30 January 2023 4 9K Report

I ran a likelihood ratio test in r and the result was as follows:

(I used GLMM)

model_full = glmer(response~condition*phase*trial + (1|ID), data=glmm, family=binomial, nAGQ=0)

mp2 = glmer(response~condition + trial + condition:phase + condition:trial + phase:trial + condition:phase:trial + (1|ID), data=glmm, family=binomial, nAGQ=0)

lrtest(model_full, mp2)

Likelihood ratio test

Model 1: response ~ condition * phase * trial + (1 | ID)

Model 2: response ~ condition + trial + condition:phase + condition:trial + phase:trial + condition:phase:trial + (1 | ID)

#Df LogLik Df Chisq Pr(>Chisq)

1 9 -227.1

2 9 -227.1 0 0 < 2.2e-16 ***

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

I understand that the p-value is < 0, which means the main effect of the phase significantly contributes to the model.

But I am confused about interpreting the degrees of freedom and Chi-square value.

More Jisoo Oh's questions See All
Similar questions and discussions