|Sponsor:||U.S. Department of Education (DoE)/National Center for Education Statistics (NCES)|
|Description:||The National Household Education Surveys Program (NHES) provides descriptive data on the educational activities of the U.S. population and offers researchers, educators, and policymakers a variety of statistics on the condition of education in the United States. The NHES surveys cover learning at all ages, from early childhood to school age to adulthood. The NHES uses a repeating cross-sectional design that allows for the study of trends related to educationally important topics.
While there are many surveys included in this system, this profile includes only the surveys that have more than 100 AI/AN/NA individuals across multiple years, and that were administered in 1995 or later (note that the first NHES survey was fielded in 1991). Surveys that are included in this profile are the Parent and Family Involvement in Education Survey (PFI-NHES), the Early Childhood Program Participation Survey (ECPP-NHES), the Adult Education Survey (AE-NHES) (also called the Adult Education and Lifelong Learning Survey (AELL-NHES)), the Before- and After-School Programs and Activities Survey (ASPA-NHES), and the 1999 Parent Survey (Parent-NHES, which includes some items from the PFI-NHES, ECPP-NHES, and ASPA-NHES).
|Relevant Policy Issues:||Educational Attainment, Educational Opportunities, and Factors Contributing to Educational Disparities.|
|Unit of Analysis:||Individual|
|Identification of AI/AN/NA:||Race is self-reported on all the surveys. The questions asking about race were changed between the 2003 and 2005 collections in order to meet new OMB requirements for collecting this information. Specifically, the questions were changed to allow respondents to be classified as more than one specific race (e.g., as both White and Black). In previous years (i.e., 1995-2003), respondents were classified as either belonging to only one racial group or as being multiracial without those specific races being identified. For example, the 2005 Adult Education Survey included the following race question:
Which of the following races do you consider yourself to be? You may name more than one.
In 2001, the AE-NHES race questions was:
The race question in all surveys prior to 2005 was structured the same way as this question from the 2001 AE-NHES Survey. Note that also prior to 2005, the categories Asian and Native Hawaiian or other Pacific Islander were combined into a single category: Asian or Other Pacific Islander.
For all NHES surveys, information about Hispanic ethnicity is collected in a separate question.
|AI/AN/NA Population in Data Set:||PFI-NHES
1996: Out of 20,792 records, 231 are AI/AN
2003: Out of 12,426 records, 108 are AI/AN
AE-NHES and AELL-NHES
|Geographic Scope:||The geographic scope of NHES is national. While zip code was collected on the surveys, it is not included in the public use data files. The files do contain the variable ZIPURBAN, which identifies records as being inside or outside an urban region. Additionally, records contain a Census region with the four values: Northeast, South, Midwest, West.|
|Date or Frequency:||The PFI-NHES survey was conducted in 1996 and 2003. The ECPP-NHES survey was conducted in 1995, 2001, and 2005. The AE-NHES survey was conducted in 1995, 1999, 2001 and 2005. In 2001, the AE-NHES survey was called the Adult Education and Lifelong Learning Survey (AELL-NHES). The ASPA-NHES survey was conducted in 2001 and 2005. The Parent-NHES was conducted in 1999. For all years of administration, the NHES data collection period runs from January to April.
The NHES will continue to be conducted regularly in the future, covering the same topics investigated in earlier collections. In 2007, surveys will cover Adult Education for Work-Related Reasons, School Readiness, and Parent and Family Involvement in Education.
|Aggregation:||It is possible to combine multiple years of data for the NHES surveys, but researchers should be aware that there is a sizable time difference between the administrations of some of the surveys (for example, the PFI-NHES survey was conducted in 1996 and 2003). In addition, in 1995, the NHES surveys switched the way in which they selected their sample (from using a modified Mitofsky-Waksberg method to a list-assisted method). For this reason, it is recommended that the samples from NHES surveys conducted prior to 1995 not be combined with samples later than 1995. As the NHES is weighted data, new weights would need to be developed in order to combine survey data across multiple years.
Researchers also should review any survey questions of interest before combining them to ensure the question text and response categories are comparable. Change in the race question text between 2003 and 2005 may affect the ability to aggregate 2005 data with data from previous years.
If researchers choose to combine data sets across multiple years, the end results would be:
|Data Collection Methodology:||NHES is conducted as a random digit dial (RDD) telephone survey using computer-assisted telephone interviewing (CATI). Each household contact begins with a screener to obtain information used to sample adults and children for extended interviews (the topical surveys). Since 1996, in order to introduce the survey, advance letters have been sent out to all sample households where an address was obtained through a commercial address matching service.|
|Participation:||Optional, with incentives|
|Response Rate:||Response rates:
1996 PFI-NHES: Weighted response rate of 62.5%
2003 PFI-NHES: Weighted response rate of 53.8%
1995 ECPP-NHES: Weighted response rate of 66.3%
1995 AE-NHES: Weighted response rate of 58.6%
2001 ASPA-NHES: Weighted response rate of 59.7%
1999 Parent-NHES: Weighted response rate of 66.7%
The NHES data user manuals provide weighted response rates as these rates give a better description of the success of the survey with respect to the population of interest. The response rate indicates the percentage of possible interviews that have been completed, taking all sampling stages into account. The weighted response rate is similar to the unweighted response rate unless the probabilities of selection vary considerably.
|Sampling Methodology:||Beginning in 1995, the NHES surveys began using a list-assisted method to select the random digit dial (RDD) sample. With the list-assisted method, an equal probability random sample of telephone numbers is selected from all telephone numbers that are in 100-banks (numbers in a 100-bank have the same first 8 digits of the 10-digit telephone number) in which there is at least one residential telephone number listed in the white pages directory (the listed stratum). Both listed and unlisted telephone numbers are included in the listed stratum. Telephone numbers in 100-banks with no listed telephone numbers (the zero-listed stratum) were not sampled.
Additionally, a within-household sampling scheme was developed to limit the number of persons sampled for extended interviews in each household in order to reduce respondent burden.
|Analysis:||Design effects (deff):
1996 PFI-NHES survey deff = 1.3 for the Full Sample
2003 PFI-NHES survey deffs = 1.3 for the Full Sample, and 1.4 for the race/ethnicity subgroups
1995 ECPP-NHES survey deff = 1.2 for the Full Sample
1995 AE-NHES survey deff = 1.3 for the Full Sample
2001 ASPA-NHES survey deff = 1.3 for the Full Sample
1999 Parent-NHES survey deff = 1.3 for the Full Sample
|Strengths:||Multiple years of data are available. Documentation is detailed and comprehensive.|
|Limitations:||Most of these studies contain a relatively small number of AI/AN/NA respondents. While it may be possible to combine the samples from multiple years of the surveys, this is a complicated procedure. One of the biggest issues related to aggregation is the variation in the race questions over the years.|
|Access Requirements and Use Restrictions:||Data are available online at no cost.|
|Contact Information:||The NHES data can be downloaded via the Internet at: http://nces.ed.gov/nhes/dataproducts.asp
The NHES staff can be contacted by sending an email to: firstname.lastname@example.org.
"report.pdf" (pdf, 1.56Mb)
"apa.pdf" (pdf, 107.76Kb)
"apb.pdf" (pdf, 88.97Kb)
"apc.pdf" (pdf, 233.07Kb)