Mean Scores of Different Ethnic Groups</p>
<p>Asian Americans. In the years since the Second World War, Asian Americans, especially those of Chinese and Japanese extraction, have compiled an outstanding record of academic and professional achievement. This record is reflected in school grades, in scores on content-oriented achievement tests like the SAT and GRE, and especially in the disproportionate representation of Asian Americans in many sciences and professions. Although it is often supposed that these achievements reflect correspondingly high intelligence test scores, this is not the case. In more than a dozen studies from the 1960s and 1970s analyzed by Flynn (1991), the mean IQs of Japanese- and Chinese American children were always around 97 or 98; none was over 100. Even Lynn (1993), who argues for a slightly higher figure concedes that the achievements of these Asian Americans far outstrip what might have been expected on the basis of their test scores.</p>
<p>It may be worth noting that the interpretation of test scores obtained by Asians in Asia has been controversial in its own right. Lynn (1982) reported a mean Japanese IQ of 111, Flynn (1991) estimated it to be between 101 and 105; Stevenson et al (1985), comparing the intelligence-test performance of children in Japan, Taiwan and the United States, found no substantive differences at all. Given the general problems of cross-cultural comparison, there is no reason to expect precision or stability in such estimates. Nevertheless some interest attaches to these particular comparisons: they show that the well-established differences in school achievement among the same three groups (Chinese and Japanese children are much better at math than American children) do not simply reflect differences in psychometric intelligence. Stevenson et a1(1986) suggest that they result from structural differences in the schools of the three nations as well as from varying cultural attitudes toward learning itself. It is also possible that spatial ability, in which Japanese and Chinese obtain somewhat higher scores than Americans, plays a particular role in the learning of mathematics.</p>
<p>One interesting way to assess the achievements of Chinese- and Japanese-Americans is to reverse the usual direction of prediction. Data from the 1980 census shows that the proportion of Chinese Americans employed in managerial, professional, or technical occupations was 55% and that of Japanese was 46%. (For whites, the corresponding figure was 34%.) Using the well-established correlation between intelligence test scores and occupational level, Flynn (1991, p.99) calculated the mean IQ that a hypothetical White group “would have to have” to predict the same proportions of upper-level employment. He found that the occupational success of these Chinese Americans, whose mean IQ was in fact slightly below 100, was what would be expected of a White group with an IQ of almost 120! A similar calculation for Japanese-Americans shows that their level of achievement matched that of Whites averaging 110. These “over-achievements” serve as sharp reminders of the limitations of IQ-based prediction. Various aspects of Chinese-American and Japanese American culture surely contribute to them (Schneider, Hieshima, Lee & Plank, 1994); gene-based temperamental factors could conceivably be playing a role as well (Freedman & Freedman, 1969).</p>
<p>Hispanic Americans. Hispanic immigrants have come to America from many countries. In 1993, the largest Latino groups in the continental United States were Mexican Americans (64%), Puerto Ricans (11%), Central and South Americans (13%), and Cubans (5%) (U.S. Bureau of the Census, 1994). There are very substantial cultural differences among these nationality groups, as well as differences in academic achievement (Duran, 1983; USNCEP, 1982). Taken together, Latinos make up the second largest and the fastest-growing minority group in America (Davis, Haub & Willette, 1983; Eyde, 1992).</p>
<p>The mean intelligence test scores of Hispanics typically lie between those of Blacks and Whites. There are also differences in the patterning of scores across different abilities and subtests (Hennessy & Merrifield, 1978; Lesser, Fifer & Clark, 1965). Linguistic factors play a particularly important role for Hispanic Americans, who may know relatively little English. (By one estimate, 25% of Puerto Ricans and Mexican Americans and at least 40% of Cubans speak English 'not well" or 'not at all"—Rodriguez, 1992). Even those who describe themselves as bilingual may be at a disadvantage if Spanish was their first and best-learned language. It is not surprising that Latino children typically score higher on the performance than on the verbal subtests of the English-based WISC-R (Kaufman, 1994). Nevertheless, the predictive validity of Latino test scores is not negligible. In young children, the WISC-R has reasonably high correlations with school achievement measures (McShane & Cook, 1985). For high school students of moderate to high English proficiency, standard aptitude tests predict first-year college grades about as well as they do for non Hispanic Whites (Pennock-Roman, 1992).</p>
<p>Native Americans. There are a great many culturally distinct North American Indian tribes (Driver, 1969), speaking some 200 different languages (Leap, 1981). Many Native Americans live on reservations, which themselves represent a great variety of ecological and cultural settings. Many others presently live in metropolitan areas (Brandt, 1984). Although few generalizations can be appropriate across so wide a range, two or three points seem fairly well established. The first is a specific relation between ecology and cognition: the Inuit (Eskimo) and other groups that live in the arctic tend to have particularly high visual-spatial skills. (For a review see McShane & Berry, 1988.) Moreover, there seem to be no substantial sex differences in those skills (Berry, 1974). It seems likely that this represents an adaptation—genetic or learned or both—to the difficult hunting, traveling and living conditions that characterize the arctic environment.</p>
<p>On the average Indian children obtain relatively low scores on tests of verbal intelligence, which are often administered in school settings. The result is a performance-test/verbal-test discrepancy similar to that exhibited by Hispanic Americans and other groups whose first language is generally not English. Moreover, many Indian children suffer from chronic middle-ear infection (otitis media), which is “the leading identifiable disease among Indians since record-keeping began in 1962” (McShane & Plas, 1984b, p.84). Hearing loss can have marked negative effects on verbal test performance (McShane & Plas, 1984a).</p>
<p>African Americans. The relatively low mean of the distribution of African-American intelligence test scores has been discussed for many years. Although studies using different tests and samples yield a range of results, the Black mean is typically about one standard deviation (about 15 points) below that of Whites (Loehlin et al, 1975; Jensen, 1980; Reynolds et al, 1987). The difference is largest on those tests (verbal or non-verbal) that best represent the general intelligence factor g (Jensen, 1985). It is possible, however, that this differential is diminishing. In the most recent re-standardization of the Stanford-Binet test, the Black/White differential was 13 points for younger children and 10 points for older children (Thorndike et al, 1986). In several other studies of children since 1980, the Black mean has consistently been over 90 and the differential has been in single digits (Vincent, 1991). Larger and more definitive studies are needed before this trend can be regarded as established.</p>
<p>Another reason to think the IQ mean might be changing is that the Black/ White differential in achievement scores has diminished substantially in the last few years. Consider, for example, the mathematics achievement of five year olds as measured by the National Assessment of Educational Progress (NAEP). The differential between Black and White scores, about 1.1 standard deviations as recently as 1978, had shrunk to .65 SD by 1990 (Grissmer et al, 1994) because of Black gains. Hispanics showed similar but smaller gains; there was little change in the scores of Whites. Other assessments of school achievement also show substantial recent gains in the performance of minority children.</p>
<p>In their own analysis of these gains, Grissmer et al (1994) cite both demographic factors and the effects of public policy. They found the level of parents’ education to be a particularly good predictor of children’s’ school achievement; that level increased for all groups between 1970 and 1990, but most sharply for Blacks. Family size was another good predictor (children from smaller families tend to achieve higher scores); here too, the largest change over time was among Blacks. Above and beyond these demographic effects, Grissmer et al believe that some of the gains can be attributed to the many specific programs, geared to the education of minority children, that were implemented during that period.</p>
<p>Test Bias. It is often argued that the lower mean scores of African Americans reflect a bias in the intelligence tests themselves. This argument is right in one sense of “bias” but wrong in another. To see the first of these, consider how the term is used in probability theory. When a coin comes up heads consistently for any reason it is said to be 'biased," regardless of any consequences that the outcome may or may"not have. In this sense the Black/White score differential is ipso facto evidence of what may be called “outcome bias.” African Americans are subject to outcome bias not only with respect to tests but along many dimensions of American life. They have the short end of nearly every stick: average income, representation in high-level occupations, health and health care, death rate, confrontations with the legal system, and so on. With this situation in mind, some critics regard the test score differential as just another example of a pervasive outcome bias that characterizes our society as a whole (Jackson, 1975; Mercer, 1984). Although there is a sense in which they are right, this critique ignores the particular social purpose that tests are designed to serve.</p>
<p>From an educational point of view, the chief function of mental tests is as predictors (Section 2). Intelligence tests predict school performance fairly well, at least in American schools as they are now constituted. Similarly, achievement tests are fairly good predictors of performance in college and postgraduate settings. Considered in this light, the relevant question is whether the tests have a “predictive bias” against Blacks, Such a bias would exist if African-American performance on the criterion variables (school achievement, college GPA, etc.) were systematically higher than the same subjects’ test scores would predict. This is not the case. The actual regression lines (which show the mean criterion performance for individuals who got various scores on the predictor) for Blacks do not lie above those for Whites; there is even a slight tendency in the other direction (Jensen, 1980; Reynolds &:Brown, 1984). Considered as predictors of future performance, the tests do not seem to be biased against African Americans.</p>
<p>Characteristics of Tests. It has been suggested that various aspects of the way tests are formulated and administered may put African Americans at an disadvantage. The language of testing is a standard form of English with which some Blacks may not be familiar; specific vocabulary items are often unfamiliar to Black children; the tests are often given by White examiners rather than by more familiar Black teachers; African Americans may not be motivated to work hard on tests that so clearly reflect White values; the time demands of some tests may be alien to Black culture. (Similar suggestions have been made in connection with the test performance of Hispanic Americans, e.g., Rodriguez, 1992.) Many of these suggestions are plausible, and such mechanisms may play a role in particular cases. Controlled studies have shown, however, that none of them contributes substantially to the Black/White differential under discussion here (Jensen, 1980; Reynolds 82 Brown, 1984; for a different view see Helms, 1992). Moreover, efforts to devise reliable and valid tests that would minimize disadvantages of this kind have been unsuccessful.