The success of science education in classroom and out-of-school settings can be influenced by parents’ behaviours and STEM-related values. The present study investigated pathways in parent-to-child transmission of STEM (science, technology, engineering, mathematics) values by examining at same time parents’ values and behaviours, along with their children’s perceptions of these parental influences. The study included 1071 students (M_age = 12.15) and the same number of their parents. Path analysis revealed that children’s importance value of the STEM school fields was best explained by their perceptions of parental values and behaviours in STEM. On the other hand, parents’ self-reported values and behaviours had a weak effect in predicting children’s values, which can be explained by inaccurate children’s perceptions of their parents. The results suggest that parents more easily convey beliefs about the utility than the attainment value of STEM. Namely, parents’ utility value had a larger effect in predicting children’s value, partly mediated through children’s perception of parents’ encouragement of STEM interests. The study highlights the role of children’s perceptions of their parents’ beliefs and behaviours and the importance of communicating STEM-related values within the family. Practical implications for parents and science educators are discussed.

Background: STEM school achievement is related to various student, parent and family characteristics, and experiences in the school setting. Studies also acknowledge the role of informal learning in STEM achievement, mentioning the importance of student digital activities at home. However, the existing studies have not examined in detail the relationship between students’ engagement in specific technology-based activities at home and STEM achievement, and whether this relationship varies depending on student gender and parental education. Purpose: This study explores the relationship between students’ engagement in technology-based activities at home and STEM school achievement, and the possible moderating role of gender and parental education in this relationship. Sample: Participants in the study are 1,205 students (580 girls) attending fifth (n = 567) and sixth grades (n = 638) in 16 primary schools in Zagreb, Croatia. For each student, one of their parents also participated in the study, by completing questionnaires for parents. Methods: Students completed questionnaires containing questions assessing the frequency of various technology-based activities at home, and their STEM school achievement was assessed by objective paper-pencil tests. Results: The frequency of writing posts on social media sites and downloading music, movies, games or software from the internet negatively predicted STEM achievement, even after controlling for parental education and student gender. The relationship between frequency of certain technology-based activities at home and STEM achievement was found to vary depending on student gender and the combination of student gender and parental education. Conclusions: Although the frequency of certain activities at home is related to STEM achievement, the general conclusion is that engagement in technology-based activities at home does not substantially contribute to STEM achievement. Additionally, the relationship between technology-based activities and STEM achievement may differ for boys and girls, and for students with parents of different education levels.