(via EurekAlert) Concordia University's Department of Psychology published a press release today summarizing the results of a 20-year study that showed that children evaluating their peers are more likely to accurately predict their adult personalities than children evaluating themselves.
"This study, known as the Concordia Longitudinal Risk Project, was started in 1976 by my colleagues in the Department of Psychology, Alex Schwartzman and Jane Ledingham, who is now at the University of Ottawa" says [Lisa] Serbin. "Over two years, Montreal students in grades 1, 4 and 7 completed peer evaluations of their classmates and rated them in terms of aggression, likeability and social withdrawal. The students also did self-evaluations."
The report only states that the peer analyses were more accurate than self-evaluations, not how much more accurate they were, or how accurate they were in general -- it could be a difference between like 10% and 20%, still significant but not particularly useful.
I'm actually pretty sure I don't want to know what my friends thought of me during my early years of socialization. I didn't think very much of myself, at the time, and I'd rather put off exploring that aspect of my psyche until I've done a better job of building a stable present-day sense of self-esteem.