The Origins of Mathematical Interest, Motivation, and Self-Perceived Ability

• In Event: Interdisciplinary and Cross-Cultural Investigations Into Cognitive and Motivational Underpinnings of Mathematical Development

Abstract

According to the commonly held view, individual differences in mathematical self-evaluation of abilities and interest (an overall motivation towards mathematics) are driven by environments, such as parental or teacher influence, rather than genes. The current study measured mathematical motivation in a large representative longitudinal sample of monozygotic (MZ) and dizygotic (DZ) twins, based in the UK, at ages 9, 12 and 16. At ages 9 and 12, twins’ self-evaluation was measured by asking ‘how much do you like?’ (perceived ability) and ‘how good do you think you are?’ (liking/interest) on aspects of National Curriculum mathematics (solving number and money problems, doing maths in your head, and multiplying and dividing). At age 16, three PISA (Programme for International Student Assessment) items measuring interest and eight PISA items assessing perceived self-efficacy were administered. At age 9 twin correlations were 0.41 for MZ twins (N=1175) and 0.13 for DZ twins (N=1022) on perceived ability; and 0.39 for MZ twins (N=1164) and 0.17 for DZ twins (N=1009) on liking scales. A very similar pattern of correlations was observed at age 12: 0.49 for MZ (N=2015) and 0.16 for DZ twins (N=1812) on perceived ability; and 0.46 for MZ (N=2015) and 0.21 for DZ twins (N=1812) on liking/interest. Overall, self-evaluation was relatively stable across years. Next, we applied standard structural equation twin model-fitting to the data to assess the extent of environmental and genetic influences on individual differences in mathematical self-evaluation and motivation. Significant heritability (approximately 40%) was found at ages 9 and 12, suggesting that observed similarity between the twins comes from similarities in their genetic make-up. The results showed no significant influence of shared environments (environmental influences that contribute to similarity between the twins growing up together in the same family or same school), but considerable non-shared environmental influences at both 9 and 12 (59% and 57% respectively). Preliminary results on the first cohort (2,274 individuals – 1,137 twin pairs) at age 16 provide further support for these findings. To further investigate the mechanisms of environmental contribution to variation in mathematical motivation, we compared twin similarity at age 9 and 12 between two groups: twins taught mathematics in the same class by the sane teacher and twins taught in different classes by different teachers. We found no teacher/classroom effect – the twin correlations were practically identical for both groups. For example, at age 9, for Liking mathematics, MZ correlation for same teacher was 0.39 (N=665) and for different teacher 0.38 (N=451); DZ correlation for same teacher was 0.18 (N=580) and for different teacher 0.18 (N=385). Overall, our study suggests that even twins growing up together and often attending the same school and class are not very similar to each other in liking, interest, and self-evaluation of mathematical activities. Any observed similarity between the twins seems to come from similarities in their genetic make-up, rather than family or school environments, whereas most differences come from non-shared environments. We discuss these findings in terms of implications for understanding gene-environment interface and for mathematical education.

Authors

• Beata Tick, King's College London

• Yulia Kovas, Goldsmiths, University of London and Tomsk State University

• Robert Plomin, Kings College, London