National Household Education Surveys Program, 2012

An initial sample of 208,000 addresses was selected, of which 159,994 were designated for the NHES:2012. The initial sample of addresses was drawn from a file of residential addresses maintained by a vendor, Marketing Systems Group (MSG), based on the United States Postal Service (USPS) Computerized Delivery Sequence File (CDSF). MSG also provided the sample for the NHES Pilot and Field Tests in 2009 and 2011.

The NHES:2012 sample is a two-stage, stratified sample. The first sampling stage selected residential addresses from the MSG file, and the second sampling stage selected an eligible child from information provided on the household mail screener. Households and children were selected with differential probabilities of selection based on the Black and Hispanic composition of the Census tract where an address is located, residential address type, and children’s survey eligibility within the household (ECPP or PFI). These differential probabilities of selection are accounted for in the NHES weighting methodology. When these weights are applied to the ECPP survey, it is nationally representative of all children from birth through age 6 who are not enrolled in kindergarten. When these weights are applied to the PFI survey, it is nationally representative of students enrolled in grades K-12, including children who are enrolled in public school, private school, and those who are homeschooled for the equivalent grades.

Survey (self- or interviewer-administered)

Irregular

Data were collected using a two-phase survey (conducted primarily by mail), which used an address-based sample covering the 50 states and the District of Columbia. As described in detail in the user's manual, households were randomly sampled, and a screening questionnaire was sent to each sampled household. Households that were eligible for further surveying received a longer, topical survey. Additional information about data collection methods can be found here: https://nces.ed.gov/pubs2018/2018100.pdf

The ECPP survey file contains data from surveys completed with the parents or guardians of 5,844 children age 6 or younger not yet enrolled in kindergarten. The PFI survey file contains data from surveys completed with the parents or guardians of 14,075 children age 20 or younger in kindergarten through 12th grade, including 13,523 students whose parents completed the PFI–Enrolled questionnaire for students enrolled in public or private school and 552 students whose parents completed the PFI questionnaire for homeschooled students. The ATES file contains data from surveys completed with 47,744 noninstitutionalized adults ages 16–65, not enrolled in grade 12 or below.

The ECPP file contains 642 variables. The PFI file contains 822 variables. The ATES file contains 354 variables.

Only Restricted-Use NHES data has variables which allow for linkages. The geographic identifiers within NHES permit data users to merge on any data available at the state, CBSA, county, NECTA, ZCTA, census tract, or census block group-level, such as population characteristics from the ACS. While this is possible it should be noted that the NHES is not designed to create sub-national estimates. Additional information about data linkage capabilities can be found here: https://nces.ed.gov/pubs2018/2018100.pdf

• Other (see Linkage Capabilities description)