A comparison of direct observation and self- report measures of parenting behaviour.

Abstract

The importance of parenting behaviour as a risk factor for the development and persistence of externalising behaviour problems, and as a focus for early interventions, has led to a desire for the accurate measurement of parenting attitudes and practices. Several methods of measuring parenting behaviour have been employed in theoretical research and program evaluation, including self-report and observational measures. Both of these methods have advantages and limitations, but a systematic comparison of self-reports and observations of parenting behaviour has not been conducted. This study compared self- reports and observations of parental responsiveness and control in a sample of 68 parents of preschool children in metropolitan Adelaide, South Australia. Videotaped observations of parents interacting with their children in four set tasks (free play, drawing, pack-up and no distraction) were interval-sampled using behavioural items that paralleled self- report items on the Parenting Scale and the Child-Rearing Practices Questionnaire. In addition, parents completed questionnaires about their child's behaviour and temperament, and factors associated with parenting behaviour including parental psychopathology, social support, marital adjustment, disagreements about child-rearing and demographic characteristics. Reports of children's behaviour at preschool were also obtained from teachers. When the correlations between corresponding behaviour domains on the self-report and observational measures were compared, the only significant correlation was for parental warmth. Observations of parental control practices (such as permissiveness, inconsistency and overreactivity) were not significantly associated with self-reports of these behaviours. The observational system used in this study yielded observations of permissive and inconsistent parenting that had good validity and reliability, whereas self-reports of harsh and overreactive parenting practices were more validly and reliably assessed using self-report methods. Self-reports and observations of responsive parenting practices demonstrated results that are more equivocal. Possible explanations for these results and the implications for the use of self-report and observational measures in parenting research are discussed.