Table 2.
Model comparison of the effects on difficulty ratings. The table reports a Likelihood ratio test between models where one predictor at a time was entered. AIC = Akaike Information Criterion. No.par = number of parameters of the model.
Predictors | No.par | AIC | logLik | LR.stat | df | Pr(>Chisq) |
---|---|---|---|---|---|---|
RT | 7 | 16,980 | −8483 | 20.70 | 1 | <.001 |
Condition | 10 | 16,982 | −8481 | 3.67 | 3 | .300 |
Cluster | 13 | 16,964 | −8469 | 23.61 | 3 | <.001 |
Condition × Cluster | 22 | 16,968 | −8462 | 14.08 | 9 | .120 |