Is Your Older Brother Smarter than You Are?
Psychologists, composing a young and mainly untried field, saw opportunity in the fear of the feebleminded, for they had a new tool, the IQ test, which they believed could ferret out the mentally retarded, and indeed plumb the biological depths of the rest of us, as well. Their efforts led to considerable misery, such as the coerced sexual sterilization of more than 60,000 low test scorers in the US, but also, strangely enough, to the much-discussed recent reports that firstborn children tend, on average, to have IQ scores of three points higher than subsequent children.
"Three points on an I.Q. test may not sound like much," reported the New York Times on June 21. "But experts say it can be a tipping point for some people — the difference between a high B average and a low A, for instance. That, in turn, can have a cumulative effect that could mean the difference between admission to an elite private liberal-arts college and a less exclusive public one."
Beware the IQ and intelligence experts. With more frequency than in other fields, they need to believe they understand us to a degree they don't, and they are often afflicted with grand and dangerous ideas for society.
Understanding how IQ tests were developed should make us cautious, at the very least, when considering exam results. The IQ tests that we use today grew from American psychologists, and sometimes doctors, testing literate and illiterate people in various contexts. In the early 1900s, xenophobic Americans were quite concerned about the effects Southern and Eastern European immigrants (presumed to be feebleminded more frequently than those from the West and North) were having on the national gene pool and society.
Responding to these nativist forces, in legislation that would have significant unintended consequences, Congress asked doctors of the US Public Health Service stationed on Ellis Island to keep out the feebleminded. The doctors did so with a new intelligence test imported from France (which was composed mainly of common sense and school-type questions), as well as questions of their own devising geared toward test-takers who hadn't attended school and didn't speak English—often from South and Eastern Europe. The doctors found children's blocks, jigsaw puzzles and boards with everyday items such as dolls and scissors attached to them to be of help. Immigrants who performed poorly on these "non-verbal" tasks were sent home.
When the US entered WWI, in 1917, psychologists tested about 1.9 million recruits with exams cobbled together from Ellis Island, France, and elsewhere, but still with this "verbal/non-verbal" distinction for literates and illiterates. In the process of creating the alpha exam, for literates, and the beta, for illiterates, the psychologists laid the structural foundation of IQ tests today, as well as the methods and tools for the entire standardized testing industry.
Two decades after the war a former army intelligence tester named David Wechsler joined together the test questions for literates and illiterates—the verbal and non-verbal—into one handy exam. Many psychologists of the day thought this strange. Why test literate people with questions for illiterates? But Wechsler's exams were so easy to administer and comprehensive feeling that they quickly became the dominant test and remain so today.
Psychologists and IQ researchers today have forgotten this history. They believe that IQ tests measure intelligence, but the exams are based on historical circumstance and necessity, not on any serious intelligence theory. While our understanding of the brain, cognitive processing, and genetics has improved, 11 of the 14 subtests that compose today's WAIS-III (the Wechsler Adult Intelligence Scale) are based on pre-WWI question types.
Add to this historical background the fact that average IQ scores have been going up every year ever since they've been developed and it becomes untenable to believe that IQ tests measure intelligence. On at least one IQ test, about 70 percent of people born in Victorian England would be considered mentally retarded today. It might be tempting to think that we are that much smarter than our ancestors, but it isn't evolutionarily feasible.
If IQ tests aren't useful in understanding individuals' intelligence, then it's a wonder that anyone finds differences in average scores between groups interesting at all. Certainly comparing African American and white scores has been a useless and extremely damaging endeavor over the past century.
Which brings us back to the sibling studies. The recent findings that firstborns score on average higher than their siblings in no way imply that they are smarter. It may be that firstborns' tendencies to follow rules (critics have long criticized IQ tests for rewarding the "good" kid) more than subsequent siblings will also lead them to do very slightly better on other exams, such as the SAT's. But how interesting is that? The SAT is a terrible predictor of freshman year grades (a University of California study found that they explain about 13 percent of the variance), and its predictive powers diminish beyond that.
Ultimately, it's not clear what we're supposed to do with the conclusion from the recent sibling IQ studies. Promote rule following in all of our children in the hopes that they might perform better on standardized tests, which, in turn, rests on the remote possibility that this might gain them entrance to Stanford rather than UC Berkeley? Instead of focusing on quirky sibling studies, let's ensure that all of our children are healthy, well educated, and enjoy learning.