Poster #77 - Gender Differences in Early Number Sense

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Education, Schooling

Integrative Statement

Early gender differences in mathematics may have important associations with later involvement in STEM careers. However, findings are equivocal. While some studies have identified differences favoring boys as early as kindergarten (e.g., Jordan et al., 2006; Penner & Paret, 2008), others suggest that gender differences do not arise until later (Hyde et al., 1990). Little research has addressed gender differences before kindergarten or trajectories between pre-k and first grade.
We investigated the effect of gender on early number competencies through analyses of data collected as part of a larger study to develop a number sense screener (SENS) for pre-k, kindergarten, and first-graders. The SENS assesses children’s competencies in number, number relations, and number operations (α >.90).
Children were assessed with the form appropriate for their grade level (Pre-k n=200; kindergarten n=300, first n=200). The sample at each grade was approximately half female (pre-k: 51.5%; k: 51%, first: 50%), and each sample was half low-income. ANOVAs at each grade on overall accuracy showed a significant and moderate effect of gender favoring boys at kindergarten and first grade but not pre-k (Table 1). Gender did not interact with income at any grade.
To test the hypothesis that gender may differentially affect performance at different skill levels (Penner & Paret, 2008), quantile regressions were run for each grade level, testing the effect of gender at the 25th, 50th, and 75th quartiles of performance (Figure 1). Pre-k quantile regression demonstrated no gender effect at any quartile of accuracy. At kindergarten, gender did not have a significant effect at the 25th or 50th quartiles, but did at the 75th quartile (p <.001, d=.46). Boys at this quartile were estimated to perform at about 77% accuracy, with girls performing at 60% accuracy. In first grade, gender had a significant effect at the 25th (p=.01, d=.36) and 50th (p=.02 d=.34), but not the 75th quartile. Girls were estimated to perform 10% below boys at the 25th and 17% below boys at the 50th quartile. The differences by grade suggest an advantage for boys at the top of the distribution earlier on that moves toward the middle of the distribution in first grade.
We also tested the effect of gender on the three subareas. In both kindergarten and first grade, number was the only area with significant gender differences (p <.01). An item analysis revealed that counting on from double-digit numbers and naming double-digit numerals were less accurate for girls. Examination of errors on numeral naming items revealed one discrepancy in responses: girls were more likely to respond “I don’t know” when presented with the numeral: 20.8% of kindergarten and first girls’ gave this response compared to only 9.7% of boys (χ2=10.26, p=.001).
Results suggest that gender differences in number sense are not apparent until kindergarten, possibly related to the introduction to formal math instruction. Furthermore, girls seem to struggle with the same types of problems (double-digit numerals) in kindergarten and first grade. The reasons for these emerging differences and their implication for practice require further investigation.

Authors

• Brianna Devlin, University of Delaware
  Presenting Author

• Haobai Zhang, University of Delaware
  Non-Presenting Author

• Amber Beliakoff, West Ed
  Non-Presenting Author

• Nancy Jordan, University of Delaware
  Non-Presenting Author

• Alice Klein, West Ed
  Non-Presenting Author