Formatter functions in the 'apa' package take the return value of a statistical test function, e.g. a call to chisq.test() and return a string formatted according to the guidelines of the APA (American Psychological Association).

apa's functions format output of statistical tests according to guidelines of the APA (American Psychological Association), ready to copy-and-paste into manuscripts.

The idea of such formatters was introduced in the schoRsch package. apa generalizes this idea by providing formatters for different output formats (text, Markdown, RMarkdown, HTML, LaTeX, docx and R's plotmath syntax).

Currently available formatters are:

`anova_apa()`

^{2}`chisq_apa()`

`cor_apa()`

`t_apa()`

Further miscellaneous functions:

`apa()`

: A wrapper around the`*_apa()`

-functions for use in inline code in RMarkdown documents.`cohens_d()`

/`cohens_d_()`

: Calculate Cohen's d effect size (from raw data, t-test or statistical parameters). Also supports Hedge's g* and Glass's Δ.`t_test`

: A wrapper around`t.test()`

that includes the original data in its return list (in order to calculate the effect size in`cohens_d()`

and`t_apa()`

directly from the data).

^{1} pandoc is required for docx output and needs to be installed manually when not using RStudio (which ships pandoc).

^{2} Supports input from `aov()`

, `ezANOVA()`

from the ez package and `aov_ez()`

/ `aov_car()`

/ `aov_4()`

from the afex package.

The development version can be installed using:

`# install.packages("devtools")devtools::install_github("dgromer/apa")`

- Fix a test that returned a wrong result in r-devel (t-test now returns a list with more elements).

- Fix a bug in
`t_test`

when the independent variable has unused factor levels. - Fix a test that assumed no empty groups present (needed for dplyr 0.8

compatibility)

- Add LaTeX math output format (#3)

- Fix error in
`anova_apa`

when specifying the`effect`

argument - Fix printing of p-values if p = 1.
- Add missing
`else`

in`anova_apa`

. (@stegmannks, #6) - Fix error in calculation of sample size from degrees of freedom in Cohen's d for dependent samples (@lcreteig, #7)

- Add support for
`aov`

in`anova_apa`

.

- Fix bug when using abbreviations "pes" or "ges" in
`anova_apa`

. - Provide same order of effects in
`anova_apa`

independent of input object - In
`anova_apa`

significance asterisks might have been incorrect when p-values were corrected for violation of sphericity.