What the opposition suggests in this position is a study that would be so enormous as to be impossible.  Acquiring independent verification from the thousands of people interviewed would be very difficult, at best, and probably impossible.  In those cases of persons identified as having physical or mental impairment, it would require an independent medical evaluation of the selected sample in order to resolve the issue of “lack of exogeneity.”  All survey research of a macro nature lacks exogeneity, but the large sample size reduces, if not eliminates, the supposed problem regarding exogeneity.  Stern (1989) tests for this problem by measuring labor force participation using self-reported disability.  He finds that any potential bias is small and that “the standard disability measures are powerful and reasonably exogenous predictors of labor force participation” (p. 392). 

The CPS relies upon answers from respondents to questions administered by trained Census personnel (self-reporting). As such, the criteria used by the Census Bureau to classify a respondent’s disability status depends upon

·                    the respondent’s ability to recognize the disability, and

·                    the truthfulness of the response

Critics speculate that one or both of these requirements are not met in enough cases as to skew the results.  The Current Population Survey is the primary source of employment data for the United States.  The entire survey is self-reported, or lacking exogeneity.  Despite this, it is relied upon by researchers, economists, demographers, and other scientists across the world for measurements of employment, earnings, education status, age, and other characteristics of the U.S. economy. 

The government does not require a CPA to verify the income reported, employers to verify employment status, or birth certificates to verify age.  Yet, vocational and economic experts and social scientists in general routinely rely upon the resulting income, employment, and age statistics both in forensic and nonforensic settings.  However, when it comes to the CPS question on whether the respondents have any physical or mental limitation in the kind or amount of work they can do, challengers contend that respondents are incapable of or unwilling to give an accurate response.  As with the heterogeneity issue, this is a nonsensical double standard.  Those who contend that lack of exogeneity is a problem themselves routinely use such data specific to earnings, level of education, and age.