Sources of Data

The study design for this project requires the use of large-scale panel studies that are representative of known populations, with a minimum of three waves of measurement separated by two-year re-interview intervals. Questions were selected for use only if they were exactly replicated (exact wording, response categories, mode of interviewing, etc.) across the three waves, and if the underlying variable measured was continuous (rather than categoric) in nature. Specifically, this research is based on ten nationally (or regionally) representative panel surveys of the American population, all involving probability samples and, with one exception, all using face-to-face interviews, as shown in the table below. These selection criteria applied to these ten panel studies yielded more than 1,300 self- and proxy-report questions that are typical of the kinds of questions employed in contemporary surveys.

Table 1. Sources of data for question-specific estimates of reliability
Sample Number of
Panel studies Acronym Size Measures
Totals 37,786 1357
1956-58-60 National Election Study Panel NES60s 2,529 47
1972-74-76 National Election Study Panel NES70s 4,455 122
1992-94-96 National Election Study Panel NES90s 2,439 114
Americans' Changing Lives Panel ACL 3,617 98
Study of American Families-Mother panel SAF-Mo 1,113 56
Study of American Families-Child panel SAF-Ch 1,113 46
1998-00-02 Health and Retirement Study HRS 18,645 154
2006-08-10 General Social Survey Panel GSS06 1,276 249
2008-10-12 General Social Survey Panel GSS08 1,295 238
2010-12-14 General Social Survey Panel GSS10 1,304 233

More information