STEM students show a higher mastery level and a higher learning gain in mathematics than non-STEM students.¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r1 · q2 · i2 medium
Currency · newest study 2018 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i2 Independent t-tests on the student profile show the STEM class has higher mean values on AveCarelessness, AveKnow and AveCorrect, with Cohen's d of -0.47, -0.45 and -0.44 respectively. → Yeung 2018 q2 i? A one-tailed hypothesis test on normalized learning gain computed from DKT+ knowledge states rejects the null hypothesis with a p-value of 0.0072, concluding STEM students have a higher knowledge gain in mathematics. → Yeung 2018 (2)

Evidence¶

Yeung 2018¶

Yeung, C.-K., Lin, Z., Yang, K., & Yeung, D.-Y. (2018). Incorporating Features Learned by an Enhanced Deep Knowledge Tracing Model for STEM/Non-STEM Job Prediction. arXiv preprint arXiv:1806.03256. https://arxiv.org/abs/1806.03256

q2 · i2 · associational · r1

Independent sample t-tests on student-profile attributes between STEM and non-STEM classes (Section 2.2, Table 1). The STEM class has higher means on AveKnow and AveCorrect; the printed Cohen's d values are -0.45 for AveKnow and -0.44 for AveCorrect, described as "a large magnitude in t-score and Cohen's d".

"Furthermore, “AveKnow”, “AveCorrect” and “AveCareless-ness” seem to be potent predictors, as they have a large magnitude in t-score and Cohen’s d as well as a small p-value, indicating the STEM and the non-STEM classes have different mean values in these attributes with a high confidence level."

Yeung 2018 (2)¶

Yeung, C.-K., Lin, Z., Yang, K., & Yeung, D.-Y. (2018). Incorporating Features Learned by an Enhanced Deep Knowledge Tracing Model for STEM/Non-STEM Job Prediction. arXiv preprint arXiv:1806.03256. https://arxiv.org/abs/1806.03256

q2 · i? · associational · r1

One-tailed hypothesis test on normalized learning gain adapted to DKT+ knowledge states (Section 5.2.3). The average NLG for STEM students is 0.065 (SD 0.172) and for non-STEM students 0.023 (SD 0.146); the null hypothesis is rejected "with a p-value of 0.0072".

"As a result, the null hypothesis H0 is rejected with a p-value of 0.0072. It is concluded that STEM students have a higher knowledge gain in mathematics than non-STEM students after the use of the ASSISTments, indicating the STEM class has a higher ability in learning mathematics."

Discussion¶

Related Claims¶

• Models trained with combined DKT/DKT+ knowledge-state and student-profile features generally outperform models trained with student profile alone on STEM/non-STEM job prediction. — related
• DKT+ knowledge states discriminate STEM from non-STEM students better than DKT knowledge states for the majority of mathematical skills. — related