This study aimed to address an ongoing question regarding the relationship between parental and laboratory ratings of shyness in preschool children (e.g., Kagan, Snidman, & Arcus, 1992). Seventy-five 5-year-old children were paired with 75 age-and sex-matched peers in a 20-minute free-play situation. Play behaviors of shyness were coded from videotapes, and parental ratings of shyness were obtained for each proband.

A step-wise linear regression analysis was conducted with a behavioral measure of shyness in the probands serving as the dependent variable, and level of shyness displayed by the peer, sex of proband, number of children in the home, and parent ratings of approach and withdrawn behaviors serving as the independent variables.

Results indicated that the behavioral measure of shyness in the proband was significantly predicted by the level to which other children interacted with the proband and by the parental rating of withdrawn behaviors in each 10-minute segment of free-play. There were no sex differences on any of the measures.