ECLS-B restricted-use data files are released approximately two years after data collection ends for each wave (e.g., 9-months, 2-years).
Due to NCES' confidentiality legislation, ECLS-B data in micro-data form (i.e., CD-ROM) are available only through a restricted-use data license agreement. When presenting, preparing manuscripts, or publishing ECLS-B results, analysts must comply with ECLS-B rounding rules. Unweighted sample sizes must be rounded to the nearest 50. For example, a cell size of 25 to 74 is rounded to 50, 75 to 124 is rounded to 100, and a cell size less than 25 is denoted by a symbol indicating "rounds to zero." Further, all presentations and manuscripts prepared using ECLS-B restricted-use data must be sent to the NCES Data Security Office (IESData.Security@ed.gov) for disclosure review, as is required by the NCES restricted-use data license terms.
The ECLS-B 9-month data are now available for analysis to the general public in the Data Analysis System (DAS). The DAS, which can be accessed at http://nces.ed.gov/das/, is a software application that allows users to build tables of weighted estimates from the ECLS-B variables included in the DAS. An on-line tool allows users to calculate t-tests from estimates produced in the DAS. Users can also produce correlation matrices for use in linear regression analyses. The DAS webpage has links to an on-line tutorial and user guide to help users navigate through the application and produce desired tables.
The ECLS-B DAS application currently contains 9-month variables that are available in the 9-month-2-year restricted-use data file (NCES 2006-044). Data from later waves of the ECLS-B (i.e., 2-year, preschool, and kindergarten collections) will be added in the future. Some variables that are not analytically useful in the context of the DAS have been omitted. Users are strongly encouraged to review the description screens for the variables they are interested in examining, because this screen has important information about the variables, including: an explanation of the variable, the questionnaire item to which the variable pertains, the respondents who were asked the question, the variable’s range of values, and weighted frequencies for each value. For composite variables, this screen also contains detailed information about how the variable was created.
In addition to reviewing the variable description screens, users are strongly encouraged to use the DAS in conjunction with the ECLS-B questionnaires, which can be accessed here: http://nces.ed.gov/ecls/birthinstruments.asp. Variables appear in the DAS in questionnaire order (after the composite variables), by questionnaire section. Reviewing the questionnaires will help users better understand the flow of the interview, the groups of respondents who answered specific questions, and the context in which specific questions were asked. This is particularly important as some of the skip patterns are complex. The label for each variable begins with the questionnaire item number, which makes it easy for users to locate the question to which each variable pertains.
A handful of variables have been transformed to be compatible with the reporting capabilities of the DAS. For example, the values for the variable indicating child‘s assessment age (in months), which are reported to one decimal place in the restricted-use data file, were multiplied by 10 to be reported as whole numbers in the DAS. All such transformations are thoroughly described in the variable‘s description screen. Please note that as a result of these transformations, estimates produced from the DAS will differ from estimates in published NCES reports and elsewhere that are produced from the restricted-use file.
9 months of age: The ECLS-B assesses children and interviews primary caregivers (usually the mother) in the children's homes. Fathers complete a self-administered questionnaire. The in-home visit includes a videotaping of parent-child interaction. Additionally, information regarding prenatal care and delivery is obtained from the National Center for Health Statistics.
2 years of age:The ECLS-B assesses children and interviews primary caregivers in the children's homes. Fathers complete a self-administered questionnaire. The in-home visit includes a videotaping of parent-child interaction. Child care providers are interviewed, and a subsample of child care settings are observed.
4 years of age/preschool: The ECLS-B assesses children and interviews primary caregivers in their children's homes. Father's complete a self-administered questionnaire. The in-home visit includes a videotaping of parent-child interaction. Child care providers are interviewed, and a subsample of child care settings are observed and rated.
Kindergarten: The ECLS-B assesses children and questions parents, early care and education providers, and teachers in two data collection waves. Data are collected for all sample children in the Fall of 2006, when approximately 75% are estimated to be age-eligible for kindergarten. In the Fall of 2007, the remainder of the sample enters kindergarten; these children are assessed and an abbreviated primary caregiver interview is administered in this second kindergarten collection. School data are drawn from existing databases supported by NCES [e.g., the Common Core of Data, the Private School Survey, etc.].
ECLS-B restricted-use data are released in CD-ROM format. The CD-ROMs are structured and formatted much like other NCES surveys (e.g., Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 and the National Household Education Survey). They include the raw data files, electronic codebooks (ECBs), user manuals, the survey instruments, and the record layouts. The ECBs can be used to 1) examine variables in a dataset, 2) examine question wording, response categories and frequency distributions, and 3) select and export variables along with the appropriate code needed to create SAS, SPSS-PC and STATA datasets.
ECLS-B data available through the DAS are structured in a similar fashion as data available on the restricted-use files. The DAS software application allows users to conduct statistical analyses on-line, without directly accessing the data file.
