One of the most highly cited articles in intelligence research is a 1996 report commissioned by the American Psychological Association’s Board of Scientific Affairs to provide an authoritative statement on the science of intelligence (Neisser et al., 1996). What many people do not know, though, is that this was not the first time APA’s Board of Scientific Affairs had commissioned such a report.
In 1975, the organization published a report that it had commissioned on the use of psychological tests with diverse diverse groups in school settings (Cleary et al., 1975). The similarities between the 1975 and 1996 reports are unmistakable. Both were commissioned in a time of controversy, and both were published in APA’s flagship journal, American Psychologist.
It is telling to compare the reports and to examine their similarities and differences. Doing so reveals a lot about the science of intelligence (especially testing) and how authoritative reports are commissioned.
Similarities Between the Two Statements
Both reports were a response to a controversy over intelligence. The 1975 report was a response to a “manifesto” (Cleary et al.’s, 1975, label–not mine) presented to APA by the Association of Black Psychologists (ABPsi). Among other things, this manifesto called for a moratorium on the use of educational tests in schools for children from disadvantaged backgrounds (see Williams & Mitchell, 1978, for a description of this statement and other anti-testing statements from the era). The 1996 report was a response to the controversy over The Bell Curve (Herrnstein & Murray, 1994).
This is not as alarming as it sounds. It usually takes a controversy to get APA to commission an official report from its Board of Scientific Affairs. The same happened over the past 25 years or so with its reports on statistical inference, video game research, and internet-based research.
Another similarity is how the reports do not deny the differences in average IQ scores across racial and ethnic groups. In both reports, the Black-White American difference is listed as about 15 points. Also, it is interesting that the reports stress (correctly) that there is no scientific justification for applying group averages in test scores to decisions about individuals.
What these similarities show is continuity in psychologists’ understanding of intelligence in the late 20th century. This is not a field that has had major upheavels that make older research obsolete overnight.
Bias in Intelligence Tests
Both APA statements discuss intelligence tests and their uses extensively. Both state unequivocally that IQ scores are excellent predictors of academic achievement (Cleary et al., p. 23; Neisser et al., pp. 81-82). They both also discuss test bias and both dismiss the simplistic argument that average score gaps across demographic groups are evidence of test bias. Both statements are also clear that IQ scores are equally good predictors of outcomes for Black and White Americans.
The 1975 statement devotes considerably more attention to the issue of test bias (Cleary et al., 1975, pp. 25-31) because it is at the heart of the “manifesto” that the ABPsi issued. The methodology of examining what is today called “predictor bias” is laboriously explained, probably because the procedures for investigating bias were only a few years old at the time. The first author of the 1975 statement, T. Anne Cleary, was the developer of these procedures (Cleary, 1968).
The discussion of bias, though, is where both statements are obsolete. Predictor bias is not the only possible form of test bias–and it is often not the most important anyway. Neither statement mentions item bias, and the major breakthrough of measurement invariance (Meredith, 1993) is not discussed in the 1996 statement.
Although James Flynn popularized the Flynn effect (and forced psychologists to grapple with some tough questions arising from it), he never claimed to have discovered it. Indeed, the 1975 refers to the Flynn effect multiple times, though does not call it by that name.
. . . a nonexperimental approach to this problem [of creating a psychosocial environment to boost intelligence] is illustrated by the comparison of intelligence of World War I and World War II draftees (Tuddenham, 1948) and of the World War II and 1963 norms of the Air Force Classification Tests (Tupes & Shaycoft, 1964). The results are quite dramatic. The associated increase in intelligence test scores amounted to approximately one standard deviation of the World War I distribution, while subsequent to World War II the increase appeared to be about one quarter of a standard deviation.Cleary et al. (1975, p. 20).
Intelligence is not a fixed trait in a population. There are good data which show that the intellectual level of Americans, as well as the levels of other Western nations . . . has risen even during the relatively short history of the use of intelligence tests.Cleary et al. (1975, p. 23).
The 1996 statement is much more detailed and devotes seven paragraphs to the Flynn effect (Neisser et al., 1996, pp. 89-90). This section is more systematic than anything in the 1975 statement, though there is little indication in that statement that pre-Flynn psychologists were aware of IQ score inflation.
Differences Between the Two Statements
Where the reports differ most is in their coverage of intelligence research. The 1975 report focuses mostly on testing and average race differences in IQ. The 1996 report is more comprehensive because that task force had a wider mandate: to provide an authoritative summary of all intelligence research; its work was not confined to testing or to differences across demographic groups.
Genetic Influences on Intelligence
One clear area of difference is in the discussion of the genetic influences on intelligence. Compare this quote from the 1975 report . . .
Studies of family relationships and other experimentally uncontrolled studies of human genetics are suggestive but not convincing.Cleary et al. (1975, p. 20)
. . . with this statement from the 1996 report:
If one simply combines all available correlations in a single analysis, the heritability (h2) works out to about .50 and the between-family variance (c2) to about .25 . . . In childhood h2 and c2 for IQ are of the order of.45 and .35; by late adolescence h2 is around .75 and c2 is quite low (zero in some studies).Neisser et al. (1996, p. 85).
Gone are any uncertainties about the influence of genes on intelligence in 1996! That being said, the authors of the 1975 report did admit that “. . . no scientist can unequivocally reject possible genetic influences [causing average race differences in IQ] on the basis of present data” (Cleary et al., 1975, p. 16).
The articles also differ in the authors’ willingness to give a definition of intelligence. The 1996 statement bats around several theories–including Sternberg’s triarchic theory of intelligence and Gardner’s theory of multiple intelligences. Those theories were popular in the 1990s (and remain so, among non-mainstream researchers and non-experts), but the empirical support for them has never been strong. This is pretty much what to expect with Robert J. Sternberg being one of the task force members in 1996. 🙄
In contrast, the 1975 statement is much clearer:
Intelligence is defined as the entire repertoire of acquired skills, knowledge, learning sets, and generalization tendencies considered intellectual in nature that are available at any one period in time.Cleary et al. (1975, p. 19).
Unlike the 1996 statement, there is no ambiguity into what the 1975 authors are talking about when they use the word “intelligence.” The lack of a definition of intelligence is one of the weakest parts of the 1996 statement. Mainstream psychologists did know what intelligence was in the 1990s, as demonstrated by the widespread agreement with the Gottfredson (1997) definition:
Intelligence is a very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience.Gottfredson (1997, p. 13).
Although both statements were commissioned in response to controversies, the 1975 statement’s authors are much more willing to discuss the social implications of intelligence research. For example:
No matter how high one draws a line for qualification on the intellectual dimension, one will find Negroes who qualify. The proportion of Negroes who qualify, however, is smaller than the proportion of Caucasians when selection is from the upper half of our population. . . . The racial difference of one standard deviation . . . does not justify segregation, [but] it does have consequences that cannot be ignored. Furthermore, if steps are to be taken to do something constructive about racial differences, it is foolish to minimize the extent of the problem.Cleary et al. (1975, p. 16)
In contrast, the authors of the 1996 statement commit themselves to . . .
. . . make clear what has been scientifically established, what is presently in dispute, and what is still unknown. In fulfilling that charge, the only recommendations we shall make are for further research and calmer debate.Neisser et al. (1996, p. 78)
While neither group makes social or policy recommendations, it is not clear in reading the 1996 statement why intelligence matters, why non-scientists should care about the topic, or why it should be studied at all. Even if The Bell Curve had never been written, intelligence would still be an important topic of study because it does impact people’s lives.
Probably because the 1975 statement was issued in response to a call to ban intelligence testing for some populations, a section of the article is devoted to test misuse. The section on test misuse makes the 1975 statement relevant reading today. There is nothing like this in the 1996 statement. It includes several passages and recommendations that would be good to keep in mind when using psychological tests in an ethical fashion:
Banning tests will not, however, ban carelessness, insensitivity, or ignorance from schools. Furthermore, when these characteristics are present, they find outlets in many classroom and school practices, attitudes, and decisions not connected with the tests. Test misuse is only one symptom of more basic difficulties.Cleary et al. (1975, p. 22).
A child should not be locked into a track system or stereotyped intellectually on the basis of a single test score.Cleary et al. (1975, p. 22).
That recommendation is part of the modern Standards for Educational and Psychological Testing (American Educational Research Association et al., 2014).
Finally, it is worth praising the 1975 report. Some sections of it sound astonishingly modern and hold up well:
Undoubtedly, the heredity-environment issue with respect to intelligence has been tied to political-social-economic beliefs. Actually there is only the vaguest sort of association in any fundamental sense between the two sets of issues, but some very inadequate notions about the nature of inheritance have been used to support some conservative beliefs, while similarly some liberals have discounted genetic contributions to human behavior on equally inadequate notions. The passionate state of mind engendered by the association of important human values . . . leads to more heat than light concerning the scientific issue.Cleary et al. (1975), pp. 15-16
Other interesting points from the 1975 report that are empirically supported today:
- No test measures “innate” ability and all tests require examinees to draw upon their prior learning and experience in some way (Cleary et al., 1975, pp. 17).
- Intelligence tests are not a measure of adherence to middle-class culture (Cleary et al., p. 23).
- The problems of operationalizing intelligence as the ability to adapt to one’s own environment (Cleary et al., p. 24).
- The superiority of testing as a tool of selection compared to other alternatives (Cleary et al., pp. 32-35).
On the other hand, the 1996 statement is much more thorough. It includes a discussion of so many topics that are not included in the 1975 statements, including:
- Biological and genetic variables that correlate with IQ (Neisser et al., pp. 83-86).
- Environmental influences on IQ (Neisser et al., pp. 86-89).
- Sex differences in mental abilities (Neisser et al., pp. 91-92).
- A discussion of race/ethnic group differences that goes beyond Black-White American differences and speculates about causes (Neisser et al., pp. 92-95).
I do not think that these lists show that either statement is better than the other (though the 1996 statement is more up-to-date, of course). Rather, each group of authors was merely giving their readers a statement on the issues of the day. Cleary et al. (1975) were focused mostly on tests because of ABPsi’s call to ban intelligence tests for some populations. The Neisser et al. (1996) statement was written because APA found a need for an authoritative report on the science of intelligence.
Both statements have their strengths. Both are worthy of attention and reading. But both are showing their age. Whenever APA’s Board of Scientific Affairs issues a new statement on intelligence, that one will also be worth reading. I think that the 1975 and 1996 statements will always have their place, even if it is to show the stability of some findings and principles that have been a part of intelligence research for decades.
American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for educational and psychological testing. American Educational Research Association.
Cleary, T. A. (1968). Test bias: Prediction of grades of Negro and White students in integrated colleges. Journal of Educational Measurement, 5(2), 115-124. https://doi.org/10.1111/j.1745-3984.1968.tb00613.x
Cleary, T. A., Humphreys, L. G., Kendrick, S. A., & Wesman, A. (1975). Educational uses of tests with disadvantaged students. American Psychologist, 30(1), 15-41. https://doi.org/10.1037/0003-066x.30.1.15
Gottfredson, L. S. (1997). Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography. Intelligence, 24(1), 13-23. https://doi.org/10.1016/S0160-2896(97)90011-8
Herrnstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life. Free Press.
Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58(4), 525-543. https://doi.org/10.1007/BF02294825
Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Ceci, S. J., Halpern, D. F., Loehlin, J. C., Perloff, R., Sternberg, R. J., & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51(2), 77-101. https://doi.org/10.1037/0003-066X.51.2.77
Williams, R. L., & Mitchell, H. (1978). What happened to ABPsi’s Moratorium on Testing: A 1968 to 1977 reminder. Journal of Black Psychology 4(1-2), 25-42. https://doi.org/10.1177/009579847800400104