In an otherwise interesting and enlightening article, Borghans, Golsteyn, Heckman, and Humphries (2011) analyzed evidence from the 1979 National Longitudinal Survey of Youth (NLSY79) to support their contention that "achievement" tests have greater power than "IQ" tests in predicting "a variety of life outcomes". A key point in their argument is their contention that scores on the Armed Forces Qualifications Test (AFQT) represent "achievement" scores and that the AFQT is qualitatively different from purported true "IQ" score data also available in the NLSY79. This contention is based on both conceptual argument and empirical analysis of NLSY79 data. This comment disputes their contention on both grounds. First, it argues that their conceptual distinction is contradicted in the educational testing literature and is based on erroneous assumptions about the nature of the purported true "IQ" test data in the NLSY79. Second, it presents evidence that their empirical findings flow from problems in true "IQ" score imputation and large gaps in calendar time between the purported true "IQ" tests and AFQT and personality test data in the NLSY79 data set.
- IQ tests
ASJC Scopus subject areas