What populations of children can be studied with the ECLS-B data?
The ECLS B describes children born in the United States in 2001, with the exception of children born to mothers younger than age 15 and children who died or were adopted prior to the 9-month data collection. Additionally, the ECLS-B sample is large enough to support the analysis of many different subgroups of children. Because the ECLS-B oversampled certain groups of children that are relatively rare in the general population (twins; Chinese, Other Asian and Pacific Islander, and American Indian/Native Alaska children; and children born with moderately low or very low birth weight), reliable estimates can be produced for these groups of children.
Do I need to use weights for my analyses? What if I am not interested in making statements about the population?
Weights are used to adjust for disproportionate sampling, survey nonresponse, and undercoverage of the target population when analyzing complex survey data. They also are necessary to produce national-level estimates of the children born in 2001.
Oversampling results in certain groups of children being represented as a larger proportion of the sample than their representation in the general population. Estimates produced in analyses that do not adjust for this oversampling may be biased if the characteristics of the oversampled groups are related to the outcomes being studied. For example, the average birth weight estimate will be lower in the ECLS B sample than in the population in general due to the oversample of twins and children born with low and very-low birth weight in the sample. Using the weights will correct for the over-representation of these groups and produce a more accurate estimate of birth weight among babies born in the United States in 2001.
Do I need to use the same weight for all analyses?
Researchers are encouraged to use the same weight throughout all analyses in a publication or paper, even when there is a different ideal weight for each analysis. Weights are assigned to cases with valid data for the component(s) contributing to the weight. Selecting different weights within the same publication or paper results in each analysis being run with a different analytic sample (i.e., the exact cases contributing to the analyses).
Can researchers produce state-level estimates?
No, the ECLS-B sample is designed to support national and regional estimates. It is not designed to estimate characteristics of children, families, and schools at or below the state level.
What statements can be made about participation in early child care and education programs, and its relation to cognitive, physical, and socioemotional development?
The ECLS-B data on child care and early education can be used to examine nonparental care and education experiences of the children born in 2001, both at a particular point in time (e.g., when they are 9 months old) and longitudinally. For example, it is possible to compare children’s nonparental care and early childhood education experiences before kindergarten to their experiences in before- and/or after-school programs and activities once they enter kindergarten.
All statements about nonparental child care and early education should be made in relation to the experiences of children born in the United States in 2001. The ECLS-B data cannot be used to make statements about nonparental child care and early education in the United States generally or about the population of providers. For example, the data cannot be used to generate estimates of the number of providers in the United States who provide different types of care and early education (i.e., center-based, relative care, or nonrelative care). Because early care and education providers were identified through their link to the ECLS-B sample of children born in 2001, as opposed to being identified through a random sample of providers from a universal list of providers, the sample of providers is not nationally representative.
To study the relationship between early care and education and children's development, it is important to have data both before children receive care and education from persons other than parents and after they begin receiving nonparental care and education. The ECLS-B captures information on children's participation longitudinally, making it possible to compare differences in key child development outcomes before and after experiencing nonparental care and education.
Which data file should I use if I want to analyze data from just one round of collection?
The longitudinal 9-month—kindergarten 2007 restricted-use data file can be used for any analysis. This longitudinal file contains data for all cases that ever participated in the study, including those that became nonrespondents at some point after the 9-month collection. (NOTE: The data file does not contain information about the originally sampled cases that never participated in the study.) This final longitudinal data file includes some important updates and corrections to errors discovered in previously released data; for this reason, researchers with previously released data files (e.g., the 9-month-—2-year restricted-use file) are strongly encouraged to obtain the most recent release of the data.