Margaret Callaghan Guest has enough younger brothers and sisters to field two baseball teams--and nearly enough of her own kids (seven) to field another. With so many relatives, Margaret and her mother put out a family newsletter to keep the Detroit-area clan informed. The news is remarkably good. Margaret and her 17 siblings all put themselves through Catholic high school and college. Several have advanced degrees and one is a millionaire. Six of Guest's own children are college graduates, and the seventh just finished his first year.
Few families can claim as much academic success. But according to a recent study by Ohio State University sociologist Douglas Downey, the Callaghans might have achieved even more. Drawing on data collected in 1988 from 24,599 eighth graders in 1,550 schools, Downey concludes, "Children in larger families don't do as well in school as children in smaller families." And this holds true, he says, regardless of race, class and socioeconomic standing. The notion that small families produce better students, Downey reports, has been settled sociological opinion for decades. But for the first time, he claims, sociologists can begin to understand why standardized test scores in math and verbal skills fall as families expand.
Children from large families generally suffer from "the diminution of parental resources," says Downey, whose study will be published in the October issue of American Sociological Review. With each additional birth, he argues, parents have less time, energy, money and materials such as books and magazines to "invest" in each child. The probability that a family will have a computer for kids to use, for instance, drops "dramatically" when there are four or more children. Parents are willing to provide educational resources until they reach a certain threshold, he finds, but after four kids, the focus seems to shift to fulfilling basic needs like food and shelter. The effects of family size on grades, he observes, "are consistent but not tremendous": where an only child achieves a B-plus, children with five siblings average a B.
The implication for very large families like the Callaghans is obvious: they might have done even better at school had there been fewer children. But don't tell that to Mary Callaghan Lynch, Margaret's sister. Their parents were not wealthy: both were teachers. When the children numbered only 10, they still had no washer, dryer or car. But, Mary recalls, "we had a 12-foot dinner table and all were expected to be there at 6 sharp." At meals, the kids were always asked about their schoolwork. "You followed the older ones," says Mary. "The question wasn't 'Are you going to private Catholic schools?' It was 'You are going, and how are you going to pay for books and clothes?' "
Downey acknowledges that without further studies, he cannot rule out other cultural factors that might mitigate his findings. Mormons, for example, typically have large extended families. They also have a theology that holds that their families will be bound together for all eternity. In a 1982 study of 15,000 students (in-eluding 1,200 siblings) at Brigham Young University, Richard Galbraith, a professor of family sciences at BYU, found no correlation between family size and standardized test scores. "The number [of children] is really irrelevant," says Galbraith, who has five children himself.
At most, Downey's study confirms that a correlation does exist between family size and scores on standardized tests by grade schoolers. But it says nothing about college achievement or IQ-much less the emergence of genius. Nor can it predict individual academic performance. In short, correlation is not causation. And as even Downey concedes, the pattern he has proved is no reason to limit family size. As the Callaghans learned, family pride can be a powerful resource.