For some drugs that have multiple names, questions regarding the use of that drug may be asked for each distinct name. For example, the use of methedrine and desoxyn are measured separately in this study even though they are both methamphetamine.
To protect the confidentiality of respondents, all variables that could be used to identify individuals have been encrypted or collapsed in the public use file. These modifications should not affect analytic uses of the public use file.
Data were collected by Response Analysis Corporation, Princeton, NJ, under contract with the National Institute on Drug Abuse. The data and codebook were prepared for release by Research Triangle Institute, Research Triangle Park, NC, and the codebook was initially distributed by National Opinion Research Center, Chicago, IL, under contracts with the Substance Abuse and Mental Health Services Administration.
Some frequencies related to interview information detailed on page 79 of the codebook (FINLRES1, VSADLTCM, PHADLTCM, VSYTHCM, PHYTHCM) reflect overlapping records. For example, although "Adult/Youth" (ADULTYTH) indicates that 4,043 adults were interviewed, "Final Adult Result" (FINLRES1) indicates that an adult interview was completed in 4,706 cases. Crosstabulation of these variables revealed that adult interviews were completed in 663 youth cases. This apparently represents households in which both an adult and a youth were interviewed.
Data were weighted based on two weight factors required in the sampling plan. The product of the sampling weight factors is equal to the inverse of each respondent's selection probability. Adjustments were made to compensate for nonresponse and sampling error. Adjustments also included a poststratification to Census population estimates.
For selected variables, statistical imputation was done following logical imputation to replace missing responses. These variables are identified by the designation "IMPUTATION-REVISED" in the variable label, and the names of these variables begin with the letters "IR". For each imputation-revised variable there is a corresponding imputation indicator variable that indicates whether a case's value on the variable resulted from an interview response, logical imputation, or statistical imputation. The names of imputation indicator variables begin with the letters "II".