Parents’ math-gender stereotypes and children’s perception of math ability at the age of 6: the role of children’s appraisal of parents’ evaluations.

Theoretical background. The expectancy-value model (see Eccles, 2007) emphasizes the role of family influences in the emergence of early gender differences in children’s self-perception of math ability. However, thus far most of the studies have addressed children and their families from late childhood through adolescence. Moreover, past research has often overlooked the role of children’s subjective appraisals of parents’ beliefs in the construction of ability beliefs. Aims and hypotheses. The present study investigated the influence of mothers’ and fathers’ math-gender stereotypes on parents’ evaluations of children’s math ability, and on children’s self-perception of math ability at the age of 6. The study also tested whether children’s appraisal of parents’ evaluations is a mediator of parental influences, as predicted by the expectancy-value model, or rather whether it represents a projection of children’s self-view, as predicted by the projected appraisal model of self-knowledge development (Felson, 1993). Method. Participants were 252 first-graders aged 6 and their fathers and mothers. Children’s self-ratings and their appraisal of both parents’ evaluation of math ability were collected via individual interviews. Parents completed self-administered questionnaires measuring math-gender stereotypes and evaluations of the child’s math ability. Teachers’ ratings of math ability were also collected. Data were analyzed usingtructural equations models with robust estimation methods (i.e. WLSMV and MLR). Results. Parents' evaluations of children's math ability fully mediated the effect of teachers' ratings children's self-perception of ability. Fathers evaluated their sons better than their daughters. Children's self-perception was affected by the interaction between mothers' math-gender stereotypes and children's gender, with a negative effect on girls’ self-perception of ability. Finally, the projected appraisal model fitted the data better than the reflected appraisal model: children's self-perceptions mediated the relationships between actual parents’ evaluations and children’s appraisals, rather than the vice versa. Importantly, girls whose mothers endorsed math-gender stereotypes at a higher extent perceived to be considered as less competent in math by both parents, regardless of parents’ actual evaluations. Discussion and implications. The direct path linking mothers’ stereotypes and daughters’ self-perception suggests that unintended, stereotype-laden non-verbal behaviors may subtly affect girls’ self-perceptions from the very beginning of schooling, regardless of parental appreciation of daughters’ actual math achievement. Overall, results suggest that actions contrasting the gender gap in STEM fields should be promoted from the earliest school grades, with theinvolvement of families as well.