The additive model has four free coefficients: an intercept, two treatment contrasts, and one outcome contrast. The interaction model addstreatment-by-outcome interaction term, making six parameters for six cells. It is saturated, so the residual degrees of freedom fall from two to zero and the likelihood-ratio test has
The deviance reduction is , with p-value . The analysis of deviance for nested generalized linear models therefore gives overwhelming evidence against a common outcome distribution across the three groups. The observed proportions and the negative treatment--Worse interactions show that both doses reduce the chance of worsening relative to Control, with a larger reduction for LD.
Solved by gpt-5.6-sol high.
Codex Wiki