Correlates of Intelligence

Genetics

The research on mindset is interesting but there can also be a temptation to interpret it as suggesting that every human has an unlimited potential for intelligence and that becoming smarter is only a matter of positive thinking. There is some evidence that genetics is an important factor in the intelligence equation. For instance, a number of studies on genetics in adults have yielded the result that intelligence is largely, but not totally, inherited (Bouchard,2004). Having a healthy attitude about the nature of smarts and working hard can both definitely help intellectual performance but it also helps to have the genetic leaning toward intelligence.

Group Differences

Carol Dweck’s research on the mindset of children also brings one of the most interesting and controversial issues surrounding intelligence research to the fore: group differences. From the very beginning of the study of intelligence researchers have wondered about differences between groups of people such as men and women. With regards to potential differences between the sexes some people have noticed that women are under-represented in certain fields. Recent data suggests, for example, that women make up between 3% and 15% of all faculty in math-intensive fields at the 50 top universities. In a comprehensive review of research on intellectual abilities and sex, Ceci and colleagues (2009) argue against the hypothesis that biological and genetic differences account for much of the sex differences in intellectual ability. Instead, they believe that a complex web of influences ranging from societal expectations to test taking strategies to individual interests account for many of the sex differences found in math and similar intellectual abilities.

So are there sex differences in intellectual ability? In a review of the research literature, Halpern (1997) found that women appear, on average, to outperform men on measures of fine motor skill, acquired knowledge, reading comprehension, decoding non-verbal expression, and generally have higher grades in school. Men, by contrast, appear (on average) to outperform women on measures of fluid reasoning related to math and science, perceptual tasks that involve moving objects, and tasks that require transformations in working memory such as mental rotations of physical spaces. However, as we will note, we must be careful about what we conclude regarding the meaning of such results.

Complications with Validity in Intelligence Testing

Stereotype Threat

Some researchers have examined various explanatory hypotheses for why sex differences in intellectual ability occur. Some studies have provided mixed evidence for genetic factors while others point to evidence for social factors (Neisser, et al, 1996; Nisbett, et al., 2012). One interesting phenomenon that has received research scrutiny is the idea of stereotype threat. Stereotype threat is the idea that mental access to a particular stereotype can have real-world impact on a member of the stereotyped group. In one study (Spencer, Steele, & Quinn, 1999), for example, women who were told that women tend to fare poorly on math exams just before taking a math test performed worse relative to a control group who did not hear the stereotype. Research on stereotype threat has yielded mixed results and we are currently uncertain about exactly how and when this effect might occur. One possible antidote to stereotype threat, at least in the case of women, is to make a self-affirmation (such as listing positive personal qualities) before the threat occurs. In one study, for instance, Martens and her colleagues (2006) had women write about personal qualities that they valued before taking a math test. The affirmation largely erased the effect of stereotype threat by improving math scores for women relative to a control group but similar affirmations had little effect for men (Martens, Johns, Greenberg, & Schimel, 2006).

Eugenics and Scientific Racism

The application and interpretation of intelligence tests historically has been problematic in other ways as well. As the United States entered World War I, there were proponents of the eugenics movement who developed and backed intelligence testing as a way to improve American society by eliminating genetic “defects” and promoting reproduction by those deemed genetically “superior.”

War was a topic of considerable debate among eugenics societies (Kühl, 2013). It was clear that many people would perish during war. On the one hand, according to the logic of eugenics, if “low-quality” people tended to perish, then war could be positive for eugenics because those people would no longer be around to breed. On the other hand, the loss of “high-quality” individuals would be a negative outcome as their genes would also be lost. Furthermore, there was no way to measure the eugenic quality of soldiers and then use that information to determine who would live or die during war. Eugenicists were advocates of using intelligence tests on soldiers to help make personnel selection decisions, such as who would become an officer, and who would have a higher probability of being killed by being sent to the front.

American Psychologist Robert Yerkes (APA president in 1917) wanted to establish a “mental census” of Americans, and then use that information to improve American society from a eugenics point of view. Yerkes advocated for the testing of all Americans and organized the largest mass mental testing of American men for the draft, administering tests to 1.75 million adults (Yerkes, 1923). There were two versions of the test (inspired by Stanford-Binet). The “alpha” test was created for soldiers who could read, and the “beta” test was created for soldiers who could not.

In his 1923 report, “Eugenic Bearing of Measurements of Intelligence in the United States Army,” Yerkes describes the methods and results from the alpha-beta testing efforts and lists these five main reasons to conduct such widespread testing:

• In the discovery of men whose superior ability recommends their advancement.
• In the prompt segregation in the Development Battalions of intellectually inferior men whose inaptitude would retard the training of the unit.
• In building organizations of equal or appropriate strength.
• In selecting suitable men for various army occupations or for special training in the technical schools.
• In eliminating the feeble-minded.

Eugenics ideology typically included racist beliefs about the superior or inferior eugenic qualities of different ethnic groups (Turda, 2010). The Alpha-beta tests of 1.75 million American men produced results that fit existing eugenic ideology about inherent differences in intelligence between ethnic groups. For example, psychologist Carl Brigham, wrote an entire book analyzing the results of the Alpha-Beta tests (Brigham, 1922). He concluded that white Americans had superior intelligence to black Americans and immigrants. He also created dire warnings about the future of America, suggesting that American intelligence was rapidly declining. He warned that although American deterioration was imminent, it could be prevented through public action and laws. For example, eugenically inferior immigrants could be kept out of the country. And, increased segregation of whites and blacks, along with laws against intermarriage could prevent further mixing of the races.

These recommendations are, of course, scary and problematic for many reasons, but let’s for now focus on the difficulties of interpreting such data. As discussed earlier, it is not entirely clear what these so-called intelligence tests measure, or what intelligence itself actually refers to. Such tests are intended to assess an objective idea of “intelligence” but in reality, they measure a researcher’s operational definition of “intelligence” (meant to specify the meaning of their own terms). Then, the two ideas of intelligence are falsely equated.

Nevertheless, proponents of eugenics were quick to claim that results from the tests really did legitimately measure supposedly intrinsic and genetically inherited qualities of humans that made some superior and others inferior. Alternative interpretations – for example that the tests measured culturally acquired aptitudes rather than intrinsic traits – took additional time to be seriously considered. Racist motivations would continue to daunt intelligence testing in America, and an extended history is beyond the current scope of this chapter.

Today, these types of controversies compel many lay people to wonder if there might be a problem with intelligence measures. It is natural to wonder if they are somehow biased against certain groups. Scientific bias (the type of bias being suggested here) is related to the psychometric properties of the test such as validity and reliability. Validity is the idea that an assessment measures what it claims to measure and that it can predict future behaviors or performance. To this end, intelligence tests are not exactly biased since they are fairly accurate measures and predictors of success in certain domains. But, we do need to be careful about the conclusions we draw from their meaning (e.g. that success on such a task is based on innate traits, or that it’s indicative of some more objective intellectual superiority). There are real biases, prejudices, and inequalities in the social world that might benefit some advantaged group while hindering some disadvantaged others.

In the video below, Hank Green gives an interesting overview of the theories and controversies surrounding intelligence testing.