In Brief

Though most siblings have somewhat similar temperaments, parents often perceive their children as different as night and day, suggests a study published in the November issue of APA's Journal of Personality and Social Psychology (Vol. 87, No. 5). Perhaps more importantly, according to the authors, the article questions the value of relying exclusively on parental reports in child studies.

Lead author Kimberly J. Saudino, PhD, a psychology professor at Boston University, had previously found that parents of twins saw little similarity in their children's temperaments, despite the twins' common genetic makeup and objective measures that found they had similar temperaments, and she was curious if the bias applied to nontwin siblings.

To find out, she and students Annie Wertz, Jeffrey Gagne and Sonia Chawla used parental ratings and objective measures to assess the activity levels and shyness of 95 pairs of siblings between 3 and 8 years old. The pairs were roughly half same- and half mixed-gender siblings.

In the lab, researchers recorded the children's level of shyness--for example, if they took a toy offered to them or clung to their parent instead. The researchers then attached motion recorders--small accelerometers measuring how often and vigorously a limb moves--to the dominant leg and arm of each child. Back home, the children wore the detectors for 48 hours. Parents also completed questionnaires about their children's temperaments.

The researchers found that the objective measures--their shyness evaluations and motion-detector readings--correlated little with parents' reports, suggesting that how parents view their children differs substantially from their children's real temperaments.

Notably, parents reported negative correlations between their children's temperament dimensions. That is, parents didn't just say that their children were dissimilar, but that they were completely opposite: If one was shy, the other was outgoing.

Why the parental bias, in spite of children's genetic similarities? "It could be a response to a society that values individuality, or it emphasizes children's individual place in the family," says Saudino, who suspects these results might disappear if studying collectivist cultures.

The answer may be even more basic, she theorizes: Defining one child through another may be a simple way to better understand and interact with each child, she says.

Perhaps more significant are the findings' theoretical implications. "This bias calls into question the validity of research using parent ratings to look at sibling differences," Saudino adds. She suggests that studies of children employ more objective measures and avoid relying exclusively on parent reports.

--M. GREER