Relations of Math Attitudes and Math Self-Concepts to Math Achievement: Evidence from Implicit and Explicit Measures

• In Event: Girls’ and Boys’ Attitudes and Beliefs about STEM: Sources, Development, and Relations to Achievement

Abstract

Children’s learning of mathematics is linked not only to their academic skills, but also to their attitudes (“I enjoy math”) and self-concepts about math (“I am a math person”). Early attitudes and self-concepts about math have been identified as powerful longitudinal predictors of later achievement in science, technology, engineering, and mathematics (STEM; Gunderson, Ramirez, Levine, & Beilock, 2012). Early individual and group differences in children’s attitudes and self-concepts about math are important to understand because these may be one developmental contributor to a current societal inequity—the underrepresentation of women in STEM at later educational stages and careers (Ceci, Ginther, Kahn, & Williams, 2014).

How early children begin to form math attitudes—positive versus negative affective reactions to math—is currently understudied due to the scarcity of developmentally appropriate instruments for use with children as early as Grade 1. Here, we report a novel measure of children’s implicit math attitudes.

The current study is the first to examine the relations among (a) implicit and explicit assessments (b) of both math attitudes and self-concepts, (c) as well as a measure of math achievement, (d) in the same children. This is accomplished by testing independent groups of Grade 1 and Grade 5 children in Croatia, a country that has put substantial effort into developing a nationalized and standardized grading system (Bušljeta & Kardum, 2019).

391 children (195 girls) attending Grade 1 (Mage = 7.53 years) and Grade 5 (Mage = 11.53 years) completed implicit and explicit measures of math attitudes and self-concepts. For the implicit measures, the Child Implicit Association Test (IAT) was used. The Child IAT is a computerized categorization task that measures strengths of associations among categories without requiring self-report. An IAT score (D) is calculated from the speed with which children categorize exemplars from four categories using only two response options (Cvencek, Meltzoff, & Greenwald, 2011). For the explicit measures, students completed validated surveys derived from the widely-used, standardized, TIMSS and PISA questionnaires, assessing students’ liking of and self-concepts about math (TIMSS International Study Center, 2014; OECD, 2012).

Multilevel analyses showed that first-grade girls held a strong negative implicit attitude about math (p = .005), despite no gender differences in math grades (p = .72) and self-reported (explicit) positivity about math (ps < .001) (Figure 1). The explicit measures significantly predicted math grades (p < .001), and implicit attitudes accounted for additional variance in boys (p = .01). Collectively, these results show that by first grade, girls have already developed a strong implicit negativity about math.

The striking contrast between the implicit (negativity for girls) and explicit (positivity for girls and boys) results suggest implicit–explicit dissociations in children, which mimics similar findings of implicit–explicit dissociations in adults (Greenwald et al., 2009) and raises important issues for developmental theory. Implicit attitudes may be acquired rapidly and effortlessly. Such early emerging implicit attitudes may not only form the groundwork for the development of explicit attitudes, but may also serve as a foundation for interests, choices, and learning.

Authors

• Dario Cvencek, University of Washington
  Presenting Author

• Ružica Brečić, University of Zagreb
  Non-Presenting Author

• Dora Gacesa, Faculty of Economics and Business, Zagreb
  Non-Presenting Author

• Andrew N Meltzoff, University of Washington
  Non-Presenting Author