75 years later, study still tracking geniuses
Posted by Dean Holden at March 10th, 2013
by Daniel Goleman, 7 March 1995
IN 1921, Dr. Lewis M. Terman, a Stanford University psychologist and a pioneer of the I.Q. test, scoured California’s schools to identify 1,521 children who scored 135 or over on his new intelligence test, the Stanford-Binet.
Terman’s little geniuses — who as the study went on took to calling themselves “Termites” — are now in their 80’s, and have been contacted by researchers every 5 or 10 years, making the Terman Study of Genius the grandfather of all life-span research.
Now entering its 75th year, the study is still going strong. Just last month the data yielded an article on links between childhood traits and longevity; in April Stanford University Press will publish a volume by Dr. Terman’s successor, the late Dr. Robert R. Sears, a psychologist at Stanford, on the Terman children late in life. Over the years more than 100 scientific articles and almost a dozen books have been based on the Terman data.
In the world of social science, such longitudinal studies are the method of choice for assaying the mysteries of the seasons of life. Such studies allow researchers to analyze large groups of people over many years, and so tease out the hidden and often murky links between cause and effect that would be missed in other kinds of studies. Recent findings from other studies include, for example, that harsh sentences for young criminals shut them off from key opportunities to avoid a criminal career, indications of the traits of preschool children that put them at heightened risk of drug use as teen-agers, and the personality traits that lead symptoms of post-traumatic stress to wane as life goes on.
“It’s only by following people over years that you can most accurately track the relationship of early traits or influences on the course of later development,” said Dr. Anne Colby, a psychologist and director of the Murray Research Center at Radcliffe College.
The Radcliffe center has become the largest repository for longitudinal studies, including a duplicate set of the Terman data and more than 150 other such studies done in this country. Typical of the information available to scholars at the center are data on the life course of 510 parolees first assessed in 1921, or 300 newlyweds first tested in 1935. The main alternative to such studies is retrospective research. Because retrospective research relies “on recollections of what happened in childhood,” Dr. Colby said, it may be “distorted by people’s notoriously imperfect memory.”
Shepherding data over the course of lifetimes is itself a logistical challenge. The Terman study, for example, is in its third generation of overseers, having outlived its first two directors. Dr. Terman was the head until 1954, when he turned the study over to Dr. Sears, who was himself a Termite. When Dr. Sears died in 1987, Dr. Al Hastorf, a colleague, became director.
The Terman study, the first to follow such a large number of people since childhood over their entire life span, continues to yield new insights about each stage of life. Only now that the Terman boys and girls have reached their 80’s have investigators discovered that those children whose parents divorced faced a 33 percent greater risk of an earlier death than those whose parents remained married until the children reached age 21.
For men whose parents divorced while they were children, the average age of death was 76, compared with 80 for those whose parents remained married, according to findings by Dr. Howard Friedman, a psychologist at the University of California at Riverside. To do the study Dr. Friedman had to track down hundreds of death certificates throughout California and 20 other states.
The Terman children were born, on average, in 1910; in those days divorce was rarer and carried more stigma than is now the case. Dr. Friedman believes, based in part on a large body of recent research, that it was the stress and anxiety caused by their parents’ strife that took its toll in later life.
Perhaps surprisingly, there was “very little” effect on mortality for those boys and girls who had a parent die, Dr. Friedman said, suggesting that the tensions of divorce were detrimental, rather than simply the absence of a parent. “The children whose parents divorced also tended to have less marital stability themselves,” Dr. Friedman said. “They followed a different track through life.”
The Terman study adds to the debate about the influence of I.Q. in life success, suggesting that intelligence itself is not the only ingredient necessary for outstanding achievement.
In 1968 Melita Oden, a research associate of Dr. Terman’s, published a study of 100 Termites who at midlife had attained the most success and 100 whose careers had foundered. The successes, whom she called A’s were in professions like law and medicine, or were university professors or business executives. The other group, the C’s, were in occupations like sales clerks, far below their intellectual potential. One, who had earned an advanced degree in engineering, was working as a technician.
The A’s, to be sure, on average had I.Q.’s seven points higher than the C’s: 157 versus 150. But small differences in scores at the extreme high end of the I.Q. curve translate into little actual difference in ability. Such a difference is “meaningless,” said Dr. Hastorf, the current shepherd of the Terman data and a psychologist retired from Stanford University.
But other differences were telling. The A’s were more motivated from the start; they skipped more grades in grammar school, and went further in their education than the C’s. As youngsters, the A’s were rated as more lively and engaged than the C’s, taking part in more extracurricular activities in school and, throughout their lives, in more sports.
Perhaps most significant in explaining the difference in career success, said Dr. Hastorf, were character traits. From childhood on, the C’s showed a lack of persistence in pursuing their goals, whether in school or work; the A’s, at an average age of 11, already showed greater “will power, perseverance and desire to excel.”
Similar findings have come from another longitudinal study, a 1992 follow-up of 379 Boston children first observed in 1951, when they were 5 years old. It found that parents’ putting pressure on the children to achieve was a stronger predictor of their earnings and job achievement than were childhood I.Q. scores. That study was begun by Dr. Sears and colleagues while he was at Yale, and ended up in the archives of the Center at Radcliffe. The follow-up was done in 1992 by Dr. David McClelland and Dr. Carol Franz, then at Boston University.
Some of the relationships teased out of such long-term data can challenge conventional wisdom. Such is the case with recent findings that suggest sentencing juvenile offenders to long prison terms may have the unintended effect of shutting them off from the most promising avenues of exit from a life of crime.
That conclusion follows from a new analysis of data first collected in the 1940’s on 1,000 teen-age boys from impoverished areas of Boston, half of whom were juvenile delinquents, the other half carefully matched case by case but with no criminal record. The boys were followed by the original researchers, Dr. Sheldon Glueck and Dr. Eleanor Glueck, Harvard psychologists, for up to 45 years.
Recent re-analysis of the Glueck data (which is in the archives at the Radcliffe center) revealed that two influences in the boys’ lives between ages 17 and 25 turned them away from committing further crimes: getting a stable job that they cared about and where the employer valued them, or marrying a woman with whom they felt a strong tie, and supporting her and any children.
For those teen-age delinquents who managed to find a stable job, only 32 percent went on to commit crimes, while the crime rate among those who did not was 74 percent. Similarly, for those delinquents who became committed to a marriage, just 34 percent went on to engage in crime, while for those without such a strong marital bond 76 percent engaged in crimes.
“Teen delinquents are more than twice as likely to be still committing crimes in their late 20’s if they have low job stability or a weak marriage,” said Dr. John Laub, a professor of criminal justice at Northeastern University in Boston and a visiting scholar at the Radcliffe center.
“From our findings I’m convinced that the current policy of locking up young criminals for long periods is counterproductive,” added Dr. Laub, who published his findings in a 1994 book, “Crime in the Making” (Harvard University Press), written with Robert Sampson, a University of Chicago sociologist. “Putting young offenders in prison cuts them off from the very opportunities that might allow them to become productive members of society.”
By allowing connections to be made between childhood traits and later habits, longitudinal data allow researchers to identify early factors that put people at risk for later problems. “From nursery school data we can identify the kids most likely to be in the drug scene 15 years later,” said Dr. Jack Block, who for more than two decades has directed a study of several hundred children in Oakland, Calif., who were recruited as subjects when they were 3.
In interviews when the children were 14, Dr. Block found that for girls, those who in preschool were sulky and whiny, prone to teasing, anxious, unkempt and showed less interest in pleasing others were most likely to be using drugs at age 14. For boys in preschool the markers for those most likely to use drugs in their teen years included being uncooperative, prone to upsets and open in their expression of negative feelings, inattentive and aggressive.
A study by Dr. Block four years later, when the teen-agers reached 18, however, found that not all adolescent drug use boded a grim future. In this study, those teen-agers who had experimented with drugs like marijuana during their teen-age years — compared both to those who used them heavily and those who abstained — were the best adjusted.
The teen-agers who used drugs most frequently were the most alienated, had the poorest impulse control and the most emotional distress, while those who had never tried any drugs were more anxious, emotionally constricted and socially inept. Other data show that nearly two-thirds of young adults in the United States have experimented with marijuana at one time or other.
Dr. Block’s conclusion was that drug use is a symptom of maladjustment, not a cause, and that it can best be understood in the context of the larger course of life.
While data collected years or even generations ago can still yield such insights, one drawback — particularly with studies begun many decades ago — is that the original measures used can sometimes be out of date or irrelevant to a contemporary investigator. The psychological measures fashionable in the 1920’s or 1940’s, like elaborate measures of body type, are sometimes museum relics to researchers in the 1990’s.
“When you walk in to a set of longitudinal data you benefit from the years of toil the original researchers put in, but you also inherit the flaws,” said Dr. Franz, a psychologist at the University of California at Berkeley.
Researchers must be resourceful in how they use what can otherwise be useless, antiquated data. That was true of Dr. George Vaillant, a psychiatrist at Harvard Medical School, when he first encountered data that had been collected on Harvard graduates in the 1940’s. Called the “Grant Study” after the W. T. Grant Foundation, which provided the original financing, “the data sat unused at Harvard for years,” Dr. Vaillant said.
But by tracking down the men, now in middle age, and conducting follow-up interviews with them, Dr. Vaillant has been able to harvest ample research results. The most recent, a study of recovery from post-traumatic stress disorder done with Dr. Glen Elder, a sociologist at the University of North Carolina, will be published this year in The American Journal of Psychiatry.
Interviews and writings of the men while in college, and after returning from combat allowed Dr. Vaillant to determine that those who were most emotionally vulnerable before their traumatic experience were most likely to continue to suffer from symptoms like flashbacks and nightmares almost a half century later.
The Grant data, when Dr. Vaillant first encountered it, “was regarded by Harvard’s psychology department as an embarrassment,” Dr. Vaillant said.
“They thought the wrong questions had been asked,” he said. “But for me it was a gold mine.”