In dgromer/psymisc: Miscellaneous functions for psychologists

library(psymisc)
library(dplyr, warn.conflicts = FALSE)
library(ez)
library(tidyr)

Format test outputs according to APA guidelines

The *_apa functions help you to format test outputs according to APA guidelines. The functions take the return value of a test function as the first argument, e.g. a call to chisq.test. Supported tests are t-test (t.test and psymisc::t_test), ANOVA (ez::ezANOVA, afex::aov_car, afex::aov_ez and afex::aov_4), chi-squared-test (chisq.test) and test of a correlation (cor.test).

x <- chisq.test(c(20, 30))
x
chisq_apa(x)

The format argument allows you to specify the output format, which can be one of "text" (default), "markdown", "rmarkdown", "html", "latex", "docx" or "plotmath".

chisq_apa(x, format = "rmarkdown")
chisq_apa(x, format = "latex")

# Opens a temporary document in your word processor
chisq_apa(x, format = "docx")

No more copy-and-paste of single values or writing APA formatted test results by hand. You can simply copy the formatted test output into your manuscript or presentation. Available output format functions are t_apa, anova_apa, chisq_apa and cor_apa.

Descriptive statistics with ds and fplot

When beginning to analyze a data set, it's usually a good idea to take a look at descriptive statistics. Base R offers the aggregate function for computing summary statistics (e.g. mean, standard deviation) of factorial data.

aggregate(trait_anx ~ group + gender, hquest, FUN = "mean")

Computing a second statistic needs to be done using a second call to aggregate:

aggregate(trait_anx ~ group + gender, hquest, FUN = "sd")

A shortcoming of this function is that it does not allow you to compute multiple statistics at once. With the dplyr package and its functions group_by and summarize, this can be done more easily:

hquest %>%
  group_by(group, gender) %>%
  summarise(mean = mean(trait_anx), sd = sd(trait_anx))

The function ds in psymisc is a wrapper for these two functions and provides an easy formula interface for descriptive statistics:

ds(hquest, trait_anx ~ group + gender)

Note that ds calculates the mean and standard error of the mean by default. This behavior can be changed by using the funs argument:

ds(hquest, trait_anx ~ group + gender, funs = c("n", "mean", "sd", "moe"))

The fplot function can be used to get a graphical overview of descriptive statistics:

fplot(hquest, acrophobia ~ group + gender)
# With standard deviation instead of standard error
fplot(hquest, acrophobia ~ group + gender, error = "sd")
# Lines instead of bar plot
fplot(hquest, acrophobia ~ group + gender, geom = "line")
# Boxplot
fplot(hquest, acrophobia ~ group + gender, geom = "boxplot")

Tables with cor_table and stats_table

Similar to cor, cor_table computes a correlation matrix but in addition indicates significance with asterisks.

cor_table(hquest[4:7])
# HTML format
cor_table(hquest[4:7], format = "html")

For comparision of multiple groups of participants, the stats_table prints a table

stats_table(hquest, iv = group, dvs = age:sens_seek)
# HTML format opens in the RStudio Viewer pane and allow copy & paste to Word
stats_table(hquest, iv = group, dvs = age:sens_seek, format = "html")

Further functions

See a list of all psymisc functions at https://github.com/dgromer/psymisc or run ls("package:psymisc") in the R console after loading psymisc. Furthermore, the help page of each function lists usage examples.

dgromer/psymisc documentation built on May 15, 2019, 7:22 a.m.