Poster #42 - The dimensionality of children’s informal mathematical knowledge in the United States, China, and Japan

• In Event: 1-169 - Poster Session 04
  In Poster Session: PS 04 Section - Cognitive Processes

Integrative Statement

Cross-national studies of children’s mathematics achievement in elementary and middle school have long demonstrated that American children perform lower than children from many other countries, including China and Japan (e.g., Provasnik et al., 2016; Stevenson, Chen, & Lee, 1993). While there are likely many contributing factors, this achievement gap in children’s informal mathematical knowledge is evident during early childhood with differences emerging as early as three years of age. That is, Chinese and Japanese children demonstrate more advanced knowledge than American children and these differences become more pronounced over the preschool years (Starkey & Klein, 2008). Given that preschool children’s early competencies in informal mathematical abilities are predictive of formal mathematics achievement in later grades (Bisanz, Sherman, Rasmussen, & Ho, 2005; Purpura, Baroody, & Lonigan, 2013), this disparity has broad implications for child outcomes.

Recent articles in the mainstream press (Chu, 2017; Qin, 2017) highlight one approach to improving the mathematical performance of American children. This recommendation is to use the same instructional methods and materials as used in higher performing countries. However intuitive, there is no evidence to suggest this would be beneficial. First there is a need to better understand the nature of mathematical knowledge and how it develops in each country. In the current study, we utilize confirmatory factor analysis (CFA) to determine whether the early mathematical abilities of children in the U.S., China, and Japan have a similar dimensional structure across the preschool years.

The participants were 639 children (48% female) who were living in the US (n=199; Mean age (Time 1) = 3.61 years, SD=0.26), China (n=237; Mean age = 3.63 years, SD=0.27), and Japan (n=202; Mean age = 3.73 years, SD=0.28). Approximately half of the children in each country were from families of lower socioeconomic status (SES), as measured either by income (the U.S.) or parental education (China and Japan). Children’s informal mathematical knowledge was assessed at four time points (fall and spring) over two consecutive years of preschool using the Child Math Assessment (CMA; Klein & Starkey, 2004). The CMA includes multiple items spanning number and operations, geometry, pattern knowledge, and measurement. It was administered to children individually by native language speakers.

Preliminary CFA model fit comparisons have yielded different models across country. In the U.S., a four factor model of Number and Operations, Measurement, Geometry, and Patterning fit the data best, with adequate fit at times 2 to 4 (e.g., RMSEAs= .024 - .031). However, no appropriate model could be achieved for Time 1, likely due to floor effects. In China, the same four-factor model fit the data better than other factor models, but the fit was still poor, suggesting a different theoretical model applies to this data. Finally, in Japan, a three-factor model of Number and Operations, Measurement, and Spatial Math (geometry + patterning items) fit the data best. These findings suggest that, as early as age three, the dimensionality of children’s mathematical knowledge varies by country. The complete analyses will be presented and the implications and future directions discussed.

Authors

• Lydia DeFlorio, University of Nevada, Reno
  Presenting Author

• Trelani F Milburn-Chapman, University of Alberta
  Non-Presenting Author