PreK=-3rd Data Resource Center: The First Six Years of Schooling and Beyond

Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K) Resource Guide

Introduction

About the Guide

This resource guide provides a brief overview of the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Kindergarten-Eighth Grade Full Sample (ECLS-K K-8) and specific instructions for obtaining the ECLS-K K-8 dataset, which you can download to your own computer. It also provides guidance in obtaining access to additional restricted-use data. ECLS-K users should refer to the User Guide, which provides greater detail on the topics discussed below.

About the Data

ECLS-K is a large, nationally representative survey of the school experiences of a cohort of U.S. children who were enrolled in public or private kindergarten during the 1998-99 school year. The ECLS-K should be distinguished from a parallel study, the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B), which follows through kindergarten children born in 2001. The National Center for Education Statistics (NCES), which is part of the U.S. Department of Education's Institute of Education Sciences, sponsors both ECLS-B and ECLS-K.

ECLS-K is a multi-method and multi-source survey examining children's development from kindergarten through high school in the context of their families, schools and communities. Data sources include parents, teachers, school officials, school administrative records, and direct student assessments. Beginning in grade three, children complete self-description questionnaires; at fifth and eighth grades, they also complete food consumption questionnaires.

Data from ECLS-K can be used to: (1) advance research on academic achievement in the elementary years, (2) assess the cognitive, motor and physical development of children at the start of their formal schooling and at critical junctures during the elementary school years, and (3) study relationships between the family, preschool and school experiences, and children's developmental processes. Research using these studies can enrich our understanding of children's early learning, development, and educational experiences.

The ECLS-K data are available in one combined K-8 public-use file with data from the base-year (kindergarten), first-grade, third-grade, fifth-grade, and eighth-grade data collections. It was created so that analysts can easily access all data that has been publicly released in any round of data collection between fall-kindergarten and spring-eighth grade. The K-8 file can be used to analyze data from any single round of data collection or from any combination of rounds (e.g., cross-year analysis). Restricted-use versions of the following files are available from NCES:

  • Eighth-Grade Restricted-Use Data File
  • Fifth-Grade Restricted-Use Data File
  • Third-Grade Restricted-Use Data File
  • First-Grade Restricted-Use Data File

Acknowledgements

This resource guide was originally prepared by Donald J. Hernandez, Senior Advisor for the Foundation for Child Development and was subsequently updated by Sara C. Lazaroff, ICPSR to reflect the final ECLS-K K-8 data file released by NCES. It was developed for the PreK-3rd Data Resource Center: The First Six Years of Schooling and Beyond, a website hosted by ICPSR with support from the Foundation for Child Development.

Sample

Base year data for ECLS-K includes a nationally representative sample of 21,260 children enrolled in kindergarten during the 1998-1999 school year. To date, seven rounds of data have been collected from the kindergarten sample. During the 1998-1999 (kindergarten) and 1999-2000 (first grade) school years, data were collected during the fall and spring semesters. The fall, first grade data collection was designed to allow nationally representative analyses of summer setback. The three most recent rounds of data collection were in 2001-2002 when the majority of the students were in the third grade; in 2003-2004 when the majority were enrolled in fifth grade; and in 2006-2007, when the majority were enrolled in eighth grade. Table 1 lists each round of data collection and a sample description. (See Chapter 4 of the User Guide for additional detailed information about sample design and implementation.)

Table 1. Round and Year of Data Collection with Sample Criteria

Round Grade / Year Eligible Sample
Round 1 (Base year) Kindergarten, Fall 1998 Full original sample (21,260) obtained using multistage probability sample design. In general, the target number of children sampled at any one school was 24.
Round 2 Kindergarten, Spring 1999 All base-year students plus an additional 1,426 children who were selected but did not participate in Round 1.
Round 3 First grade, Fall 1999 Subsample of schools selected by staged probability sampling at the regional and school levels. All participating children in selected schools were eligible for this round. A 50% subset of "movers"1 from these schools was also included.
Round 4 First grade, Spring 2000 All base-year students freshened2 with a sample of first graders (165) from the population of children who did not attend kindergarten in the U.S. during the 1998-1999 school year. 50% of base-year "movers" are included in this round of data collection.
Round 5 Third grade, Spring 2002 All students included in the first grade-spring sample.
Round 6 Fifth grade, Spring 2004 All students included in the third grade-spring sample. Children who moved between third and fifth grades were subsampled at a lower rate than earlier rounds. Sample size for Round 6 is 11,820 students.
Round 7 Eighth grade, Spring 2007 All students included in the fifth grade-spring sample.

1 Movers are children included in the base year sample due to the attendance at a sampled school but who have since transferred to another school. The ECLS-K makes efforts to retain in the sample a larger proportion of movers who are in special populations, such as those for whom English is not their first language and for students with a disability.

2 Freshening integrates children into the study that were not part of the sampling frame during the base year because they did not attend kindergarten in the U.S. in 1998-99. This includes students who were out of the country in 1998-1999, children that had immigrated after the 1998 collection cycle or children who were not enrolled in kindergarten.

At first grade, the sample was freshened to include children who had not been enrolled in kindergarten in the U.S. in 1998-99. This allows estimates to be made of first graders in 1999-2000. Note that each year small portions of students either repeat or skip a grade and thus are not in the same grade year as the majority of the cohort. However, the sample was not freshened at third or fifth grade. Therefore, the data, when properly weighted allow researchers to make national estimates about the sample cohort of kindergarteners in 1998-99 (or first graders in 1999-2000) when the majority of them were in third and fifth grade, respectively; it does not allow users to make estimates about third or fifth graders.

Appropriate weighting is essential to generalize from ECLS-K sample data to the U.S. national population. Weighting of data adjusts estimates for unequal selection probabilities, (such as due to oversampling of private schools, and Asian/Pacific Islander children) and sample protections for student movers who have an Individualized Education Plan for a disability. Weights also adjust estimates for instrument non-response across rounds. If weights are not used, estimates will not be nationally representative of the target population. See the "Learn More" section of this document for additional information about using weights in analyses of the ECLS-K data.

In addition, the ECLS-K utilizes a clustered sample design to reduce field costs. A clustered sample design can distort true sample variance if the level of clustering is not taken into account. Therefore, estimates of variance must be adjusted appropriately before testing hypotheses. This can be done using SUDAAN, AM, SAS survey procedures, WesVar, and other software programs. It is advised that analysts verify their preliminary weight selection with ECLS-K experts who can ensure they have made the proper selection. Researchers are encouraged to email ECLS with questions about selecting and using weights. See Appendix A of this document for information on weighting of ECLS-K data. Chapter 4 of the User Guide provides additional details on sampling methods, the calculation and use of weights, and adjusting variance estimates.

Data
Elements

Data for ECLS-K are collected from multiple sources. At each round, direct cognitive and physical assessments were administered, parents were interviewed, and teachers and school administrators completed self-administered questionnaires. ECLS-K field staff completed a school facilities checklist. Starting in the third grade round, students answered a short self-description questionnaire about their interests in academics, peer relationships, and self-esteem. Starting in the fifth grade round, students also completed a food consumption survey. Table 2 indicates the data collection instruments for each survey year.

Table 2: Survey Components by Grade of Data Collection

InstrumentK - Fall 1998K - Spring 19991st - Fall 19991st -Spring 20003rd - Spring 20025th - Spring 20048th - Spring 2007
Parent InterviewXXXXXXX
Child AssessmentsXXXXXXX
Teacher Questionnaire XXXXX
Teacher Questionnaire (teacher level)XX
Reading Teacher QuestionnaireXX
Mathematics Teacher QuestionnaireXX
Science Teacher QuestionnaireXX
Special Education Teacher Questionnaire XXXXX
Adaptive Behavior ScaleXX
Self-Description QuestionnaireXXX
Food Consumption QuestionnaireXX
School Administrator QuestionnaireXXXXX
Student Record AbstractXXXX
School Fact SheetX
School Facilities ChecklistXXXX
Salary and Benefits QuestionnaireX
Head Start VerificationX

Users can access more specific technical information on instruments, assessments, and other methodological issues at the following websites:

Table 3 briefly summarizes the topics available for each data component. Variables and measures are administered in the same form at each wave to the maximum extent possible given children's development stage at the time of data collection. Therefore, keep in mind that some of these data elements are present in only one or two of the data collection rounds.

Table 3. Data Instrument with Summary of Topical Content1

Data ComponentTopical Variable Content
Child assessmentsLanguage, reading, mathematics, general knowledge (focus on social studies and knowledge of the physical world for grades K-1 only), science (grade 3-5 only), psychomotor skills (kindergarten only), and height and weight measurements.
Student Questionnaires Self-description of academic interest and competence; peer relationships; self esteem. Consumption of high fat/sugar/salt foods, fast food, fruits and vegetables, availability of these foods (for 5th and 8th grade); socio-emotional development (children answered some of these questions themselves beginning in 3rd grade).
Parent interviewFamily structure; immigration status; languages spoken; child care; child's health; child's social skills; home environment; parental educational expectations; neighborhood characteristics; parent's educational attainment, employment status, and annual income; demographic background data; information on nonresidential parents.
Teacher questionnairesClass description; class organization; classroom characteristics; instructional information; parental involvement; professional development; teacher's grading practices; perceptions of school climate; teacher's demographic background; evaluations of child's cognitive skills, social skills, and academic ability.
Special Education Teacher QuestionnaireTeacher's characteristics; child disability; level of services; assisted technology; IEP goals.
School Administrator QuestionnaireInformation about the school; student body; teachers; school policies; characteristics of administrator; length of school year; courses taught.
School facilities checklistPresence of security measures; neighborhood characteristics.
School records abstract formChild's attendance; child's IEP.

1 Appendix E, Exhibit 3 of the Eighth Grade Documentation provides variables names

Variable Names

The variable name can be used as a guide to determine the data collection round and instrument from which it is derived. With few exceptions, such as child identifiers and select demographic background items, variables begin with a single character indicating a data source. Table 4 presents a guide to data sources based on the initial character of variable names. The number (1-7) following the initial character in variables names usually indicates the round of data collection (See Tables 1 and 2 of this document).

Table 4. Data Sources by Variables' Initial Characters

Variable Instrument Code Data Source Variable Instrument Code Data Source
APrimary teacher questionnaire A (kindergarten, 1st, 3rdgrades)LSchool fact sheet (3rd grade)
BPrimary teacher questionnaire B (kindergarten, 1st, 3rd grades)MMathematics teacher child level questionnaire
CDirect child assessment and weight variablesNScience teacher child level questionnaire
DSpecial education teacher questionnaire APParent interview
ESpecial education teacher questionnaire BRDerived child demographic or child status variables
FField Management SystemSSchool administer questionnaire
GReading teacher child level questionnaireTPrimary teacher questionnaire C (kindergarten, 1st, 3rd grades)
IFImputation flagsUStudent record abstract
JPrimary teacher questionnaire (5th grade)W1Parent composite variables
KSchool facilities checklist

1 "W" coded variables follow a different numbering convention from the rest of the data catalogue. Variables beginning with "W" are followed by the grade number (K, 1, 3, 5, 8) to which they apply rather than the round number.

How to Obtain Data and
Documentation Files

Downloading Data and Documentation from the PreK-3rd Data Resource Center

Data from the Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K), Kindergarten-Eighth Grade are made available through the Child Care and Early Education Research Connections project, a data archive within ICPSR.

Researchers interested in downloading analysis ready data and documentation files can do so free of charge through the PreK-3rd Data Resource Center's Datasets & Resource Guide Web Page. Data are available in 3 formats: SAS, SPSS and STATA. Raw ASCII data are also provided with accompanying setup (syntax) files. Documentation is provided in PDF format.

To download the K-8 data and/or documentation, researchers must agree to the terms and conditions of use. To download files, select the Datasets & Resource Guides from the menu options located in top right corner of the PreK-3rd Data Resource Center home page. Locate the Access Data link for the Early Childhood Longitudinal Study, Kindergarten Class of 1998-1999, Kindergarten-Eighth Grade Full Sample. Clicking on the link will redirect you to the Research Connections website and the Resource Page where the user can choose to download files, browse documentation, etc.

Downloading and Ordering Data and Documentation from NCES

Researchers can also obtain the public-use data and corresponding documentation from the Early Childhood Longitudinal Study Kindergarten Cohort by requesting the ECLS-K Longitudinal Kindergarten-Eighth Grade Public Use data file on CD-ROM free of charge from the EDPubs on-line ordering system or by downloading it from the NCES website. Chapter 8 of the User Guide has step-by-step instructions for using the Electronic Codebook (ECB), an application that allows the user to generate programs in SAS, STATA or SPSS. Prior public-use longitudinal data files (i.e. K-3) and cross-sectional data files are no longer disseminated.

Cross-sectional and Longitudinal Data Files

Unlike the public-use longitudinal files released in previous rounds, this final K-8 public-use file contains all data and weights for all ECLS-K sample cases that have been publicly released in any of the rounds. Thus, it can be used for within-year (cross-sectional) analyses of any round of data collection and cross-year (longitudinal) analyses of combinations of rounds. The ECLS-K restricted-use data are available in cross-sectional format. It is recommended that users always turn to either this final K-8 public-use file, or to the restricted data files. For most research needs, the final K-8 public-use file is most appropriate.

Cross-sectional files contain records for every respondent in that particular round. The longitudinal files contain records only for those who participated at the end of the round range specified (i.e., a K-5 file would contain records only for children who responded through the fifth grade round; it would not include records for children who left the study prior to fifth grade). Therefore, the longitudinal files did not contain records for all ECLS-K participants over time; rather, they included the sample of children who have parent or child data in spring kindergarten and also have parent or child data for at least one round of the first grade data collections. Sampling weights adjust for non-response over time.

Further, child assessment scores are recalibrated at every round. Therefore, assessment scores from prior rounds contained on the most recent longitudinal file will vary somewhat from scores on previously released files. Accordingly, if the user is interested in longitudinal outcomes at third grade, it would be more appropriate to use the K-8 public data file.

There is another difference between cross-sectional files and longitudinal files. Not all variables were retained from cross-sectional files when converted to a longitudinal file format. Therefore, the restricted cross-sectional files have more records (children) and more variables for a given round than the public combined K-8 longitudinal file.

Restricted-Use Data Files

NCES releases the ECLS-K data restricted-use files. Restricted-use files differ from the public-use version in that they contain variables that have been suppressed in the K-8 public-use version to protect respondent confidentiality. Nonetheless, suppressed variables are labeled on the K-8 public-use file; data values are simply set to a reserve code. Other modifications found in the K-8 public-use file include composite variables describing children's family structure and family member characteristics. (See Table 7-16 of the User Guide for a list of variables that are modified or suppressed in the K-8 public-use version of the data). If those data are of particular value to the analyst, the user can apply on-line for a restricted-use data license.

First Steps toward Obtaining Your Analytic File

Before downloading the data or beginning analysis, it is important for the user to become familiar with the User Guide and Questionnaires.

User Guide. The ECLS-K User Guide provides the most detailed information about the coding and construction of variables, particularly composite variables, as well as information about child assessments and how they were administered. The User Guide can be downloaded from the documentation page. There is one User Guide per round and each guide separates information about data elements according to topical headings, with a Combined Eighth Grade and K-8 User Guide. While the Combined Eighth Grade and K-8 User Guide contains the most up-to-date information regarding ECLS-K, researchers may still need to refer to the previous User Guides for information specific to the round(s) they are studying. Users interested in learning more about a particular assessment or questionnaire can review the table of contents of the User Guide to locate the appropriate subheading.

Questionnaires. The survey instruments represent most fully the data actually collected for ECLS-K. It is strongly recommended that a user review the questionnaires to assure that all relevant variables for their research have been identified and that skip patters are clear to the user. Users interested in reviewing the wording of individual questions also should consult the questionnaires which can be downloaded from the documentation page.

Questions in each questionnaire are numbered; however, due to the complex fashion in which many composite variables (e.g., assessment test scores) are obtained, users may not be able to determine from which questions a specific variable has been constructed without also reviewing Chapter 7 of the round in question's corresponding User Guide, which discusses composite variable formulation in detail.

Learn
More

Appendix A

The Importance of Using Weights

This section provides information about using weights in analyses of the ECLS-K data. It is adapted from material developed for NCES training seminars. It is advised that analysts verify their preliminary weight selection with ECLS-K experts who can ensure they have made the proper selection. Researchers are encouraged to email ECLS with questions about selecting and using weights.

Why use weights? Weights are used for the following reasons:

  1. To make inferences to the population being studied--e.g., the ECLS-K weights can be used to estimate the total number of kindergartners in the United States who participated in a specific type of program.

  2. To adjust for differential sampling rates--e.g., certain groups of children sampled at a higher rate.

  3. To adjust for differential nonresponse--e.g., not all parents agreed to be interviewed and whether or not a parent agrees can vary by characteristics of the parent and family.

Weights Adjust for Differential Sampling Rates

ECLS-K uses a complex sample design. To provide acceptable precision for estimates about certain subgroups of the population, these subgroups were oversampled (sampled at a higher rate than they appear in the population). Asian children and children attending private schools were oversampled. The initial base weights adjust for the sample design and the final weights that appear on the data files also take into account differential non-response. Table 1 shows how distributions can change dependent on weighting the data. To illustrate the separate influence of the sampling design, the distributions of race/ethnicity and school type are shown unweighted, weighted by the base weight, and by the final weight (in this case, the fall- kindergarten child weight). Notice that the unweighted distribution of race/ethnicity overestimates the percentage of Asian children and the unweighted distribution of school type overestimates the number of children attending private schools. It is important to apply weights for all analyses with the ECLS-K data, regardless of the data collection round or the variables included in the analyses.

Table 1. Unweighted and weighted distributions of ECLS-K sample children, by race/ethnicity and school type: Fall 1998

Characteristic Unweighted
% Distribution
Weighted
(BASEWT)
% Distribution
Weighted
(C1CW0)
% Distribution
Race/ethnicity
White575658
Black151616
Hispanic182019
Asian633
School Type
Public788785
Private221315

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Base Year Public-use Data File, Fall 1998.

Weights Adjust for Differential Non-response

Not all potential respondents are equally likely to provide data. Some children and parents are not located or do not agree to participate. The more dissimilar respondents are from non-respondents, the greater the risk of non-response bias in the data. To illustrate some of the ways respondent and non-respondent parents differ, Table 2 shows the percent of spring first-grade respondents and non-respondents by selected family characteristics (data about these characteristics were obtained in the kindergarten year).

Table 2. Characteristics of first-grade children with (respondents) and without (nonrespondents) ECLS-K spring first-grade parent interview data: Spring 2000

CharacteristicRespondents (%)Nonrespondents (%)
Child's race-- white5941
Two-parent family7561
Food secure household9187
Non-English speaking household1118
Mother's education-- BA or Higher2111
Top 20 % of socioeconomic status 1910

Note: Estimates are weighted by the base weight.

Source: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, First Grade Data File, Spring 2000.

The ECLS-K weights will reduce the bias that is present in unweighted estimates. To illustrate the separate influence of the differential non-response, the distribution of selected family characteristics at first grade are shown unweighted, weighted by the base weight, and by the final weight (in this case, the spring first-grade parent weight).

Table 3. Unweighted and weighted distributions of first-grade children, by selected family characteristics: Spring 2000

Characteristic Unweighted
% Distribution
Weighted
(BASEWT)
% Distribution
Weighted
(C4PW0)
% Distribution
Socioeconomic status
Bottom 20%171920
Middle 60%596060
Top 20% 242120
Family type
Two parents787673
Single parent202224
Other222
Mother's education
Less than high school131516
High school diploma, no BA626363
BA or higher252221
Location
Large and mid-sized city393836
Suburbs and large towns394043
Small town and rural212321

Note: Details may not sum to 100 percent due to rounding.

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, First Grade Public-use Data File, Spring 2000.

The Importance of Choosing the Appropriate Weight

All weights in the ECLS-K base-year database were developed so that they sum to the population of kindergarten children, kindergarten teachers, or schools offering kindergarten programs in the 1998-1999 school year. First-grade cross-sectional weights were developed such that they sum to the population of first-graders. Weights are based, in part, on responses to certain instruments. For example, cases with a non-zero C1PW0 (fall-kindergarten cross-sectional parent weight) are respondents to the fall-kindergarten parent interview. Similarly, cases with a non-zero C2PW0 (spring-kindergarten cross-sectional parent weight) are respondents to the spring parent interview. Cases with non-zero weights, which are based on multiple sources (e.g., C1CPTW0 - child, parent and teacher), are respondents to each of the multiple sources (e.g., for C1CPTW0 - they are respondents to the fall child assessment AND the fall parent interview AND the fall teacher questionnaire). Detailed descriptions of how the base-year, first-grade, third-grade, and fifth-grade ECLS-K weights were calculated and how to choose the appropriate weight for your analysis are found in chapter 4 of either the Base Year, First Grade, Third Grade, or Fifth Grade User's Manual. Details on the kindergarten-first grade longitudinal weights can be found in Chapter 9 of the First Grade User's Manual or in the user's guides accompanying the ECLS-K Kindergarten-First Grade Longitudinal ECB. Details on the kindergarten-third grade longitudinal weights can be found in Chapter 9 of the Third Grade User's Manual and the kindergarten-fifth grade longitudinal weights are described in Chapter 9 of the Fifth Grade User's Manual.

Tables 4 through 7 show how the number of cases used in an analysis changes depending on the weight selected (table 4 is adapted from table 4-10 in the Base Year User's Manual; table 5 is adapted from tables 4 through 11 and table 9-2 in the First Grade User's Manual; table 6 is adapted from tables 4 through 12 and table 9-1 in the Third Grade User's Manual, and table 7 is adapted from tables 4 through 17 and table 9-1 in the Fifth Grade User's Manual.

Tables 4 through 7 do not reflect item non-response. For most variables in the database, item non-response is low. There are some exceptions and users need to decide how to deal with these in their analyses.

Table 4. Base-year weights, number of cases, and population sizes (sum of weights)

WeightNumber of cases
(Sample n)
Weighted number of cases
(Population N)
CROSS-SECTIONAL WEIGHTS
C1CW019,1733,865,946
C1PW018,0973,865,946
C1CPTW017,1243,865,946
C2CW019,9673,863,512
C2PW018,9503,863,512
C2CPTW017,4543,863,512
LONGITUDINAL WEIGHTS
BYCW018,2113,863,204
BYPW016,9063,863,204
BYCPTW015,4203,863,204
BYCOMW017,0603,863,204

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Base Year Public-use Data Files.

Table 5. First-grade weights, number of cases, and population sizes (sum of weights)

WeightNumber of cases
(Sample n)
Weighted number of cases
(Population N)
CROSS-SECTIONAL WEIGHTS
C3CW05,2913,858,882
C3PW05,0713,858,850
C4CW016,7273,938,490
C4PW015,6263,937,097
C4CPTW013,4913,935,870
LONGITUDINAL WEIGHTS
C34CW05,0473,850,650
C34PW04,6823,849,405
C23CW05,2163,858,997
C23PW04,8613,858,805
C123CW04,7293,858,824
C123PW04,2953,858,493
C24CW016,3713,844,009
C24PW014,9383,842,784
C124CW015,0013,844,473
C124PW013,4133,841,464
C1_4CW04,5423,850,620
C1_4PW04,0123,847,785
Y2COMW013,9833,842,961

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, First Grade Public-use Data Files.

Table 6. Third-grade weights, number of cases, and population sizes (sum of weights)

WeightNumber of cases
(Sample n)
Weighted number of cases
(Population N)
CROSS-SECTIONAL WEIGHTS
C5CW014,4703,938,513
C5PW013,4893,937,758
C5CPTW010,3953,937,125
LONGITUDINAL WEIGHTS
C45CW013,9643,935,960
C45PW012,6523,934,550
C245CW013,6943,843,642
C245PW012,2043,843,272
C1_5FC012,5583,843,607
C1_5FP010,9983,842,954
C1_5SC04,0323,841,183
C1_5SP03,5223,840,278

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Third Grade Restricted-use Data Files.

Table 7. Fifth-grade weights, number of cases, and population sizes (sum of weights)

WeightNumber of cases
(Sample n)
Weighted number of cases
(Population N)
CROSS-SECTIONAL WEIGHTS
C6CW011,3463,936,156
C6PW010,9963,935,007
C6CPTR010,1203,935,344
C6CPTM05,0173,946,324
C6CPTS05,1033,931,381
LONGITUDINAL WEIGHTS
C56CW011,1363,936,880
C56PW010,0793,935,347
C456CW010,8523,932,021
C456PW09,5683,931,097
C2_6FC010,6733,837,337
C2_6FP09,2673,836,988
C1_6FC09,7963,837,337
C1_6FP08,3703,836,478
C1_6SC03,0003,822,526
C1_6SP02,5663,823,589

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Fifth-Grade Restricted-use Data Files.

8. Eighth-grade weights, number of cases, and population sizes (sum of weights)

WeightNumber of cases
(Sample n)
Weighted number of cases
(Population N)
CROSS-SECTIONAL WEIGHTS
C7CW09,3583,943,827
C7PW08,8093,944,166
C7CPTE08,2943,943,318
C7CPTM0 4,1303,945,141
C7CPTS04,164 3,941,257
LONGITUDINAL WEIGHTS
C67CW08,9603,944,055
C67PW08,5443,944,048
C567CW08,8273,943,678
C567PW08,0703,943,290
C4_7CW08,6333,939,414
C4_7PW07,7643,939,255
C2_7FC08,5033,840,561
C2_7FP07,5583,841,500
C1_7FC07,8033,840,438
C1_7FP06,8613,840,784
C1_7SC02,3693,836,993
C1_7SP0 2,0633,839,514

SOURCE: U.S. Department of Education, National Center for Education Statistics, Early Childhood Longitudinal Study, Kindergarten Class of 1998-99, Kindergarten-Eighth Grade Data Files.

Choosing the Appropriate ECLS-K Base Year Weight

ECLS-K Base-Year Cross-sectional Weights

Fall-Kindergarten

Child-level weights

C1CW0
fall-kindergarten direct child assessment data and child characteristics, alone or when in conjunction with teacher/classroom data
C1PW0
fall-kindergarten parent interview data, alone or in combination with child assessment data)
C1CPTW0
fall-kindergarten direct child assessment data combined with fall-kindergarten parent interview data and fall-kindergarten teacher data

Teacher-level weights

B1TW0
fall-kindergarten teacher data, questionnaire part A or B

Spring-Kindergarten

Child-level weights

C2CW0
spring-kindergarten direct child assessment data and child characteristics, alone or in conjunction with teacher/classroom data
C2PW0
spring-kindergarten parent interview data, alone or in combination with child assessment data
C2CPTW0
spring-kindergarten direct child assessment data combined with spring- kindergarten parent interview data and spring-kindergarten teacher data

Teacher-level weights

B2TW0
spring-kindergarten data from questionnaire part A; fall- or spring-kindergarten data from questionnaire part B; or combination of data from both rounds

School-level weights

S2SAQW0
school administrator data or facility checklist data

Note: There are no separate weights for Teacher Questionnaire, parts A or C; Student Records Abstract Form; Special Education Teacher Questionnaires; Adaptive Behavior Scales; or Facilities Checklists.

ECLS-K Base-Year Longitudinal Weights

Child-level weights

BYCW0
child direct assessment data and child characteristics from both fall- and spring-kindergarten, alone or in conjunction with teacher/classroom data
BYPW0
parent interview data from BOTH fall- and spring-kindergarten (alone or in combination with child assessment data)
BYCPTW0
any cross-round combination of child direct assessment data AND parent interview data AND teacher data
BYCOMW0
Both rounds of child assessment data in conjunction with one or more rounds of parent and/or teacher data

Note: There is no panel weight for teachers.

ECLS-K First-Grade Year Cross-sectional Weights

Fall-First Grade

Child-level weights

C3CW0
fall-first grade direct child assessment data alone or in conjunction with a) a limited set of child characteristics (e.g. age, sex, race-ethnicity), and b) school or teacher data collected in spring-first grade.
C3PW0
fall-first grade parent interview data alone or in combination with a) fall-first grade child assessment data, and b) school or teacher data collected in spring-first grade.

Spring-First Grade

Child-level weights

C4CW0
spring-first grade direct child assessment data, alone or in conjunction with any combination of a) a limited set of child characteristics (e.g. age, sex, race-ethnicity), b) any spring-first grade teacher questionnaire A, B or C data, and c) data from the school administrator questionnaire or facilities checklist
C4PW0
spring-first grade parent interview data alone or in combination with a) spring-first grade child assessment data, b) spring-first grade teacher questionnaire A, B, or C data, and c) data from the school administrator questionnaire or facilities checklist. Exception: If data from the parent AND child assessment AND teacher questionnaire A or B (not C) are used then C4CPTW0 should be used
C4CPTW0
spring-first grade direct child assessment data combined with spring-first grade parent interview data AND spring-first grade teacher data alone or in conjunction with data from the school administrator or facilities checklist

Note: There are no separate weights for the Teacher Questionnaires, School Administrator Questionnaire, Student Records Abstract Form; Special Education Teacher Questionnaires; Adaptive Behavior Scales; or Facilities Checklists.

ECLS-K First-Grade Longitudinal Weights

Within First Grade

Child-level weights

C34CW0
child direct assessment data from BOTH fall-first grade and spring-first grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race-ethnicity)
C34PW0
parent interview data from BOTH fall-first grade and spring-first grade alone or in combination with fall- and/or spring-first grade child assessment data.

Kindergarten - First Grade

Child-level weights

C23CW0
child direct assessment data from BOTH spring-kindergarten and fall-first grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity)
C23PW0
parent interview data from BOTH spring-kindergarten and fall-first grade (alone or in combination with child assessment data)
C123CW0
child direct assessment data from fall- AND spring-kindergarten AND fall-first grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity)
C123PW0
parent interview data from fall- AND spring-kindergarten AND fall-first grade
C24CW0
child direct assessment data from BOTH spring-kindergarten and spring-first grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity)
C24PW0
parent interview data from BOTH spring-kindergarten and spring-first grade
C124CW0
child direct assessment data and from fall- AND spring-kindergarten AND spring-first grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity)
C124PW0
parent interview data from fall- AND spring-kindergarten AND spring-first grade
C1_4CW0
child direct assessment data from all four rounds of data collection, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity)
C1_4PW0
parent interview data from all four rounds of data collection
Y2COMW0
child direct assessment data from fall- AND spring-kindergarten AND spring-first grade, in conjunction with parent and/or teacher data from spring-first grade, AND one or more base year rounds of parent and/or teacher data.

ECLS-K Spring-Third Grade Year Cross-sectional Weights

Child-level weights

C5CW0
third-grade direct child assessment data, alone or in conjunction with any combination of a) a limited set of child characteristics (e.g. age, sex, race-ethnicity), b) any third-grade teacher questionnaire A, B or C data, and c) data from the school administrator questionnaire or school fact sheet
C5PW0
Third-grade parent interview data alone or in combination with a) third-grade child assessment data, b) third-grade teacher questionnaire A, B, or C data, and c) data from the school administrator questionnaire or school fact sheet. Exception: If data from the parent AND child assessment AND teacher questionnaire A or B (not C) are used then C5CPTW0 should be used
C5CPTW0
Third-grade direct child assessment data combined with third-grade parent interview data AND third-grade teacher data alone or in conjunction with data from the school administrator or school fact sheet or facilities checklist

Note: There are no separate weights for the Teacher Questionnaires, School Administrator Questionnaire, Student Records Abstract Form; Special Education Teacher Questionnaires; Adaptive Behavior Scales; or Facilities Checklists.

ECLS-K Third-Grade Longitudinal Weights

First Grade - Third Grade

Child-level weights

C45CW0
child direct assessment data from BOTH spring-first grade and spring-third grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity).
C45PW0
parent interview data from BOTH spring-first grade and spring-third grade.

Kindergarten - Third Grade

Child-level weights

C245CW0
child direct assessment data from spring-kindergarten AND spring-first grade AND spring-third grade, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity).
C245PW0
parent interview data from spring-kindergarten AND spring-first grade AND spring-third grade.
C1_5FC0
child direct assessment data from four rounds of data collections involving the full sample of children (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade), alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, race/ethnicity).
C1_5FP0
parent interview data from four rounds of data collections involving the full sample of children (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade)
C1_5SC0
child direct assessment data from all five rounds of data collection, alone or in conjunction with any combination of a limited set of child characteristics (e.g., age, sex, and race/ethnicity).
C1_5SP0
parent interview data from all five rounds of data collection.

ECLS-K Spring-Fifth Grade Year Cross-sectional Weights

Child-level Weight

C6CW0
Fifth-grade direct child assessment data, alone or in conjunction with any combination of (a) a limited set of child characteristics (e.g., age, sex, race/ethnicity), (b) teacher-level data from any fifth-grade teacher questionnaire, and (c) data from the school administrator questionnaire or school facilities checklist.
C6PW0
Fifth-grade parent interview data alone or in combination with (a) fifth-grade child assessment data, (b) data from any fifth-grade teacher questionnaire (teacher-level or child-level), and (c) data from the school administrator questionnaire or school facilities checklist. Exception: If data from the parent AND child assessment AND teacher (child- and/or teacher-level) are used together, then one of the C6CPTR0, C6CPTM0 or C6CPTS0 should be used.
C6CPTR0
Fifth-grade direct child assessment data combined with fifth-grade parent interview data AND fifth-grade data from reading teacher (child- and/or teacher-level), alone or in conjunction with data from the school administrator or facilities checklist.
C6CPTM0
Fifth-grade direct child assessment data combined with fifth-grade parent interview data AND fifth-grade child data from mathematics teacher (child- and/or teacher-level), alone or in conjunction with data from the school administrator or facilities checklist.
C6CPTS0
Fifth-grade direct child assessment data combined with fifth-grade parent interview data AND fifth-grade child data from science teacher (child- and/or teacher-level), alone or in conjunction with data from the school administrator or facilities checklist.

ECLS-K Fifth-Grade Longitudinal Weights

Third grade - Fifth grade

Child-level weights

C56CW0
child direct assessment data and child characteristics from BOTH spring-third grade and spring-fifth grade, alone or in conjunction with any of the school, teacher, or classroom data.
C56PW0
parent interview data from BOTH spring-third grade and spring-fifth grade, alone or in conjunction with any of the child assessment, school, teacher, or classroom data.

First grade - Fifth grade

Child-level weights

C456CW0
child direct assessment data and child characteristics from spring-first grade AND spring-third grade AND spring-fifth grade, alone or in conjunction with any of the school, teacher, or classroom data.
C456PW0
parent interview data from spring-first grade AND spring-third grade AND spring-fifth grade, alone or in conjunction with any of the child assessment, school, teacher, or classroom data.

Kindergarten - Fifth grade

Child-level Weight

C2_6FC0
child direct assessment data and child characteristics from spring-kindergarten AND spring-first grade AND spring-third grade AND spring-fifth grade, alone or in conjunction with any of the school, teacher, or classroom data.
C2_6FP0
parent interview data from spring-kindergarten AND spring-first grade AND spring-third grade AND spring-fifth grade, alone or in conjunction with any of the child assessment, school, teacher, or classroom data.
C1_6FC0
child direct assessment data and child characteristics from five rounds of data collections involving the FULL sample of children (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade), alone or in conjunction with any of the school, teacher, or classroom data.
C1_6FP0
parent interview data from five rounds of data collections involving the FULL sample of children (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade) alone or in conjunction with any of the child assessment, school, teacher, or classroom data.
C1_6SC0
child direct assessment data and child characteristics from all six rounds of data collection, alone or in conjunction with any of the school, teacher, or classroom data.
C1_6SP0
parent interview data from all six rounds of data collection, alone or in conjunction with any of the child assessment, school, teacher, or classroom data.

ECLS-K Spring-Eighth Grade Year Cross-sectional Weights

Child-level Weight

C7CW0
child direct assessment or student questionnaire data from spring-eighth grade, alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-eighth grade teacher questionnaire (teacher-level or child-level), or (c) data from the spring-eighth grade school administrator questionnaire.
C7PW0
parent interview data from spring-eighth grade, alone or in combination with (a) spring-eighth grade child assessment or student questionnaire data, (b) data from any spring-eighth grade teacher questionnaire (teacher-level or child-level), or (c) data from the spring-eighth grade school administrator questionnaire. Exception: If data from the parent interview AND child assessments AND teacher-level (with or without child-level teacher) questionnaires are used together, then C7CPTE0, C7CPTM0, or C7CPTS0 should be used.
C7CPTE0
child direct assessment or student questionnaire data from spring-eighth grade with spring-eighth grade parent interview data and spring-eighth grade English teacher-level data with or without child-level data from the English teacher, alone or in combination with data from the spring-eighth grade school administrator questionnaire.
C7CPTM0
child direct assessment or student questionnaire data from spring-eighth grade with spring-eighth grade parent interview data and spring-eighth grade English or mathematics teacher-level data with or without child-level data from the mathematics teacher, alone or in combination with data from the spring-eighth grade school administrator questionnaire. This weight is to be used only if the analytic sample is restricted to the subset of children who were sampled to have a mathematics teacher questionnaire.
C7CPTS0
child direct assessment or student questionnaire data from spring-eighth grade with spring-eighth grade parent interview data and spring-eighth grade English or science teacher-level data with or without child-level data from the science teacher, alone or in combination with data from the spring-eighth grade school administrator questionnaire. This weight is to be used only if the analytic sample is restricted to the subset of children who were sampled to have a science teacher questionnaire.

ECLS-K Eighth-Grade Longitudinal Weights

Fifth grade - Eighth grade

Child-level weights

C67CW0
child direct assessment data from BOTH spring-fifth grade and spring-eighth grade, alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-fifth grade or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-fifth grade or spring-eighth grade school administrator questionnaire, or (d) data from spring-fifth grade school facilities checklist.
C67PW0
parent interview data from BOTH spring-fifth grade or spring-eighth grade, alone or in combination with (a) spring-fifth grade or spring-eighth grade child assessment data, (b) data from any spring-fifth grade or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-fifth grade or spring-eighth grade school administrator questionnaire, or (d) data from spring-fifth grade school facilities checklist.

Third grade - Eighth grade

Child-level weights

C567CW0
child direct assessment data from THREE rounds of data collection (spring-third grade, spring-fifth grade and spring-eighth grade) alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) school administrator questionnaire data from any of these three rounds, or (d) data from any spring-third grade or spring-fifth grade school facilities checklist.
C567PW0
parent interview data from THREE rounds of data collection (spring-third grade, spring-fifth grade and spring-eighth grade), alone or in combination with (a) child assessment data from any of these three rounds, (b) data from any spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-third grade or spring-fifth grade school facilities checklist.

Spring First grade - Eighth grade

Child-level weights

C4_7CW0
child direct assessment data from FOUR rounds of data collection (spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade) alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.
C4_7PW0
parent interview data from FOUR rounds of data collection (spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth), alone or in combination with (a) child assessment data from any of these four rounds, (b) data from any spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.

Spring Kindergarten - Eighth grade (no Fall First grade)

Child-level weights

C2_7FC0
child direct assessment data from FIVE rounds of data collection (spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade) alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.
C2_7FP0
parent interview data from FIVE rounds of data collection (spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade), alone or in combination with (a) child assessment data from any of these five rounds, (b) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.

Fall Kindergarten - Eighth grade (No Fall First grade)

Child-level weights

C1_7FC0
child direct assessment data from SIX rounds of data collection (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade) alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.
C1_7FP0
parent interview data from SIX rounds of data collection (fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade), alone or in combination with (a) child assessment data from these any of these six rounds, (b) data from any fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.

Fall Kindergarten - Eighth grade

Child-level weights

C1_7SC0
child direct assessment data from ALL SEVEN rounds of data collection (fall-kindergarten, spring-kindergarten, fall-first grade, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade) alone or in combination with (a) a limited set of child characteristics (e.g., age, sex, and race/ethnicity), (b) data from any fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.
C1_7SP0
parent interview data from ALL SEVEN rounds of data collection (fall-kindergarten, spring-kindergarten, fall-first grade, spring-first grade, spring-third grade, spring-fifth grade, and spring-eighth grade), alone or in combination with (a) child assessment data from any of these seven rounds, (b) data from any fall-kindergarten, spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade teacher questionnaire (teacher-level or child-level), (c) data from any spring-kindergarten, spring-first grade, spring-third grade, spring-fifth grade, or spring-eighth grade school administrator questionnaire, or (d) data from any spring-kindergarten, spring-first grade, spring-third grade, or spring-fifth grade school facilities checklist.