Unilevel OLS regression and four-level multilevel analysis of the same content analysis data yield differing p-values, especially for parameters that are not highly significant.¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2007 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Compared to the four-level model, unilevel regression p-values vary from 1.60 times smaller to 1.76 times larger, which might have led to erroneous conclusions. → De Wever 2007

Evidence¶

De Wever 2007¶

De Wever, B., Van Keer, H., Schellens, T., & Valcke, M. (2007). Applying multilevel modelling to content analysis data: Methodological issues in the study of role assignment in asynchronous discussion groups. Learning and Instruction, 17, 436-447. https://doi.org/10.1016/j.learninstruc.2007.04.001

q2 · i? · causal · r2

Comparison in Table 7 of parameter estimates and p-values from the four-level model C and a unilevel OLS regression on the same data; the authors report standard errors and most parameters are quite similar and both procedures lead to the same conclusions.

"Compared to the p-values of the four-l evel model, the p-values of the unilevel regression vary from 1.60 times smaller to 1.76 times larger."

Discussion¶

Related Claims¶

• The ML-LVM yields lower parameter bias and higher coverage rates for regression weights predicting learning outcomes than the PS two-step method — related