffing The telephone portion of NSFH II was managed by Lisa Famularo, with assistance from the CATI supervisors. 1. Selection of CATI Supervisors Initially, three supervisors were selected, all of whom had prior ISR CATI interviewing experience. The following criteria were considered the most important in selecting telephone supervisors for this study: * experience as a CATI interviewer * excellent CATI skills and the ability to solve CATI related problems * the patience and ability to assist interviewers having difficulties * ability to keep careful records * ability to manage a group of 10 or more interviewers * ability to accurately evaluate an interviewer and provide feedback * ability and willingness to learn how to answer calls on the NSFH-II Computer Hot Line The supervisors were trained by Lisa Famularo. The supervisor training included several mock interviews, handling CATI problems (such as locked cases and cases not in entry), guidelines on managing sample and record keep- ing procedures. The supervisors were also given Computer Hot Line training by Marcia Murphy. They learned how to assist and provide support to field interviewers having computer problems, difficulties using the software and/or problems transmitting data. The three original supervisors were also daytime clerical workers on the project, which provided them with the knowledge to answer other study-related questions. These supervisors answered daytime calls to the Computer Hotline as well. When a supervisor left the project, a new supervisor was selected from the telephone interviewing staff. There were a total of seven supervisors over the course of the study. Three of these supervisors also interviewed on shifts they were not scheduled to supervise. 2. Recruiting and Selecting CATI Interviewers Staffing efforts for the telephone portion of the study began in August 1992 with the preparation of a list of interviewers who had performed well on earlier ISR telephone studies. To this group we added persons recommended by experienced interviewers. In September 1992, we posted flyers on campus that advertised the available positions. We also posted a notice with Career Ser- vices. Both postings stated that ISR was looking for reliable individuals to work part-time in the evenings and on weekends. Due to the fact that most college students have at least some basic computer skills, computer experience was not mentioned in the postings. The focal point of the flyer was the $7.00 per hour rate of pay plus bonuses. People immediately responded to the flyers. The flyer and the posting at Career Services asked all individuals interested in the position to respond by telephone, not in person. Everyone who responded by telephone was screened by either Lisa Famularo, Yvonne Shands or Dolores Williamson. If the person sounded professional, had a clear speaking voice and good telephone manners, we set up an appointment for the person to come into the office, fill out an application and have a brief, informal interview. Screening potential inter- viewers by telephone provided us with the opportunity to allow only those with good telephone skills to fill out applications and cut down on the amount of traffic in and out of the office. After a candidate filled out an application, he/she met with Lisa Famularo for a brief in-person interview. Applicants were given a brief job description, asked about past work experience and asked to read aloud the introductory paragraph to the interview. Those candidates who were considered suitable for the position were sent an Availability Memo that described the project and outlined what would be expected of interviewers who worked on it. Those interested in the project were requested to submit an Availability Form to indicate their interest and to provide information about their desired working hours. Staffing decisions were made based on all information available about a candidate: records of previous work for ISR or other employers, application, resume, Availability Form and communication skills. Interviewers from this pool of candidates staffed the first five training conferences. All selected applicants were sent a pre-conference packet of study materials including: a Telephone Interviewer's Instruction Manual, a CATI Home Study Quiz and a cover memo, which welcomed them to the project and in- structed them to read the manual and bring the completed quiz to training. The purpose of the Home Study Quiz was to highlight important points and to encourage the trainees to read the Telephone Interviewer's Instruction Manual prior to training. The quizzes were collected on the first day of training, graded and returned to the interviewers on the second day of train- ing. No one was ever dismissed from training on the basis of their quiz score. A second round of staffing was deemed necessary after the telephone work had been underway for almost a year. We needed to replace interviewers who had left the project, increase the size of the entire interviewing staff and increase the number of interviewers working on main respondent, 1987-88 spouse/partner and new spouse/partner cases. Availability Memos were mailed to individuals who were interested but not available for earlier conferences, to experienced interviewers who had recently finished working on other ISR telephone studies and to interviewers who were currently working on the pro- ject as part of the tracing staff. From this group an additional 40 inter- viewers were selected, five of whom were bilingual in English and Spanish. Throughout the field period, weekend shifts were difficult to staff. Friday evening and Sunday evening shifts were the most difficult to fill. During the second round of staffing, we tried to recruit interviewers who wanted to work weekends. All interviewers hired during this round of staffing were required to work at least one weekend shift per week. On average, between 25 and 30 interviewers worked on the study at any given time. During most of the field period, our goal was to have at least eight interviewers per shift Monday through Thursday and at least five inter- viewers per shift Friday through Sunday. In other words, we wanted a total of at least 50 staffed stations per week. In order to figure out how many inter- viewers were needed, we assumed each interviewer would work at least two shifts per week which meant 25 interviewers were needed to meet our goal of 50 staffed stations per week. E. CATI Training Telephone interviewer training conferences were held in rooms 503 and 505 of the University Services Building. Training conferences were usually held on Saturday and Sunday from 9 a.m. to 5 p.m. Telephone supervisors Ellen Greenberg and Amanda George assisted Lisa Famularo at the trainings. CATI mock interviews used at training were created by Marcia Murphy. Other training materials were prepared by Dolores Williamson and Lisa Famularo. 1. Training on the Parent and Child Questionnaires On the first day of the conference the group assembled in a standard training classroom. The training began with a description of the background and purpose of the study. The agenda continued with a presentation on tech- niques for administering questionnaires, probing, avoiding bias, guidelines for telephone interviewing and interacting with respondents, overcoming ob- jections and avoiding refusals. The next presentation was an explanation of the Call Report Form, Contact Information Sheet and Parental Consent Form. The morning ended with a question and answer session. The group assembled in the telephone interviewing room after break. An introduction to the CATI commands and conventions was presented, then each interviewer individually completed a tutorial on CATI. The rest of the after- noon consisted of group role-playing, providing interviewers with practice using CATI to record responses to one mock interview of each type. Each mock interview began with a role-playing exercise on how to handle a certain situation. The first mock (a younger child respondent) began with the interviewers practicing how to record calls that resulted in no contact on the CRF and in CATI. The second mock (an older child respondent) began with the trainer playing the role of a suspicious household member. The inter- viewers practiced probing the household member and making mock directory assistance calls and other tracing calls. The third mock (a parent respond- ent) began with interviewers practicing how to handle a household refusal. (The parent mock was ended as a partial interview, to be continued on the sec- ond day of training.) After the role-playing at the start of each mock, the interviewers went through the remainder of each mock interview as a group. The interviewers took turns asking the questions, and all keyed in the train- er's responses to scripted interviews. On the second conference day, the group met in the training classroom. After a review of the home study quiz, the interviewers were given general instructions on study procedures such as filling out payment records, and rules and guidelines concerning breaks, schedules and calling out sick were explained. The group then moved into the telephone interviewing room. After learning how to get back into a partial interview on CATI, the group finished the partial parent interview they had begun on the first day of training. The fourth mock (a younger child respondent) began with interviewers practicing how to contact the main respondent for parental permission to interview the younger child respondent. After break, the fifth mock (an older child respondent) began with inter- viewers practicing refusal conversion techniques. The final mock (a parent respondent) began with a presentation on how to make corrections in the very sensitive "rostering section" of the questionnaire. This section was referred to as "very sensitive" because there was only one way to correctly make changes in this section. Interviewers who used a method other than the one described at training ended up with more errors than they started out with. The training ended with a question and answer session and interviewers filling out training fee payment records, signing confidentiality pledges and confirming their permanent work schedules. 2. Training on the Main Questionnaire Almost all telephone interviewers who were trained on the main respondent and spouse/partner questionnaires had previously been trained and were working on the parent and child questionnaires. The fact that the interviewers at- tending the main questionnaire training were already interviewing on the study enabled us to spend less time covering the basics -- purpose of the study, interviewing techniques, CATI tutorial, etc. The mock interviews that were used to train field interviewers were also used to train telephone inter- viewers. However, the mocks were used for different purposes and to emphasize different points. On the first day of the conference the group assembled in a standard training classroom. The training began with a brief review of the background and purpose of the study. The training continued with an explanation of the field call report form, household roster, Self Enumerated booklet (SE book- let) and other forms needed to complete an interview. After a brief question and answer session, the group moved to the tele- phone interviewing room. The mock referred to as "Mock 2" at field interview- er training was the first mock interview used at the telephone interviewer training and was used as the teaching mock. A large amount of time was spent teaching interviewers how to properly administer the Self Enumerated booklet. The questions in the Self Enumerated Booklet were not designed to be read aloud by an interviewer, so much of the training during this mock focused on reading those questions properly. After break, the group reassembled in the telephone interviewing room to go through two mock interviews. The second mock demonstrated the 1987-88 spouse/partner version of the questionnaire and the third mock demonstrated the new spouse/partner version. On the second conference day, the group met in the training classroom. The morning began with a presentation on tracing. Special emphasis was placed on techniques for tracing through references that were given by the main re- spondent during the 1987-88 interview. The next presentation was a review of a procedure, similar to one used in the parent interview, to make corrections in the very sensitive "rostering section" of the questionnaire. After this presentation, the group moved to the telephone interviewing room to practice the process described in the presentation and complete a fourth mock inter- view. After break, the group met in the training classroom for a role-playing session. Answering respondent questions and converting refusals were prac- ticed. Following the role-playing session, the group went into the telephone interviewing room and each person completed his/her own self-interview. The training ended with a question and answer session. 3. Proxy Training In September 1993, all telephone interviewers who had previously been trained on the main questionnaire met for a session on administering the proxy questionnaire. In October 1993, proxy training was added to the main quest- ionnaire training agenda. 4. CATI Interviewer Training The following table lists the training dates, the type of training, the number of experienced ISR telephone interviewers who attended, the number of interviewers new to ISR who attended, the number of people who did not show up for training and the number of interviewers who worked for less than one month after training. TABLE 4: CATI INTERVIEWER TRAINING Telephone Types of Number Trained No Worked Training Dates Interviews *ISR *New to Show Less exper. ISR Than 1 Month ----------------------------------------------------------------------------- Nov. 21-22, Parents and 4 6 1 1 1992 Children ----------------------------------------------------------------------------- Feb. 25-26, Parents and 3 6 1 2 1993 Children ----------------------------------------------------------------------------- March 18-19, Main Rs and 8 0 0 0 1993 Spouses ----------------------------------------------------------------------------- April 3-4, Parents and 0 10 0 3 1993 Children ----------------------------------------------------------------------------- May 22-23, Parents and 0 9 1 1 1993 Children ------------------------------------------------------------------------------ Oct. 16-17, Main Rs and 10 0 0 0 1993 Spouses ------------------------------------------------------------------------------ March 5-6, Parents and 1 10 0 2 1994 Children ------------------------------------------------------------------------------ March 19-20, Parents and 3 6 0 1 1994 Children ------------------------------------------------------------------------------ April 20-21, Main Rs and 10 0 0 0 1994 Spouses ------------------------------------------------------------------------------ May 14-15, Main Rs and 16 0 1 1 1994 Spouses ------------------------------------------------------------------------------ Totals * * 4 11 ---------------------------------------------------------- * The sums of these columns would double count interviewers who were trained on both the main respondent and parent/child questionnaires. See Table 5 (Time worked on the project by ISR experience). F. Preparing Parent and Child Sample A number of steps were involved in preparing the parent and child sample. An NSFH II clerk would attach all of the necessary forms to a case before the case was released to the telephone interviewing staff. 1. Contact Information Sheet (CIS) The most important component to the parent and child sample was the Contact Information Sheet (CIS). A CIS was generated for each focal child and each parent selected for a telephone interview. In a main respondent inter- view, when a focal child or a parent was selected for a telephone interview, we asked the main respondent for the name and telephone number of the selected parent or child and for permission to contact that person for a telephone in- terview. As the main respondent gave the name and telephone number, the in- terviewer typed this information into his/her computer. Field interviewers were told to transmit their completed cases to ISR on the same day or the day after the interview was completed. If a field inter- viewer did not transmit right away or had difficulty transmitting, the release of the parent and/or focal child case was delayed. Parent and child cases were only created after field interviewers did one of the following: success- fully transmitted the completed main respondent case, sent the case to ISR on a backup disk, or sent in their computer so the case could be retrieved. After the completed main respondent interview arrived at ISR, the parent or focal child case was set up on CATI. Once a week, contact information sheets were generated by David Tucker. The CIS for a parent case included: * case number * date of the 1987-88 interview with the main respondent * date of the follow-up interview with the main respondent * name of the main respondent (R1) * name of the selected parent * telephone number of the selected parent * relationship of the selected parent to the main respondent The CIS for the children included: * case number * date of the 1987-88 interview with the main respondent * date of the follow-up interview with the main respondent * name of the main respondent (R1) * address of the main respondent * focal child's first name and in some instances his/her last name * sex of the focal child * age of the focal child * whether or not the focal child is the biological child of the main respondent's spouse * whether the focal child was living in the main respondent's household at the time of the follow-up interview * relationship of the focal child to the main respondent at the time of the follow-up interview * focal child's address at the time of the follow-up interview, if the child was not living with the main respondent and did not have a telephone The parent and child cases were assigned the same five digit case number as the corresponding main respondent case. The letter following the case number was different for each type of respondent. The parent case numbers ended in P, older child case numbers ended in O and younger child case numbers ended in Y. 2. Call Report Form (CRF) After the contact information sheets were generated, a blank Call Report Form (CRF) was attached to each one. A clerk would then write the case number and the time zone on each CRF. The CRFs used for the three types of cases were identical in function and in content. The CRFs were color coded so each type of case could be easily identified. The CRFs for parent cases were print- ed on gray paper, older child CRFs were printed on lavender paper and younger child CRFs were printed on blue paper. 3. Parent/Child Address Form The next item attached to a case was the Parent/Child Address Form. This form was created in June of 1993 to obtain the addresses of parent and child respondents and written documentation of the spelling of the respondent's name and telephone number. The main respondent filled out the Parent/Child Address Form as the in- terviewer was entering the information about the selected parent or selected child into the computer. The field interviewer mailed this form to ISR, along with other forms for that case. When the completed Parent/Child Address Form arrived at ISR, it was attached to the corresponding parent or child case. 4. Parental Consent Form for Child's Participation The final item that was attached to a case before being released to the telephone staff was the Parental Consent Form for Child's Participation. At the end of the main respondent interview, the interviewer was given a prompt by the computer to hand the main respondent a Parental Consent Form to sign and date. Parental Consent Forms were mailed to ISR by field inter- viewers, along with other forms for that case. At ISR the Parental Consent Forms were matched up with the corresponding cases. In some cases, parental consent was not obtained or the Parental Consent Form was not returned to ISR. If a parental consent form was not on file for a case, the telephone inter- viewer was required to call the main respondent and obtain parental permission by telephone before contacting the child. G. Preparing Main Respondent and Spouse/Partner Sample A few steps were involved in preparing main respondent and spouse/partner sample being transferred to the telephone staff from the field. When cases arrived from the field, Lisa Famularo or a telephone super- visor would look over each case. The cases were reviewed to find the best address and telephone number, which were then written on the front of the screening form and highlighted. Cases without telephone numbers and cases which resulted in a field refusal were separated from the rest of the sample. All cases were passed along to an NSFH II clerk who attached a blank telephone study Call Report Form to each case before the case was released to the tele- phone interviewing staff. RC, for "refusal conversion", was written in red in the upper right corner of the screening form of all cases which had resulted in a refusal for a field interviewer. A refusal letter was mailed to each of these respond- ents. The cases were released to the telephone staff a few days after the letter was mailed. Cases without telephone numbers were reviewed for tracing leads. If a lead was found, the case was released to the telephone staff. If leads were not found, the case was turned over to the tracing staff. H. Data Collection 1. Assigning Cases Assignments were made by the supervisor at the beginning of each shift. When assigning cases, priority was given to old cases. Non-final sample was returned to the supervisor at the end of each shift to ensure that cases which resulted in answering machines and no answers would be called on different days and at varying times. Interviewers kept, in their personal files, cases on which they had built up a rapport with a respondent or an informant, as well as partial interviews, definite appointments and tracing cases. Supervisors checked interviewers' files on a weekly basis to make sure that interviewers were following up on partial interviews and trac- ing cases, and keeping scheduled appointments. 2. Handling Cases With Non-Interview Results Interviewers were trained to provide detailed information in the comments section of the Call Report Form (CRF) after every call. Supervisors used this information to evaluate and reassign cases. If a respondent was too ill to be interviewed or hard of hearing, the case was put aside. In November of 1993, Jim Sweet gave us permission to be- gin conducting interviews with the second biological parent of the main re- spondent if the selected parent was too ill to be interviewed or hard of hear- ing. If a respondent was living overseas, we asked the informant for the over- seas address, the telephone number and information about when this respondent might return to the United States for a visit. At first, we only conducted interviews with respondents living overseas while they were visiting the United States. We began making international calls in November 1993. If a respondent was institutionalized in jail, prison, mental hospital, nursing home, or other long-term care facility, the interviewer recorded the name and type of institution on the CRF. If the respondent was physically and mentally able to be interviewed, we attempted to complete the interview. Parent respondents who spoke a language other than English or Spanish were final coded as language barriers. Child respondents who spoke a language other than English were final coded as language barriers (there were only three). If the main respondent refused to give locating information about the selected parent or child or refused/revoked permission to interview the selected parent or child, the case was final coded as "main respondent refuses permission." Cases which resulted in a household member refusing for the re- spondent were held for a few weeks before being called again. Cases with calls resulting in a refusal or a break off were put aside for several weeks before a refusal conversion attempt was made. 3. Converting Parent and Child Refusals By March 1993, most of the interviewers were skilled and confident enough to start attempting to convert refusals. The cases from earlier months which resulted in refusals were reviewed and reassigned. We matched the gender and age of the interviewer to that of the respondent in the hopes that the sim- ilarities would make the respondent feel at ease and increase the inter- viewer's chance of completing the interview. From this point on, the process of evaluating and reassigning refusals was done on a regular basis. When reviewing the refusals, we found that most of the parent respondents refused to participate because they had never heard of the study and did not understand the purpose of it. Most parent and child respondents stated that the main respondent did not mention the study to them. Parent and child re- spondents did not receive introductory letters so it was up to the interviewer to explain the study and how the respondent was selected to be part of it. In June 1993, Allan Walters wrote a letter for parent respondents who refused to be interviewed. The letter was mailed to respondents who initially refused but later agreed to give us their address so that study information could be sent to them. Letters could not be mailed to all respondents who refused to participate in the study because address information on parent and child respondents was not obtained during the main respondent interview. For those parent respond- ents who refused to give us their address, another mailing strategy was used. We had on file the reference information the main respondent provided at the 1987-88 interview. In many cases, the parent selected for the telephone in- terview was listed as a 1987-88 reference. If the telephone number on the CIS was the same telephone number listed for that parent on the 1987-88 reference screen, we assumed that the 1987-88 address was the current address and mailed a letter to that address. The Parent/Child Address Form was created so address information for parent and child respondents would be obtained at the same time permission to do the telephone interview was obtained. By September 1993, about half of the new parent and child cases being released to the telephone staff had Parent/ Child Address Forms attached to them. Many Parent/Child Address Forms were not returned to ISR. This probably occurred because the interviewer conduct- ing the main interview forgot to give the form to the respondent to fill out. Unlike the other forms, there was no prompt in the questionnaire reminding the interviewer to give the Parent/Child Address Form to the respondent. In September 1993, Ellin Spector wrote a revised refusal letter. The revised refusal letter was mailed to parent respondents with an NSFH brochure which showed results from the 1987-88 study in the form of bar graphs, pie charts and statistical summaries. Interviewers were instructed to offer to send the letter and brochure to respondents who were hesitant about doing the interview or on the verge of refusing. The refusal letter was effective. The information provided in the letter and brochure apparently proved to suspicious respondents that the study was legitimate. The brochure was also an effective way of showing respondents how the information would be used and in what form it would be reported. Final results indicate that 43% (470/1084) of the parent refusals and 40% (50/126) of the child refusals were converted. 4. Problems Caused by Data Entry Errors a. Misspelling From the beginning, telephone interviewers complained about inaccurate information on the CIS. The information appeared on the CIS exactly as the interviewer who completed the main respondent interview typed it into his/her computer. Unfortunately, interviewers sometimes misspelled respondents' names and typed incomplete telephone numbers. Telephone interviewers received cases with telephone numbers, but no area codes; telephone numbers with transposed digits; telephone numbers missing digits; cases with "refused," "don't know," or "///" instead of a telephone number; and respondents named "refused," "don't know," "ca," "b" or "///". Telephone interviewers tried to trace those respondents but lack of address information and misspelled names made it al- most impossible to trace through directory assistance. Telephone interviewers also had difficulties figuring out the area codes without city and state in- formation. After the interviewers conducting main respondent interviews started using the Parent/Child Address Forms, the completed forms were attached to the corresponding cases before the cases were released to the telephone staff. The name and address information provided on the forms helped the telephone interviewers in several ways: 1) The respondent's address allowed the tele- phone staff to trace respondents through directory assistance. 2) The hand- written information acted as a back up to the information that was often in- correctly typed into the computer by the interviewer. 3) The respondent's address allowed us to send letters asking respondents who did not have tele- phones or had non-published telephone numbers to call in on the toll-free number. b. Incorrect Coding If the main respondent did not know the whereabouts of the selected parent or child, or refused permission to interview the selected parent or child, the case should not have been released to the telephone staff. Some of these cases were released to the telephone staff due to errors in data entry on the part of the interviewer who conducted the main respondent interview. At the end of the main respondent interview, there is a question which tells the main respondent we want to do a telephone interview with the select- ed parent or child and asks for his/her telephone number and name. On this question, the interviewer can choose one of the following response categories: 1) Respondent gives the telephone number 2) Selected parent/child has no telephone number 3) Respondent refuses to give the telephone number 4) Re- spondent doesn't know the telephone number 5) Respondent doesn't know the whereabouts of selected parent/child. Without waiting to hear the respond- ent's answer, some interviewers would automatically type in code 1 (respondent gives the telephone number) after reading the question. If the respondent answered with something other than the telephone number and name, the inter- viewer would type in something like "refused," "don't know," or "///" on the question which asks for the telephone number and name, rather than going back to the previous question and correcting their entry to reflect the ap- propriate code. Some interviewers did attempt to go back to the previous question and change the answer, but they did not do it successfully. This is why telephone numbers or respondents were listed as "b" (the CAPI command to back up) or "ca" (the CAPI command to change an answer.) c. Incorrect Data Entry Telephone interviewers received cases where reference names were listed instead of the name of the selected parent or child. We believe this occurred because interviewers conducting the main respondent interview did not read the question in which they should have asked for the parent or child's name. Instead, the interviewer assumed that this was the question which asked for the name and telephone number of a reference. When a telephone interviewer received one of these cases, he/she had to call the main respondent back to find out the name and telephone number of the selected parent or child. After this problem was discovered, we began emphasizing the importance of reading and answering these questions properly at training. 5. Tracing Parent and Child Respondents When the study was planned, it was not anticipated that telephone inter- viewers would have difficulty contacting parent and child respondents. It was thought that the parent and child respondents would be contacted soon enough after the main respondent provided us with information about how to reach them that very few respondents would move during this time period. Some respond- ents did move before we had a chance to contact them and tracing was required in order to find the correct telephone number at which to contact the respond- ents. However, many of the cases which required tracing were not of this type. In fact, many needed tracing because of the data entry errors described in the previous section: * information on the CIS was incomplete or inaccurate * cases were released to the telephone staff in error * parent and child respondents without telephones were included in the sample In the beginning, telephone interviewers did minimal tracing on parent and child cases. Interviewers made a directory assistance call looking for a name match for the respondent and/or called the main respondent for the in- formation. In many cases, the main respondent could not be reached because we did not have a telephone number on file or the telephone number we had was disconnected or no longer the correct number. If a telephone number for the main respondent could not be obtained through directory assistance, the inter- viewer turned the case in to the supervisor. In the beginning of the study, these cases were put aside as "traced out." In the fall of 1993, interviewers began tracing parent and child respond- ents through references given by the main respondent in the 1987-88 interview. In November of 1993, a tracer was hired to do intensive tracing on these re- spondents. This individual traced parent, child and main respondents through the 1987-88 references. If the telephone numbers for the 1987-88 references were wrong numbers or disconnected, she traced the references through dir- ectory assistance. In February 1994, we began using the reference information from the Wave 2 interview with the main respondent to locate parent and child respondents. Unlike the 1987-88 reference information, the Wave 2 reference information was not readily available to the telephone interviewers and tracers because it was not part of the Unify database. A supervisor and a clerk were taught how to look up this information. Looking up Wave 2 reference information required one supervisor to go into manager mode in CATI and move the case into "middle"; enter the main interview as a coder and jump to the end of the in- terview where the reference information was recorded, then hand write the reference information on the tracing form. This process was time consuming, yet very beneficial. The Wave 2 reference information was usually more help- ful than the 1987-88 reference information because it was recent. When looking up the Wave 2 reference information, we occasionally found other, even more valuable information. When moving through the questions in the main interview, we would sometimes notice "note exists" at the top of the screen, which means that the interviewer who conducted the main interview left a note. Often the notes were in reference to the selected parent or focal child. When reading these notes, we found useful information, such as a sel- ected parent or focal child was living overseas or in a prison, a dormitory, fraternity house, boarding home, nursing home or in some unusual living sit- uation. If the selected parent or child was living in one of these places and had a telephone number that included an extension number or a room number, the interviewer could not enter the entire telephone number into the computer be- cause the telephone number field only allowed a three digit area code and seven digit telephone number. In some cases the note would give a beeper/ pager number, the telephone number of a neighbor with instructions on who to ask for to get in touch with the respondent, or a prison telephone number in- cluding the name of the parole officer, warden or contact person. The only way that the interviewer could leave additional information about the tele- phone number was as a note. However, information left as a note did not appear on the CIS. The questionnaire was not designed to facilitate the entering of detailed information about a respondent's whereabouts so that it would appear on the CIS. 6. Parent and Child Respondents Without Telephones At the Wave 2 interview, if the main respondent said that the selected parent or child did not have a telephone, we collected the address of the sel- ected parent or child. The case was later released to the telephone staff without a telephone number. Telephone interviewers were instructed to call directory assistance for these cases in the event the respondent did have a telephone. If a telephone number was not obtained through directory assist- ance, the telephone interviewer would call the main respondent for additional information on how to get in touch with the respondent. Telephone inter- viewers were trained to ask main respondents and other informants the follow- ing types of questions: * Is there a work telephone number where she can be reached? * How often does she visit you? Could we call her at your house? * Is there a neighbor or a friend's telephone number where she can be reached? * Please ask her to call our toll-free number and give this case number, so we can interview her. * What is her current address? * Could we mail a letter to you so you can pass it on to her? In September of 1993, a letter asking respondents to call us on the toll- free number was written. At that time, we began mailing these letters, along with an NSFH brochure, to respondents who did not have telephones and to re- spondents we could not get telephone numbers for. When we did not have an address for a respondent, we mailed the letter in care of the main respondent. Cases with only an address and no contact person were held in a file marked "waiting for respondent to call in" after the letter was mailed. The letter did encourage some respondents to call in. It is difficult to estimate how many calls came in on the toll-free number as the result of this particular letter. Calls on the toll-free number were also the result of refusal letters and other letters mailed to respondents which included the toll-free number. Giving the number to references and family members resulted in calls to the 800 number as well. The CATI supervisors answered the 800 number in the evenings and on week- ends. When a respondent called in to do an interview, the supervisor immed- iately transferred the call to an interviewer to conduct the interview. To ensure that respondents who called in during the day could be interviewed right away, members of the daytime clerical staff were trained to conduct in- terviews. Follow up calls were made to main respondents and other informants who promised to pass along the 800 number or a letter. We found that making follow up calls around the holidays, when people typically get together or hear from family and friends they have not seen or heard from in a while, provided us with information we may not have gotten had we called at another time. 7. Spanish Language Data Collection Telephone interviewing of main respondent and spouse cases in Spanish began in May 1994. Interviewing of parent respondents in Spanish began when the questionnaire translation was finished late in May 1994. All bilingual interviewers were first trained in English. After comp- leting several interviews in English, each interviewer went through a mock interview in Spanish on his/her own. Bilingual interviewers were then paired up for one-on-one mocks. One interviewer assumed the role of the respondent while the other assumed the role of the interviewer. For the second mock, the interviewers reversed roles. Five bilingual interviewers were recruited. One had a lot of difficulty with the English questionnaire and was never trained on the Spanish question- naires. Four interviewers were trained on the Spanish questionnaires but only two completed Spanish interviews. Main respondent and spouse interviews conducted in Spanish were admin- istered using a laptop computer, with the interviewer reading the questions in the SE booklet to the respondent. Spanish parent interviews were conducted on CATI. I. Monitoring of Telephone Interviewers At first, telephone interviewers were not monitored by telephone. In- stead, from her desk a supervisor listened to an interviewer conducting an interview or stood behind the interviewer and filled out an evaluation form based on what could be heard. In September 1993, a statement alerting respondents to the fact that a supervisor may be listening to the interview was added to the introductory script. In October 1993, Lisa Famularo held a meeting with the telephone study supervisors to discuss monitoring procedures and interviewer evaluat- ions. Supervisors began telephone and screen monitoring interviewers at that time. Supervisors used the CASES monitor screen, which displays data on all terminals that are logged in to the questionnaire, including the interviewer's name, the case number he/she is working on, the current position in the quest- ionnaire, the last item answered and the code that the interviewer entered. In March 1994, video monitoring of an individual interviewer was made possible by Barry Hollander. This program allowed the supervisor to view the screen of any interviewer while listening in on the telephone. The supervisor could observe every entry the interviewer made, as he/she was making it, while the interview was conducted. This system allowed the supervisors to better evaluate interviewers. Interviewers were evaluated on interviewing skills and questionnaire administration. Interviewers were given immediate feedback about their per- formance and prompt action was taken to correct problems. Interviewers were monitored and paper evaluations were filled out soon after interviewers were trained. Paper evaluations were filled out again after an interviewer had worked on the project for several months. J. Telephone Interviewer Payment and Bonus Telephone interviewers were paid hourly, with a starting rate based on their ISR interviewing experience. Rates of pay ranged from $7.00 to $8.50 per hour. Twenty five cent raises were given to interviewers after working on the project for a year. A bonus system was used to motivate interviewers. A bonus was paid for each completed interview and each completed refusal conversion. Bonus pay- ments were made to interviewers quarterly. Full-time Temple University students were paid on the weekly student pay- roll. Full-time students were permitted to work up to 20 hours per week while school was in session and 40 hours per week during holiday breaks and summer vacation. Non-students submitted hours on an Interviewer's Weekly Payment Record. Non-students were permitted to work up to 40 hours per week. Payment records were approved by Lisa Famularo. K. Differences Between CATI and CAPI The CATI and CAPI portions of the study differed in the materials that were used, the focus of interviewer training and certain procedures. 1. Materials Some of the materials used by CATI interviewers were different from those used by CAPI interviewers. CATI interviewers had a different Interviewer Manual, Home Study Quiz, Call Report Form and Tracing Form. The CATI inter- viewers did not use a Computer Guide, a Record Of Work or show cards. The telephone portion of the study had a telephone Call Report Form with codes that were different from those used by the field interviewers. At CATI training on the main questionnaires, the interviewers were taught how to read the field Screening/Call Report Form, which included a Tracing Form. CATI interviewers used the information on this form but never recorded calls on it. Telephone interviewers always used the telephone Call Report Form, the tele- phone codes, a separate telephone Tracing Form and a tracing checklist. CATI interviewers did not use show cards. The show card categories appeared on the computer screen in the CATI version of the questionnaire. Telephone interviewers read all of the response categories to the respondent. 2. Focus of Training Another difference between the CATI and CAPI portions of this study had to do with interviewer training. Training CATI interviewers was in many ways easier than training CAPI interviewers. Most of the telephone interviewers were college students with, at minimum, basic computer skills and typing skills. Telephone interviewers learned the CATI commands quickly and without much difficulty. As a result, less time was spent on learning computer skills and practicing commands at CATI trainings. CATI interviewers needed less practice on the computer so fewer mock interviews were needed and one-on-one mocks were not necessary. 3. Procedures Questions in the Self Enumerated Booklet (SE) were read aloud and ans- wers were filled in by the CATI interviewer. The questions in the SE booklet were not programmed into the CATI version of the main respondent and spouse/ partner questionnaires, and were not designed to be read by an interviewer, so much of the CATI training focused on reading those questions properly. CATI interviewers also read to the respondent and filled out the Partic- ipation Form and Parental Consent Form for Child's Participation. CATI inter- viewers wrote "obtained over telephone" in the space on the form designated for the respondent's signature. CATI interviewers did not give a weekly report to the supervisor. In- stead, supervisors reviewed cases in interviewers' files on a weekly basis. Contact results and final codes for cases assigned to the telephone staff were not entered in the Mac database. CATI interviewers had immediate access to information that CAPI inter- viewers had to obtain through their Coordinators or directly from ISR. CATI interviewers had access to Unify information for every case. CATI inter- viewers would use Unify for several reasons: * to look up 1987-88 reference information. * to look up a case to make sure it was not already finalized * to look up the status of a spouse/partner case if it was not with the main respondent case * to look at the address correction screen to see if any new in- formation from post office updates or tracing had been entered L. Coordination of Work Between Field and Telephone After the original group of cases was assigned to the telephone staff in March of 1993, transferring of cases to and from telephone continued until the end of the project. Several types of cases were transferred to the telephone staff. Cases that resulted in field refusals in areas where a clean up trip was not planned were transferred to the telephone room for a final refusal conversion attempt. Non-final cases that remained in a field interviewer's assignment at the end of the interviewer's field period were transferred to the telephone staff if no other interviewers were scheduled to work in that area. If a respondent moved to an area where the field interviewers had finished working, the case was transferred to telephone. A large number of cases were pulled from telephone and transferred back to the field on two or three different occasions. Other than those few times, cases were usually transferred out of telephone in groups of five or six at the request of the Coordinators. Pulling cases from telephone was a time consuming task. There was not an easy way to know which telephone interviewer had a case. All cases assign- ed to the telephone staff appeared in Unify and the Mac database assigned to D. Williamson. For a short time, the telephone supervisors had a list of all cases assigned to telephone. Every time the supervisor assigned a case she would write the interviewer's name next to the case number. This system was discontinued because it was difficult to maintain. The first problem was that the list of cases assigned to telephone was out of date every two or three weeks because cases were continually being transferred to and from telephone. The second problem was cases were reassigned to different CATI interviewers so frequently that record keeping mistakes were inevitable. A small number of cases assigned to telephone were transferred back to the field by Lisa Famularo. These cases were usually transferred because a telephone number could not be obtained but we had what we believed was a good address, the respondent was hard of hearing and could not do a telephone in- terview, or the respondent said he/she would only do a face-to-face interview. M. Final Results A total of 74 CATI interviewers were trained on this project. Of those interviewers, 78% were trained on the parent and child questionnaires, and 59% were trained on the main respondent and spouse/partner questionnaires. Most interviewers (82%) worked on the study for less than 12 months. As Table 5 shows, 36% of all interviewers trained on the study had pre- vious ISR CATI experience. Twenty-two percent of the ISR CATI experienced interviewers and 15% of the non-experienced interviewers worked on the study for over 12 months. Two interviewers, one with ISR CATI experience and one without, worked for the entire 21 month field period. TABLE 5: TIME WORKED ON THE PROJECT BY ISR EXPERIENCE Number of Months ISR CATI Experienced Interviewers Without Worked Interviewers ISR Experience --------------------------------------------------------------------------- 0-3 12 21 --------------------------------------------------------------------------- 4-6 7 17 --------------------------------------------------------------------------- 7-9 1 1 --------------------------------------------------------------------------- 10-12 1 1 --------------------------------------------------------------------------- 13-15 1 5 --------------------------------------------------------------------------- 16-18 4 1 --------------------------------------------------------------------------- 19-21 1 1 --------------------------------------------------------------------------- Totals: 27 47 --------------------------------------------------------------------------- All supervisors and 73% of the interviewing staff were female. As Table 6 shows, 19% of the female interviewers and 15% of the male interviewers work- ed on the study for over 12 months. The two interviewers that worked the en- tire 21 month field period were both female. TABLE 6: TIME WORKED ON THE PROJECT BY GENDER Number of Months Number of Male Number of Female Worked Interviewers Interviewers -------------------------------------------------------------------------- 0-3 8 25 -------------------------------------------------------------------------- 4-6 8 16 -------------------------------------------------------------------------- 7-9 1 1 -------------------------------------------------------------------------- 10-12 0 2 -------------------------------------------------------------------------- 13-15 1 5 -------------------------------------------------------------------------- 16-18 2 3 -------------------------------------------------------------------------- 19-21 0 2 -------------------------------------------------------------------------- Totals: 20 54 -------------------------------------------------------------------------- As Table 7 shows, the number of interviews each interviewer completed on the study varied widely. TABLE 7: NUMBER OF COMPLETED INTERVIEWS PER INTERVIEWER Number of Number of Completed Interviews Interviewers* ------------------------------------------------------------------ Less than 5 7 ------------------------------------------------------------------ 5-25 21 ------------------------------------------------------------------ 26-50 9 ------------------------------------------------------------------ 51-100 14 ------------------------------------------------------------------ 101-150 11 ------------------------------------------------------------------ 151-200 5 ------------------------------------------------------------------ 201-250 5 ------------------------------------------------------------------ 251-300 2 ------------------------------------------------------------------ 301-350 2 ------------------------------------------------------------------ 351-400 1 ------------------------------------------------------------------ 401-500 3 ------------------------------------------------------------------ 500+ 1 ------------------------------------------------------------------- * Total number of interviewers listed here is 81. In addition to the 74 interviewers trained at CATI trainings, 4 supervisors did some interviews, 1 person trained as a CAPI interviewer did CATI work also, and 1 experienced CATI interviewer was trained individually by Lisa Famularo. N. Problems and Their Solutions 1. Dissemination of Information A problem on this study was the dissemination of information to the tele- phone interviewing staff concerning changed procedures or other clarifications or reminders. We usually hung a sign in the interviewing room or at the supervisor's desk notifying interviewers of shift cancellations, deadlines for early submission of payment records and things of that sort. When there was a new policy or change in procedure, a memo was written and placed in each in- terviewer's file. Interviewers kept so many things in their files (sample, area code maps, bonus sheets, tracing sheets, etc.) that the memos, despite their brightly colored paper, were often overlooked. This did not seem to be the best way to keep interviewers informed. Having the entire interviewing staff get together for a weekly or monthly meeting was not possible and rely- ing on the supervisors to pass along information had not worked out well in the past. We thought a weekly newsletter would work best because it could be distributed on the same day each week so the interviewers would know to look for it in their files. With a little bit of creativity, a newsletter could be more interesting than a memo and maybe the interviewers would actually want to read it. In the beginning of November 1993, Lisa Famularo developed a newsletter that was distributed to interviewers every Monday. The newsletter kept inter- viewers informed of all changes and kept them up to date on the study's pro- gress. The newsletter provided them with necessary information about shift cancellations and other administrative details, as well as information they rarely had received previously, such as the total number of completions to date and total number of completions for the week. In an attempt to motivate the staff without using money, the newsletter was used to inform them of such things as weekly or monthly goals and the names of the interviewers who had the most completed interviews, the most completed interviews in one shift, the most refusal conversions and things like that. The newsletter was well re- ceived by the interviewers; they enjoyed having a systematic way of receiving information and liked being kept up to date on their progress. 2. Motivating Interviewers In an attempt to arouse the competitive side of some of the interviewers, at the start of the bonus period which began on November 15, 1994 three charts were hung in the interviewing room: one for completed parent interviews, one for completed child interviews and one for completed main respondent and spouse interviews. Each time an interviewer completed an interview, the supervisor would give that person a sticker to place next to his/her name on the appropriate chart. We thought this would be a new and fun way to keep track of progress, however, as expected, there was a mixed response from the interviewers. Some of the interviewers did not like the charts because they thought they were juvenile, while others enjoyed them and competed with one another to see who would end up with the most stickers. Some interviewers did not participate in the competition because money and prizes were not involved, while many did participate just for the fun of it. In April 1994, another attempt was made to motivate the interviewers. The bonus for completing a main respondent or spouse/partner interview was increased from $3 to $5. The bonus for completing a refusal conversion was increased from $5 to $7. Again, the response from the interviewing staff was mixed. Those that enjoyed doing the main respondent and spouse/partner inter- views were happy about the bonus. Interviewers that preferred to do parent or child interviews still wanted to work on those cases. However, the increase in the bonus for main respondent and spouse/partner interviews made the fact that we were requiring them to work on main respondent and spouse/partner cases more tolerable. At the same time the bonus was increased, interviewers working on main respondent and spouse/partner cases received a $5 bonus for every Friday even- ing and Sunday evening shift on which they worked at least 4 hours during the month of April. A $5 bonus was not enough to encourage interviewers who were not scheduled for those shifts to come in, but was enough to encourage those that were scheduled to come in and stay for the entire shift. VIII. THE CLEANUP PHASE OF THE STUDY Cleanup began on the National Survey of Families and Households-II in August 1993, when two waves of the field work were finished and the third wave was in progress. During the field period prior to that point, we had re- assigned refusals to other interviewers in the same areas and tried as much as possible to restaff or redistribute the work of interviewers who were term- inated or quit. Cleanup efforts continued through May 1994 in some areas of the country, and telephone interviewers continued interviewing respondents from the field sample until July 1, 1994. A. Changes Made During Cleanup We improved field interviewers' productivity and morale with three major policy changes in August 1993. With Jim Sweet's approval, we removed the limit on the number of interviews that could be conducted by any one inter- viewer, allowing our best interviewers to continue working on the project as much and as long as they were available. We began sending tracing reports containing reference names and addresses for those respondents known to be tracing problems directly to the interviewers for their use in finding re- spondents. Also, when necessary, we allowed interviewers to pay respondents incentives in cash at the time of the interviews. As planned, we reduced the number of Coordinators working on the project and consolidated the supervisory work after the training conferences were over in July. All Coordinators were sent a memo on July 28, 1993, detailing the planned changes and instructing them to take final reports and send their Macs and backup disk to the office on August 11. The remaining four Coordinators each handled two to four regions of the country after the consolidation. Using the four Macs that had greater memory capacity than the other seven machines, Allan Walters reloaded the databases to the appropriate machines and the Macs were sent back to the Coordinators. The 11 original regions were retained, because there was no pressing reason to change them, and collapsing regions would have meant major changes in the pro- gramming and setup of the Mac database. Linda DeWild, Marlene Blomberg, Sandy Lewis, and Elaine Trull remained as coordinators for the Cleanup Phase of the study. All of the four remaining Coordinators had tremendous work loads, but Linda DeWild and Marlene Blomberg received the most difficult and largest assignments. Interviewers were told about the cleanup changes by their original Co- ordinators. They were called within a week by the new Coordinator as planned, and a mutually acceptable reporting time was agreed upon. The changeover from one Coordinator to another caused little difficulty in the field. In September 1993, we instructed interviewers to send all screening forms that did not result in interviews directly to their Coordinators instead of to the office. In October 1993, the Coordinators began making all of their own cleanup assignments; until that point, Elaine Trull had made the assign- ments after conferring with the Coordinators. At the beginning of Cleanup, we pulled approximately 250 cases from the telephone room and returned them to the field, because the staff of inter- viewers trained to do field interviews by phone was small. Cleanup inter- viewers were going back to areas where those cases were located, and the cases from the telephone room were added to the others in the field interviewers' assignments. The respondent introduction and refusal letters were revised by Ellin Spector, who also wrote new letters to fit some of the special circumstances that had arisen. Coordinators, instead of office staff, began sending refusal letters when they reassigned refusals. When an interviewer did work in another Coordinator's area, it was with the approval of his/her original Cleanup Coordinator. The interviewer made travel plans with the help of the Cleanup Coordinator for that area, and re- ported only the work done in the new region to her. Interviewers were assigned a separate cleanup ID number for each assign- ment by Yvonne Shands, as she entered the cleanup assignment into the central office database. These IDs identified the Coordinator and specific cleanup assignment to office staff and Coordinators, and were not used by the inter- viewers. New ID numbers were needed to separate assignments in the Mac, because all respondents assigned to one ID number were listed alphabetically, not by address, in the Coordinator database. The cleanup IDs helped keep assignments separate for easier evaluation and allowed the Coordinator to readily figure the interviewer's bonus for a cleanup trip. All Interviewer Payment Records for cleanup were sent directly to the Coordinators, rather than to the office. Coordinators approved payment and used the IPR in figuring the appropriate bonus for that cleanup assignment. Coordinators sent a separate IPR for the bonus to the office. A Continuation Call Report Form was designed and distributed to inter- viewers, so that they would have room to report their contacts with respond- ents whose original Call Report Forms had been filled with contacts made be- fore Cleanup began. The Continuation Call Report Forms were color coded to match the original screening forms. Coordinators provided interviewers with receipt books to use when making cash incentive payments from their own funds to respondents. When inter- viewers used their discretion to pay a cash incentive, they had the respondent sign a receipt and attached it to their payment record for reimbursement. Interviewers did not receive cash advances for this purpose. Approximately 1264 incentives were paid in cash. Interviewers and Coordinators were required to return all screening forms to the office in December 1993 for a comprehensive inventory over the Christ- mas holidays. The last interviewers completing assignments from Wave 3 turned in any remaining work also, so that all work assigned after January 1, 1994 was designated as cleanup work. B. The Big Push--January through March, 1994 In January 1994, the Cleanup Coordinators came to Philadelphia for an all day meeting with ISR study personnel to evaluate the status of the study and plan strategies for completing the remaining work. Although Sandy Lewis was ill and could not make the trip, she participated in the meeting via a con- ference call. The inventory, completed in early January 1994 by Lisa Famularo and Yvonne Shands, accounted for all remaining cases without final results. New screening forms were created for missing screeners and duplicates were elimin- ated. After the strategy meeting, the Coordinators spent the rest of their visit resolving problems with the screening forms. Problems were checked in Unify and in the Macs to resolve discrepancies and the screeners were returned to the appropriate Coordinators. Coordinators combined remaining cases into cleanup assignments of about 25-30 cases, matched them with the appropriate interviewer, and sent cases to the office for transfer as quickly as possible. (A large number of these assignments were ready for transfer when they were sent in for the December inventory.) Interviewers were expected to complete assignments in their own areas before they were allowed to travel. The goal was to keep interviewers traveling and working these cases as quickly as possible. At the beginning of January 1994, our goal was to complete an additional 3500 main respondent and spouse cases. We maximized our cleanup effort in several of our most difficult loca- tions by sending a team of interviewers with a Coordinator to work together for one week. In January 1994, Linda DeWild took a team of 14 interviewers to New York City for concentrated effort in all of the boroughs. At the same time, Elaine Trull supervised a 14 member team working in South Jersey and Philadelphia. The team efforts yielded 100 interviews in NYC and 112 in South Jersey/Philadelphia, as well as numerous transfers and other final codes. In February, Marlene Blomberg guided a Los Angeles team of 14 for another 122 interviews. To make sure that proxy interviews were conducted for as many of the deceased or too ill main respondents as possible, we searched the records in the Coordinators' databases for more information about the proxy respondents. From notes found in the Macs, giving next of kin names and phone numbers and other pertinent information, proxy interviewers completed another 200 inter- views. Additional telephone tracers worked in the office in December 1993 to contact the references for all main respondent cases that were not completed. Their phone calls caused problems for field interviewers, who were contacting the same references at the same time, making some respondents complain of harassment. The Coordinators asked that the office tracing of main respondent references be stopped. The telephone tracing contact of references from the office was abandoned, but only 10 - 15% of the work had not been done. Telephone tracers continued to work, however, on tracing through a var- iety of other standard ways. They used all sources available for the case, and systematically tried every lead. At the same time, Cindy Geesey, a trac- ing supervisor at ISR, worked directly with interviewers on Cleanup trips and with their Coordinators. C. Payment for Cleanup In August 1993, we sent Cleanup Availability Memos to 50 interviewers whose earlier work on the study made us consider them as candidates for extra assignments. An additional 26 interviewers were sent Cleanup Availability Memos as they successfully finished their assignments. The notices contained the new payment schedule as well as information on the requirements for clean- up assignments. In December 1993, we sent an additional commitment memo to interviewers we wanted to continue working in January and beyond. Marlene Blomberg and Linda DeWild devised a new payment system using bonuses to motivate interviewers to maintain their best production. Under the new plan, basic pay of $50 and $45 for the main respondent and spouse interviews remained the same, but an additional bonus was added for inter- viewing at least 20% of the new cases assigned. The larger the percentage of completed interviews to assigned cases, the greater the bonus per case. We awarded bonuses of $10 to $60 per interview, and most cleanup assignments generated at least some bonus. For good addresses for respondents who had moved beyond an interviewer's area, payment was $20. As a reward for tackling the difficult work in NYC, cleanup interviewers who worked the January 1994 blitz were given an unexpected treat of the matinee of a Broadway show. D. Cleanup Problems and Their Solutions We learned too late that we did not have enough interviewers to handle the most difficult work of cleanup. The problem was having enough topnotch cleanup interviewers, not just having interviewers that we could travel to other locations to work. We learned that some of our best-producing inter- viewers in their own locations performed poorly outside their home terri- tories. We also had a number of good interviewers who had limitations on their travel time because of other jobs or commitments. In January 1994, we conducted training conference 16 to train experienced interviewers for traveling cleanup assignments. Most of the interviewers were new to ISR, and the difficult cleanup that was left was too much for many of them. By the time of cleanup, respondents we had been tracing from the begin- ning of the study were even harder to find. All of the factors that had hind- ered us in locating respondents earlier--the errors in names, the bad infor- mation from MetroMail, etc.--were still causing problems. However, beginning in January 1994, all cleanup interviewers were given tracing reports for all respondents in their assignments. Interviewers or Coordinators worked directly with Cindy Geesey, a tracing supervisor, to compare notes and develop leads while interviewers were traveling. Although the inventory in December 1993 helped solve part of the problem of duplicate screeners that were in circulation, both in the field and in the telephone room, other duplicates continued to enter the system as we strived to have every remaining case worked. At the end of May 1994, when all remain- ing cases were assigned to telephone, Lisa Famularo conducted a mini-inventory of screening forms to remove the duplicates that had been introduced into the system since December.