Description Usage Arguments Value Note References See Also Examples
View source: R/tadaa_pairwise_t.R
This is an extension of stats::pairwise.t.test that's meant to deal with interactions our of the box, while also performing pairwise tests for the primary terms. The output of the function is modeled after stats::TukeyHSD, unfortunately without confidence intervals or test statistic though.
1 2 3 4 5 6 7 8 9 10 11 |
data |
A |
response |
The response variable, i.e. the dependent numeric vector. |
group1 |
The grouping variables, typically a |
group2 |
(Optional) second grouping variable. |
p.adjust |
The p-adjustment method, see stats::p.adjust.methods, passed
to stats::pairwise.t.test. Additionally, |
paired |
Defaults to |
pool.sd |
Defaults to the inverse of |
alternative |
Defaults to |
print |
Print method, defaults to |
A data.frame
with columns term
, comparison
and adj.p.value
.
The adjustment method is applied within each term
, meaning that the number of
pairwise t-tests counted for the adjustment is only equal to the number of rows
per term
of the output.
The additional Sidak adjustment method uses the following method:
p_adj <- 1 - pbinom(q = 0, size = length(p_values), prob = p_values)
And is sometimes preferred over Bonferroni.
The Sidak-like (1987) step-down procedure (sidakSD
) is an improvement
over the Holm's (1979) step-down procedure.
https://stats.stackexchange.com/questions/20825/sidak-or-bonferroni
https://rdrr.io/rforge/mutoss/man/SidakSD.html
Other Tadaa-functions:
tadaa_aov()
,
tadaa_chisq()
,
tadaa_kruskal()
,
tadaa_levene()
,
tadaa_nom()
,
tadaa_one_sample()
,
tadaa_ord()
,
tadaa_pairwise_tukey()
,
tadaa_t.test()
,
tadaa_wilcoxon()
1 2 3 | tadaa_pairwise_t(ngo, deutsch, jahrgang, geschl, p.adjust = "none", print = "console")
tadaa_pairwise_t(ngo, deutsch, jahrgang, geschl, p.adjust = "bonf", print = "console")
tadaa_pairwise_t(ngo, deutsch, jahrgang, geschl, p.adjust = "sidak", print = "console")
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.