A Table of Models

including mapping between linear models and classical tests

Table 0.1: Linear models and extensions of linear models covered in this text.
Design Formula NHST Chapter
single factor with 2 levels, different mice in each treatment ~ treatment t-test 10.1.1
single factor with > 2 levels, different mice in each treatment ~ treatment one-way ANOVA 10.1.3
single factor with ≥ 2 levels and 1 continuous covariate, different mice in each treatment ~ treatment + weight ANCOVA 14
two factors, different mice in each treatment combination ~ treatment * genotype two way ANOVA 15
single factor with 2 levels, heterogeneity, different mice in each treatment ~ treatment, weights = varIdent(form = ~ 1 | treatment) Welch t-test 12.2
single factor with ≥ 2 levels, heterogeneity, different mice in each treatment ~ treatment, weights = varIdent(form = ~ 1 | treatment) multiple Welch t-tests 12.2
single factor with 2 levels, all treatments measured in each batch, n = 1 for each batch by treatment combination ~ treatment + (1 | id) paired t-test 12.1.1
single factor with > 2 levels, all treatments measured in each batch, n = 1 for each batch by treatment combination ~ treatment + (1 | id) repeated measures ANOVA (univariate model) 16.7
single factor with > 2 levels, all treatments measured in each batch, n = 1 for each batch by treatment combination ~ treatment + (1 | id), correlation = CorSymm(form = ~ 1 | id), weights = varIdent(form = ~ 1 | treatment) repeated measures ANOVA (multivariate model) 16.7
single factor with ≥ 2 levels, all treatments measured in each batch, n > 1 for each batch by treatment combination ~ treatment + (1 | experiment_id) + (1 | experiment_id:treatment) two-way mixed-effect ANOVA 16.6
single factor with ≥ 2 levels, all treatments measured in each batch, n > 1 for each batch by treatment combination ~ treatment + (treatment | experiment_id) none 16.6