The Cattell-Horn-Carroll (CHC) model of cognitive ability as represented in the Woodcock-Johnson Tests of Cognitive Ability-III (WJ-III) was examined for Black and White adults matched on various demographic variables. Although Whites performed higher than Blacks (i.e., race differences were found in test scores and accompanying factor means), the results of multisample confirmatory factor analyses found that the same constructs are measured in different groups. Therefore results are directly comparable, and in this sense measured differences can be interpreted as "real" differences on the dimensions that the test is meant to measure.
Keywords: Group differences; Intelligence; Bias; Group Differences in Cognitive Ability: A CHC Theory Framework.
Measures of intelligence and other cognitive abilities may constitute psychology's most important technical and conceptual contribution to the behavioral sciences (Oakland, 2009). However, their use has not been without controversy. In light of the high-stakes decisions that often accompany IQ testing (Scheuneman & Oakland, 1998), measures of intelligence when used cross-culturally and cross-nationally have received considerable scrutiny and often criticism (Carroll, 1997; Jensen, 1981; Lynn, 2001; Mensh & Mensh, 1991; Ogbu, 2002; Rushton & Jensen, 2005).
Some believe measures of intelligence are inherently biased against non-white groups in the US, especially Blacks, (Kamin, 1995; Snyderman, & Rothman, 1988; Williams, 1974) despite considerable evidence to the contrary (Jensen, 1980; Meile, 2002). Historically, adherents of this belief typically pointed to mean IQ differences between Blacks and Whites as prima facie evidence of a test's racial discrimination. These group IQ differences have been somewhat stable and substantial, typically on the order of 10 to 15 points (Jensen, 1998; Lynn, 1996; Neisser, et al., 1996).
However, current views recognize mean group differences do not necessarily provide evidence of test discrimination. Contemporary views favor the use of more sophisticated statistical techniques that examine potential sources of variance that may constitute test bias (e.g., Kane & Tangdhanakanond, 2008). As such, bias in assessment has been defined as occurring "when deficiencies in the test itself or the manner in which it is used result in different meanings for scores earned by members of different identifiable subgroups" (American Educational Research Association [AERA], American Psychological Association, & National Council on Measurement in Education, 1999, p. 74). To this end, an essential step in ensuring that a test is a fair and appropriate instrument for its intended audience is to examine the factorial invariance across racial/ethnic groups. These methods are intended to have two purposes: to inform test users as to the clarity of the theoretical constructs measured by a test (e.g. Is the CHC theory valid for both Blacks and Whites?) as well as possible test bias (e.g., Does the WJ-III show empirical evidence of psychometric bias?).
This study investigates the integrity of the CHC theory as well as factorial invariance of the Woodcock-Johnson Tests of Cognitive Abilities, Third Edition (WJ-III; Woodcock, McGrew, & Mather, 2001) for Black and White adults. Relatedly, the study attempts to identify possible sources of variance that may account for any observed race differences in the measurement of intelligence.
Sample. The total sample of 412 Black and White adults was drawn from the WJ-III standardization sample. Individuals from the two groups were matched on age, primary and first language, sex, occupational status, and parental education. Hispanics were excluded from the analyses. Thus, the sample consisted of 206 adults from each race, with 96 males and 110 females in each group. Average ages for the Black and White samples were 27.1 and 27.9 years, respectively.