Models including task-dependent random effects fit CPS log data better than models with only fixed effects for all three planning indicators¶

Claim · All claims
Evidence · 1 study · associational r2 · q3
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? For all three indicators, models including both overall (fixed) and task-dependent random effects fit the data significantly better than models including only overall effects, according to likelihood ratio tests and AIC. → Eichmann 2019

Evidence¶

Eichmann 2019¶

Eichmann, B., Goldhammer, F., Greiff, S., Pucite, L., & Naumann, J. (2019). The role of planning in complex problem solving. Computers & Education, 128, 1-12. https://doi.org/10.1016/j.compedu.2018.08.004

q3 · i? · associational · r2

Likelihood ratio tests comparing nested GLMMs for each indicator on log data from N = 1346 German students on 21 interactive PISA 2012 CPS tasks showed all more complex models fit better (all p < 0.001), confirming that "the models including both overall ( fixed) and task- dependent random effects fit the data better."

"The model comparisons for research questions 1 –3 show that the models including both overall ( fixed) and task- dependent random effects fit the data better than models including only overall e ffects, as shown in Table 2. In all model comparisons the more complex model has the better fit according to the likelihood ratio test and AIC."

Discussion¶

Related Claims¶

• The duration of the longest planning interval shows no significant fixed effect on CPS performance but a significant task-dependent random effect — a narrower finding that bears on this claim
• Early planning in complex problem solving is generally beneficial, with a significant negative fixed effect of the delay indicator — related
• The variance of planning time shows no significant fixed effect on CPS performance but a significant task-dependent random effect correlated with task difficulty — related
• Significant interaction effects among duration, delay, and variance indicators show that the effects of planning behaviors on CPS performance are highly intertwined — related