This may be an {emmeans} question, but if I run a one-way repeated measures ANOVA and give it to estimate_contrasts(), is it doing paired t-tests for each pair? I'm confused by the dfs listed.
data("selfesteem", package = "datarium")
selfesteem
#> id t1 t2 t3
#> 1 1 4.005027 5.182286 7.107831
#> 2 2 2.558124 6.912915 6.308434
#> 3 3 3.244241 4.443434 9.778410
#> 4 4 3.419538 4.711696 8.347124
#> 5 5 2.871243 3.908429 6.457287
#> 6 6 2.045868 5.340549 6.653224
#> 7 7 3.525992 5.580695 6.840157
#> 8 8 3.179425 4.370234 7.818623
#> 9 9 3.507964 4.399808 8.471229
#> 10 10 3.043798 4.489376 8.581100
library(easystats)
#> # Attaching packages: easystats 0.6.0
#> ✔ bayestestR 0.13.1 ✔ correlation 0.8.4
#> ✔ datawizard 0.9.0 ✔ effectsize 0.8.6
#> ✔ insight 0.19.5 ✔ modelbased 0.8.6
#> ✔ performance 0.10.5 ✔ parameters 0.21.2
#> ✔ report 0.5.7 ✔ see 0.8.0
sel <- data_to_long(selfesteem, select = c(t1, t2, t3))
fit <- aov(value ~ name + Error(id), data = sel)
estimate_contrasts(fit, contrast = "name", method = "pairwise")
#> Warning: Could not recover model data from environment. Please make sure your
#> data is available in your workspace.
#> Trying to retrieve data from the model frame now.
#> Note: re-fitting model with sum-to-zero contrasts
#> Marginal Contrasts Analysis
#>
#> Level1 | Level2 | Difference | 95% CI | SE | t(26.00) | p
#> ------------------------------------------------------------------------
#> t1 | t2 | -1.79 | [-2.83, -0.76] | 0.40 | -4.45 | < .001
#> t1 | t3 | -4.50 | [-5.53, -3.46] | 0.40 | -11.16 | < .001
#> t2 | t3 | -2.70 | [-3.73, -1.67] | 0.40 | -6.71 | < .001
#>
#> Marginal contrasts estimated at name
#> p-value adjustment method: Holm (1979)
If each contrast is a paired $t$-test, then shouldn't its df be the number of paired differences minus 1 (e.g., 9 in this case)? Instead, it lists the df as 26 (presumably n - g - 1). Is this a mistake or am I misunderstanding something about how contrasts work?
This may be an {emmeans} question, but if I run a one-way repeated measures ANOVA and give it to
estimate_contrasts()
, is it doing paired t-tests for each pair? I'm confused by the dfs listed.Created on 2023-10-07 with reprex v2.0.2
If each contrast is a paired $t$-test, then shouldn't its df be the number of paired differences minus 1 (e.g., 9 in this case)? Instead, it lists the df as 26 (presumably n - g - 1). Is this a mistake or am I misunderstanding something about how contrasts work?