> fit4 <- glm(categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, + family = binomial(link = 'logit'), data = indvar_FODs) > summary(fit4) Call: glm(formula = categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, family = binomial(link = "logit"), data = indvar_FODs) Deviance Residuals: Min 1Q Median 3Q Max -1.315 -1.159 -1.111 1.191 1.268 Coefficients: Estimate Std. Error z value Pr(>|z|) Condition_FODsA:Gender_FODsFemale 0.06351 0.10292 0.617 0.537 Condition_FODsB:Gender_FODsFemale -0.04256 0.10317 -0.413 0.680 Condition_FODsC:Gender_FODsFemale -0.03175 0.10288 -0.309 0.758 Condition_FODsD:Gender_FODsFemale -0.10648 0.10329 -1.031 0.303 Condition_FODsA:Gender_FODsMale 0.07733 0.09837 0.786 0.432 Condition_FODsB:Gender_FODsMale -0.15937 0.09849 -1.618 0.106 Condition_FODsC:Gender_FODsMale -0.11496 0.09798 -1.173 0.241 Condition_FODsD:Gender_FODsMale 0.03865 0.09831 0.393 0.694 Condition_FODsA:Gender_FODsNon-Binary / Genderqueer 0.31845 0.32855 0.969 0.332 Condition_FODsB:Gender_FODsNon-Binary / Genderqueer 0.21131 0.32625 0.648 0.517 Condition_FODsC:Gender_FODsNon-Binary / Genderqueer 0.21131 0.32625 0.648 0.517 Condition_FODsD:Gender_FODsNon-Binary / Genderqueer -0.21131 0.32625 -0.648 0.517 (Dispersion parameter for binomial family taken to be 1) Null deviance: 4603.9 on 3321 degrees of freedom Residual deviance: 4595.2 on 3309 degrees of freedom (199 observations deleted due to missingness) AIC: 4619.2 Number of Fisher Scoring iterations: 3 > m2 <- clm(categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, + random = ~1|subject, data = indvar_FODs) > summary(m2) formula: categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1 data: indvar_FODs link threshold nobs logLik AIC niter max.grad cond.H logit flexible 3321 -2910.89 5847.79 6(0) 1.75e-12 1.4e+03 Coefficients: (1 not defined because of singularities) Estimate Std. Error z value Pr(>|z|) Condition_FODsA:Gender_FODsFemale 0.32677 0.33910 0.964 0.335 Condition_FODsB:Gender_FODsFemale 0.03530 0.33820 0.104 0.917 Condition_FODsC:Gender_FODsFemale 0.08655 0.33833 0.256 0.798 Condition_FODsD:Gender_FODsFemale 0.11344 0.33915 0.334 0.738 Condition_FODsA:Gender_FODsMale 0.30283 0.33753 0.897 0.370 Condition_FODsB:Gender_FODsMale -0.06136 0.33716 -0.182 0.856 Condition_FODsC:Gender_FODsMale -0.02626 0.33692 -0.078 0.938 Condition_FODsD:Gender_FODsMale 0.29300 0.33773 0.868 0.386 Condition_FODsA:Gender_FODsNon-Binary / Genderqueer 0.65051 0.45817 1.420 0.156 Condition_FODsB:Gender_FODsNon-Binary / Genderqueer 0.27418 0.44615 0.615 0.539 Condition_FODsC:Gender_FODsNon-Binary / Genderqueer 0.22538 0.44405 0.508 0.612 Condition_FODsD:Gender_FODsNon-Binary / Genderqueer NA NA NA NA Threshold coefficients: Estimate Std. Error z value ascending|descending 0.1625 0.3235 0.502 descending|identity 2.8620 0.3299 8.677 (199 observations deleted due to missingness) what is the difference between these two tests above? null code-to-explanation

Generation

code to explanationThu, 20 Apr 2023

> fit4 <- glm(categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, + family = binomial(link = 'logit'), data = indvar_FODs) > summary(fit4) Call: glm(formula = categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, family = binomial(link = "logit"), data = indvar_FODs) Deviance Residuals: Min 1Q Median 3Q Max -1.315 -1.159 -1.111 1.191 1.268 Coefficients: Estimate Std. Error z value Pr(>|z|) Condition_FODsA:Gender_FODsFemale 0.06351 0.10292 0.617 0.537 Condition_FODsB:Gender_FODsFemale -0.04256 0.10317 -0.413 0.680 Condition_FODsC:Gender_FODsFemale -0.03175 0.10288 -0.309 0.758 Condition_FODsD:Gender_FODsFemale -0.10648 0.10329 -1.031 0.303 Condition_FODsA:Gender_FODsMale 0.07733 0.09837 0.786 0.432 Condition_FODsB:Gender_FODsMale -0.15937 0.09849 -1.618 0.106 Condition_FODsC:Gender_FODsMale -0.11496 0.09798 -1.173 0.241 Condition_FODsD:Gender_FODsMale 0.03865 0.09831 0.393 0.694 Condition_FODsA:Gender_FODsNon-Binary / Genderqueer 0.31845 0.32855 0.969 0.332 Condition_FODsB:Gender_FODsNon-Binary / Genderqueer 0.21131 0.32625 0.648 0.517 Condition_FODsC:Gender_FODsNon-Binary / Genderqueer 0.21131 0.32625 0.648 0.517 Condition_FODsD:Gender_FODsNon-Binary / Genderqueer -0.21131 0.32625 -0.648 0.517 (Dispersion parameter for binomial family taken to be 1) Null deviance: 4603.9 on 3321 degrees of freedom Residual deviance: 4595.2 on 3309 degrees of freedom (199 observations deleted due to missingness) AIC: 4619.2 Number of Fisher Scoring iterations: 3 > m2 <- clm(categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1, + random = ~1|subject, data = indvar_FODs) > summary(m2) formula: categorical_FOD_FODs ~ 0 + Condition_FODs:Gender_FODs - 1 data: indvar_FODs link threshold nobs logLik AIC niter max.grad cond.H logit flexible 3321 -2910.89 5847.79 6(0) 1.75e-12 1.4e+03 Coefficients: (1 not defined because of singularities) Estimate Std. Error z value Pr(>|z|) Condition_FODsA:Gender_FODsFemale 0.32677 0.33910 0.964 0.335 Condition_FODsB:Gender_FODsFemale 0.03530 0.33820 0.104 0.917 Condition_FODsC:Gender_FODsFemale 0.08655 0.33833 0.256 0.798 Condition_FODsD:Gender_FODsFemale 0.11344 0.33915 0.334 0.738 Condition_FODsA:Gender_FODsMale 0.30283 0.33753 0.897 0.370 Condition_FODsB:Gender_FODsMale -0.06136 0.33716 -0.182 0.856 Condition_FODsC:Gender_FODsMale -0.02626 0.33692 -0.078 0.938 Condition_FODsD:Gender_FODsMale 0.29300 0.33773 0.868 0.386 Condition_FODsA:Gender_FODsNon-Binary / Genderqueer 0.65051 0.45817 1.420 0.156 Condition_FODsB:Gender_FODsNon-Binary / Genderqueer 0.27418 0.44615 0.615 0.539 Condition_FODsC:Gender_FODsNon-Binary / Genderqueer 0.22538 0.44405 0.508 0.612 Condition_FODsD:Gender_FODsNon-Binary / Genderqueer NA NA NA NA Threshold coefficients: Estimate Std. Error z value ascending|descending 0.1625 0.3235 0.502 descending|identity 2.8620 0.3299 8.677 (199 observations deleted due to missingness) what is the difference between these two tests above?

------------------------- **Annotated code** -------------------------