Greater mathematics content in a practice test improves subsequent mathematics performance by a small amount (~2% of a standard deviation per hour of practice)¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2–r3 · q3
Currency · newest study 2020 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Students randomly assigned to practice tests with more mathematics content scored higher on a subsequent mathematics test, with 1 hour of additional mathematics practice improving scores by about 2% of a standard deviation (b = 2.29, SE = 0.84, p = 0.007). → Avvisati 2020 q3 i? Both the 1-hour group (G2) and the 1.5-hour group (G3) outperformed the 0.5-hour group (G1) on the target mathematics task, with no evidence of non-linearity in the dose–response relationship (p = 0.178). → Avvisati 2020 (2)

Evidence¶

Avvisati 2020¶

Avvisati, F., & Borgonovi, F. (2020). Learning Mathematics Problem Solving through Test Practice: a Randomized Field Experiment on a Global Scale. Educational Psychology Review, 32, 791–814. https://doi.org/10.1007/s10648-020-09520-6

q3 · i? · causal · r2

Randomized field experiment within PISA 2012 across 32 countries (N = 19,355 in study 1). Students were randomly assigned to paper-based test forms containing 0.5, 1, or 1.5 hours of mathematics content within a 2-hour practice test. "1 h of additional problem-solving practice in mathematics improved students' ability to solve mathematics problems by about 2% of a standard deviation."

"1 h of additional problem-solving practice in mathematics improved students' ability to solve mathematics problems by about 2% of a standard deviation ( b = 2.29, SE = 0.84, p = 0.007 )"

Avvisati 2020 (2)¶

Avvisati, F., & Borgonovi, F. (2020). Learning Mathematics Problem Solving through Test Practice: a Randomized Field Experiment on a Global Scale. Educational Psychology Review, 32, 791–814. https://doi.org/10.1007/s10648-020-09520-6

q3 · i? · causal · r3

Same randomized field experiment, non-parametric group comparisons from model 2 in Table 3. "both G2 and G3 outperformed G1 on the target task" with the reported coefficients and standard errors; the linear dose–response assumption could not be rejected (p = 0.178).

"Mean comparisons that do not assume a functional form for the dose –response relationship confirmed that both G2 and G3 outperformed G1 on the target task ( bG2-G1 = 2.40, SEG2-G1 = 1.09; bG3-G1 = 2.41, SEG3-G1 = 0.87 )"

Discussion¶

Related Claims¶

• Mathematics test practice does not improve domain-general problem-solving ability — related
• Brief practice on non-symbolic approximate numerical addition or comparison tasks speeds subsequent exact symbolic arithmetic in first graders compared with non-numerical magnitude tasks — related
• Studying worked examples between two rounds of a computer puzzle game produced significantly greater knowledge map improvement than no worked examples for adult players — related