The instruction-by-time interaction favoring the simulation group was only marginally significant (one-tailed p = 0.085, g2 = 0.05), and the main effect of instruction was not significant¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1 · q2
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In a two-way mixed-design ANOVA, the instruction-by-time interaction was marginally significant with a one-tailed test at alpha = 0.10 (p = 0.085, g2 = 0.05), while the between-subject effect of instruction method was not significant. → Luo 2016

Evidence¶

Luo 2016¶

Luo, W., Pelletier, J., Duffin, K., Ormand, C., Hung, W., Shernoff, D., Zhai, X., Iverson, E., Whalley, K., Gallaher, C., and Furness, W. (2016). Advantages of Computer Simulation in Enhancing Students' Learning About Landform Evolution: A Case Study Using the Grand Canyon. Journal of Geoscience Education 64, 60-73. https://doi.org/10.5408/15-080.1

q2 · i? · causal · r1

Two-way mixed-design ANOVA on pretest-posttest scores of 43 students (20 simulation, 23 paper), with alpha set at 0.10 due to small sample. The article reports the interaction "marginally significant" with p = 0.085 and g2 = 0.05 under a one-tailed test.

"With a one-tailed test, the interaction is marginally significant (at a = 0.10), with p = 0.085 and a moderate effect size ( g2 = 0.05) (Cohen, 1988). Relative to the paper group, the simulation group showed moderately greater growth from pretest to posttest"

Luo 2016 (2)¶

Luo, W., Pelletier, J., Duffin, K., Ormand, C., Hung, W., Shernoff, D., Zhai, X., Iverson, E., Whalley, K., Gallaher, C., and Furness, W. (2016). Advantages of Computer Simulation in Enhancing Students' Learning About Landform Evolution: A Case Study Using the Grand Canyon. Journal of Geoscience Education 64, 60-73. https://doi.org/10.5408/15-080.1

q2 · i? · causal · r1

Same ANOVA: the between-subjects instruction effect was not significant, F(1, 41) = 0.04, p = 0.95, so combined mean scores did not differ significantly between the simulation and paper groups.

"The between-subject effect of intervention method was not significant [ F(1, 41) = 0.04, p = 0.95]; mean test scores (combined across pretest and posttest) were not significantly different between the two groups."

Discussion¶

Learner Variables¶

• Time and Continuity — moderator: an instructional effect differs with it

Related Claims¶

• HD-tDCS combined with mental rotation training shows no significant group effect or Group × Time interaction on accuracy, response time, or P300 GFP — related
• On higher-level application questions, percentage-correct growth was 15.0% in the simulation group versus 2.6% in the paper group — related
• In a quasiexperiment, the interactive simulation produced a large pretest-to-posttest improvement (Cohen's d = 1.06) while the paper-based exercise produced a small-to-moderate one (d = 0.40) — related