Background: Accuracy of temperament assessment is a prerequisite in research studies. To identify the extent to which parental assessment of child temperament is biased by their personal attributes, we proposed a new structural equation model, in which biases of parental attributes in their assessment of child temperament can be separated from the true (i.e. non-biased) associations between the two.
Methods: We examined 234 father-mother pairs using questionnaires including Emotionality, Activity, Sociability, and Impulsivity; Social Desirability Scale; Hospital Anxiety and Depression Scale; Temperament and Character Inventory; and State-Trait Anger Expression Inventory.
Results: Paternal Depression and Persistence, maternal Trait Anger, and parental Novelty Seeking showed significant bias in assessment of Emotionality. Maternal Self-transcendence showed significant bias in assessment of child Impulsivity.
Conclusion: Researchers should be cautious about biases in parental assessment of children's Emotionality and Impulsivity, but other temperament traits may be free from such biases.
Keywords: assessment bias; child temperament; emotionality, activity, sociability, and impulsivity; parental assessment; structural equation model.
© 2015 Japan Pediatric Society.