Manipulating Money in Math: (Whom) does it help ?

Math manipulatives are suggested to enhance learning outcomes (Carbonneau et al., 2013), and bridge abstract concepts with real-world scenarios (Moch, 2008). Their effect appears to vary across student’s prior language and math achievement levels (Bresser, 2009, Moyer-Packenham, & Suh, 2012). Our study aimed to further investigate these effects using real money during math problem-solving and considering potential gender effects known to influence performance in measurement related math problems (e.g., Vasilyeva et al. 2009; National Center for Educational Statistics, 2004). Participants included 97 3rd graders (59 girls, age: M = 9.45, SD = 0.54) who had previously taken a national standardized achievement test (prior achievement in math and in language. Students solved money-related math problems with and without money (condition) in a between-subject design. After the initial assessment, students participated in a 45-minute training session on money decomposition and interactive coin activities (day 1), followed by a repeated assessment the next day (day 2). Repeated Measures Anova controlling for socioeconomic status revealed no effect of condition on students’ problem-solving performance, but a significant day × condition interaction (p = .047). Condition did not interact with prior achievement, but we observed a marginal triple interaction of day × condition × gender (p = .052). The manipulation of money generally interfered with performance before the training and training improved performance in both gender groups. However, boys profited more from the training, such that the detrimental effect of manipulating money tended to disappear in boys on day 2.