Associate professor Timothy Konold, coordinator of research, statistics and evaluation at U.Va.'s Curry School of Education, will report his findings on April 8 at the annual American Educational Research Association (AERA) meeting in San Francisco.
"Our results indicate that both parents and teachers are important considerations when assessing a child's overall behavioral disposition," Konold said.
"The results have important implications for the manner in which we collect information on child behavior problems that are used to inform instruction and counseling decisions," he said.
Konold based his research on ratings given by mothers, fathers and teachers of a representative sample of 562 first-graders in the National Institute of Child Health and Human Development Study of Early Child Care. U.Va. is one of 10 sites of this national, 15-year research project, led by Robert Pianta, an authority on early childhood education, who is also a Curry School education professor.
One of the study's commonly used questionnaire rating 96 behaviors typically reveals different scores on the same behaviors when mothers, fathers and teachers evaluate their children, said Konold, who sought to determine which rating was a better assessment.
What he found is that parents' scores are better indicators of emotional behaviors, such as being anxious, sad or lonely, or making physical complaints (real or imagined).
"Parents do a much better job of assessing internalizing behaviors, so we should use their ratings when there's a discrepancy," he said.
When it comes to aggressive or other delinquent behaviors, called "externalizing," teachers make a better assessment, he said. Examples of those behaviors include arguing, teasing, threatening, cheating, swearing and lying. These are behaviors children may have learned from their parents, so the parents don't necessarily recognize them as deviant as consistently as teachers do.
Konold said researchers have not found these results before because they didn't have an approach for disentangling the variations attributable to the method being used to collect the ratings from observers and what the ratings were actually designed to measure (behavior problems). He used a more modern methodology developed in the last decade to make it possible for educators to interpret the measurements more effectively.
Tim Konold can be reached by e-mail at konold@virginia.edu or by phone at (434) 924-0824.