|Sponsor:||U.S. Department of Education (DoE)/National Center for Educational Statistics (NCES)|
|Description:||The Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K) is an ongoing study that focuses on childrens early school experiences beginning with kindergarten and following children through 12th grade. In the fall of 1998, ECLS-K began following a nationally representative sample of kindergarteners. The ECLS-K provides descriptive information on childrens status at entry to school, their transition into school, and their progression through 8th grade. (Initially, the ECLS-K was designed to follow children through their fifth grade year in school however plans have been made to extend the study to follow the ECLS-K children through their eighth grade of school. The study will end with the data collection scheduled for school year 2006-2007.)
The longitudinal nature of the ECLS-K data enables researchers to study how a wide range of family, school, community, and individual factors are associated with school performance. Researchers can request the child-level files for each year of data collection, as well as the longitudinal kindergarten to fifth grade data file. Data are collected from a direct child assessment, from parent interviews, from school administrators and teachers, and from student records and a school facilities checklist.
|Relevant Policy Issues:||Educational Attainment, Educational Opportunities, Factors Contributing to Educational Disparities, and Identification of Evidence-based Practices and Programs that Produce Positive Educational Outcomes and are Generalizable/replicable.|
|Unit of Analysis:||Individual child|
|Identification of AI/AN/NA:||Race information is obtained during the parent interviews using the following question: What is your race? (The same question is asked concerning the child: What is [NAME OF CHILDS] race?)
When the parent interview is not completed, race information is obtained from school records. The parent interview is considered the best source of information regarding race/ethnicity.
|AI/AN/NA Population in Data Set:||Completed interviews with AI/AN/NA children, parents, and school informants by type of data collection for the school year 2003-2004 (fifth-grade):
School administrator questionnaires
School facilities checklist
Student records abstract
|Geographic Scope:||The geographic scope of the study is national. The ECLS-K public-use data also contain information on the regional location of the childs school (i.e., Northeast, Midwest, South, and West.) The ECLS-K sample was designed to support national and regional estimates. It was not designed to estimate characteristics of children, teachers, families and schools at the state level. Variables such as the childs home and school zip code are suppressed on the public use files to ensure respondent confidentiality.|
|Date or Frequency:||The ECLS-K is a longitudinal study. The same children are followed periodically from kindergarten through the 8th grade. Information was collected in the fall and the spring of kindergarten (1998-99), the fall and spring of first grade (1999-2000), the spring of third grade (2002) and the spring of 5th (2004). Future data collections will include 8th grade (2007).|
|Data Collection Methodology:||To collect information from children, a trained assessor visits the children in their schools. Children are assessed, un-timed, one-on-one within their school. The direct child assessment collects information about childrens reading and mathematics skills, their general knowledge (i.e., science and social studies) in kindergarten and first grade, and their science knowledge in third and fifth grade. In addition, the assessment includes height/weight measurements, and in fall kindergarten only, childrens psychomotor skills (e.g., hopping, skipping, jumping, manipulating blocks, drawing figures) are assessed. The direct child assessments are administered using computer assisted personal interviews (CAPI).
To collect information from parents, a trained interviewer phones the parent at their home and administers a 45-50 minute interview. Computer assisted telephone interviewing (CATI) methods are used to record the parents answers. If the childs family does not have a telephone, the interview is conducted in person.
To collect information from schools, teachers and school administrators complete paper and pencil surveys and retrieve information from school records.
|Participation:||Optional, with incentives|
|Response Rate:||Overall unweighted response rates for the fifth grade cohort (school year 2003-2004):
Child assessments: 93.6%
Parent interview: 90.7%
School administrator questionnaire: 89.6%
Facilities checklist: 91.4%
Student records abstract: 82.1%
Teacher level-questionnaire: 90.6%
There are also some differences in response rates by race. For example, in the spring-fifth grade data collection, the AI/ANs had the lowest child assessment weighted completion rate (78.3 percent) and the highest parent interview weighted completion rate (95.2 percent).
|Sampling Methodology:||The ECLS-K study uses a multistage probability sample design. Primary sampling units (PSUs) consist of counties and groups of counties. The second stage units are schools within the sampled PSUs. In the base year of the ECLS-K, a total of 1,277 schools were selected for the sample. The third and final stage is students within the selected schools. In the base year of the ECLS-K, a total of 22,666 students were selected. The children in ECLS-K attended both public and private schools, including both full-day and part-day kindergarten programs.|
|Oversample of AI/AN/NA Population:||Asians and Pacific Islanders were oversampled for the ECLS-K study. To create this oversample, two independent sampling strata were formed within each school. One stratum consisted of Asian/Pacific Islander children, while the second stratum consisted of the remaining children.|
|Analysis:||The ECLS-K data are weighted to compensate for differential probabilities of selection at each sampling stage and to adjust for the effects of nonresponse. The Users Manual provides a detailed description of how weights were calculated and how they should be applied to the data set.
Each survey item in the ECLS-K has its own design effect that can be estimated from the survey data. The median design effects, as reported in the ECLS-K 5th Grade Users Manual, for the race/ethnicity variable for all six rounds of data collection are:
|Strengths:||Data are collected on a key policy issue, education. There are multiple years of data available. The study includes some oversampling of Pacific Islanders.|
|Limitations:||There are a limited number of AI/AN/NA in these data sets.|
|Other:||While the ECLS-K study includes children that attend Bureau of Indian Affairs (BIA) schools, the number of children attending these schools is suppressed to protect respondent confidentiality. Additionally, attendance at BIA schools is not included as an identifier in the data files.|
|Access Requirements and Use Restrictions:||Unlike the Early Childhood Longitudinal Study Birth Cohort data, the ECLS-K data has a public use data file. There is also a restricted ECLS-K data set available to researchers with a National Center for Education Statistics (NCES) license. The restricted data set contains a few values and variables that are suppressed on the public use data set. For most research, the ECLS-K public use data set should suffice. For researchers who feel they need the restricted use data set, the steps for obtaining an NCES license are detailed here: http://nces.ed.gov/pubsearch/licenses.asp. There is no cost associated with use of the data.|
|Contact Information:||Public use data file can be ordered on CD-ROM from www.edpubs.org.|
"report.pdf" (pdf, 1.56Mb)
"apa.pdf" (pdf, 107.76Kb)
"apb.pdf" (pdf, 88.97Kb)
"apc.pdf" (pdf, 233.07Kb)