Title:	Convenience Functions for Psychology
Version:	0.1.9
Date:	2025-02-01
Description:	Make your workflow faster and easier. Easily customizable plots (via 'ggplot2'), nice APA tables (following the style of the *American Psychological Association*) exportable to Word (via 'flextable'), easily run statistical tests or check assumptions, and automatize various other tasks.
License:	GPL (≥ 3)
URL:	https://rempsyc.remi-theriault.com
BugReports:	https://github.com/rempsyc/rempsyc/issues
Depends:	R (≥ 3.6)
Imports:	rlang, dplyr (≥ 1.1.0)
Suggests:	flextable (≥ 0.9.1), ggplot2 (≥ 3.4.0), effectsize (≥ 0.8.5), performance (≥ 0.10.0), insight (≥ 0.18.4), correlation, datawizard (≥ 0.5.0), report (≥ 0.5.1), modelbased, see, lmtest, ggrepel, boot, bootES, ggsignif, qqplotr (≥ 0.0.6), broom, emmeans, ggpubr, interactions, openxlsx2 (≥ 0.8), patchwork, psych, VennDiagram, Rmisc, methods, tidyr, testthat (≥ 3.0.0), knitr, markdown, rmarkdown
VignetteBuilder:	knitr
Config/testthat/edition:	3
Encoding:	UTF-8
RoxygenNote:	7.3.2
NeedsCompilation:	no
Packaged:	2025-02-01 17:10:50 UTC; coding
Author:	Rémi Thériault [aut, cre]
Maintainer:	Rémi Thériault <remi.theriault@mail.mcgill.ca>
Repository:	CRAN
Date/Publication:	2025-02-01 23:40:05 UTC

rempsyc: Convenience Functions for Psychology

Description

Make your workflow faster and easier. Easily customizable plots (via 'ggplot2'), nice APA tables (following the style of the *American Psychological Association*) exportable to Word (via 'flextable'), easily run statistical tests or check assumptions, and automatize various other tasks.

Author(s)

Maintainer: Rémi Thériault remi.theriault@mail.mcgill.ca (ORCID)

See Also

Useful links:

• https://rempsyc.remi-theriault.com

• Report bugs at https://github.com/rempsyc/rempsyc/issues

Choose the best duplicate

Description

Chooses the best duplicate, based on the duplicate with the smallest number of missing values. In case of ties, it picks the first duplicate, as it is the one most likely to be valid and authentic, given practice effects.

Usage

best_duplicate(data, id, keep.rows = FALSE)

Arguments

Details

For the easystats equivalent, see: datawizard::data_duplicated().

Value

A dataframe, containing only the "best" duplicates.

Examples

df1 <- data.frame(
  id = c(1, 2, 3, 1, 3),
  item1 = c(NA, 1, 1, 2, 3),
  item2 = c(NA, 1, 1, 2, 3),
  item3 = c(NA, 1, 1, 2, 3)
)

best_duplicate(df1, id = "id", keep.rows = TRUE)

Easy export of correlation matrix to Excel

Description

Easily output a correlation matrix and export it to Microsoft Excel, with the first row and column frozen, and correlation coefficients colour-coded based on effect size (0.0-0.2: small (no colour); 0.2-0.4: medium (pink/light blue); 0.4-1.0: large (red/dark blue)), following Cohen's suggestions for small (.10), medium (.30), and large (.50) correlation sizes.

Based on the correlation and openxlsx2 packages.

Usage

cormatrix_excel(
  data,
  filename,
  overwrite = TRUE,
  p_adjust = "none",
  print.mat = TRUE,
  ...
)

Arguments

Details

For the easystats equivalent, see: correlation::cormatrix_to_excel().

Value

A Microsoft Excel document, containing the colour-coded correlation matrix with significance stars, on the first sheet, and the colour-coded p-values on the second sheet.

Author(s)

Adapted from @JanMarvin (JanMarvin/openxlsx2#286) and the original cormatrix_excel (now imported from correlation::cormatrix_to_excel).

Examples

# Basic example
cormatrix_excel(mtcars, select = c("mpg", "cyl", "disp", "hp", "carb"), filename = "cormatrix1")
cormatrix_excel(iris, p_adjust = "none", filename = "cormatrix2")
cormatrix_excel(airquality, method = "spearman", filename = "cormatrix3")

Extract all duplicates

Description

Extract all duplicates, for visual inspection. Note that it also contains the first occurrence of future duplicates, unlike duplicated() or dplyr::distinct()). Also contains an additional column reporting the number of missing values for that row, to help in the decision-making when selecting which duplicates to keep.

Usage

extract_duplicates(data, id)

Arguments

Details

For the easystats equivalent, see: datawizard::data_unique().

Value

A dataframe, containing all duplicates.

Examples

df1 <- data.frame(
  id = c(1, 2, 3, 1, 3),
  item1 = c(NA, 1, 1, 2, 3),
  item2 = c(NA, 1, 1, 2, 3),
  item3 = c(NA, 1, 1, 2, 3)
)

extract_duplicates(df1, id = "id")

# Filter to exclude duplicates
df2 <- df1[-c(1, 5), ]
df2

Identify outliers based on 3 MAD

Description

Identify outliers based on 3 median absolute deviations (MAD) from the median.

Usage

find_mad(data, col.list, ID = NULL, criteria = 3, mad.scores = TRUE)

Arguments

Details

The function internally use scale_mad() to "standardize" the data based on the MAD and median, and then check for any observation greater than the specified criteria (e.g., +/-3).

For the easystats equivalent, use: ⁠performance::check_outliers(x, method = "zscore_robust, threshold = 3)⁠.

Value

A list of dataframes of outliers per variable, with row numbers, based on the MAD. When printed, provides the number of outliers, selected variables, and any outlier flagged for more than one variable. More information can be obtainned by using the attributes() function around the generated object.

References

Leys, C., Ley, C., Klein, O., Bernard, P., & Licata, L. (2013). Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. Journal of Experimental Social Psychology, 49(4), 764–766. https://doi.org/10.1016/j.jesp.2013.03.013

Examples

find_mad(
  data = mtcars,
  col.list = names(mtcars),
  criteria = 3
)

mtcars2 <- mtcars
mtcars2$car <- row.names(mtcars)
find_mad(
  data = mtcars2,
  col.list = names(mtcars),
  ID = "car",
  criteria = 3
)

Easily format p or r values

Description

Easily format p or r values. Note: converts to character class for use in figures or manuscripts to accommodate e.g., "< .001".

Usage

format_value(value, type = "d", ...)

format_p(
  p,
  precision = 0.001,
  prefix = NULL,
  suffix = NULL,
  sign = FALSE,
  stars = FALSE
)

format_r(r, precision = 0.01)

format_d(d, precision = 0.01)

Arguments

Details

For the easystats equivalent, see: insight::format_value().

Value

A formatted p, r, or d value.

Examples

format_value(0.00041231, "p")
format_value(0.00041231, "r")
format_value(1.341231, "d")
format_p(0.0041231)
format_p(0.00041231)
format_r(0.41231)
format_r(0.041231)
format_d(1.341231)
format_d(0.341231)

Get required version of specified package dependency

Description

Get required version of specified package dependency

Usage

get_dep_version(dep, pkg = utils::packageName())

Arguments

Easy grouped bar charts for categorical variables

Description

Make nice grouped bar charts easily.

Usage

grouped_bar_chart(
  data,
  response,
  label = response,
  group = "T1_Group",
  proportion = TRUE,
  print_table = FALSE
)

Arguments

Value

A bar plot of class ggplot.

Examples

# Make the basic plot
iris2 <- iris
iris2$plant <- c(
  rep("yes", 45),
  rep("no", 45),
  rep("maybe", 30),
  rep("NA", 30)
)
grouped_bar_chart(
  data = iris2,
  response = "plant",
  group = "Species"
)

Install package if not already installed

Description

Install package if not already installed

Usage

install_if_not_installed(pkgs)

Arguments

Easy assumptions checks

Description

Test linear regression assumptions easily with a nice summary table.

Usage

nice_assumptions(model)

Arguments

Details

Interpretation: (p) values < .05 imply assumptions are not respected. Diagnostic is how many assumptions are not respected for a given model or variable.

Value

A dataframe, with p-value results for the Shapiro-Wilk, Breusch-Pagan, and Durbin-Watson tests, as well as a diagnostic column reporting how many assumptions are not respected for a given model. Shapiro-Wilk is set to NA if n < 3 or n > 5000.

See Also

Other functions useful in assumption testing: nice_density, nice_normality, nice_qq, nice_varplot, nice_var. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Create a regression model (using data available in R by default)
model <- lm(mpg ~ wt * cyl + gear, data = mtcars)
nice_assumptions(model)

# Multiple dependent variables at once
model2 <- lm(qsec ~ disp + drat * carb, mtcars)
my.models <- list(model, model2)
nice_assumptions(my.models)

Easy planned contrasts

Description

Easily compute planned contrast analyses (pairwise comparisons similar to t-tests but more powerful when more than 2 groups), and format in publication-ready format. In this particular case, the confidence intervals are bootstraped on chosen effect size (default to Cohen's d).

Usage

nice_contrasts(
  response,
  group,
  covariates = NULL,
  data,
  effect.type = "cohens.d",
  bootstraps = 2000,
  ...
)

Arguments

Details

Statistical power is lower with the standard t test compared than it is with the planned contrast version for two reasons: a) the sample size is smaller with the t test, because only the cases in the two groups are selected; and b) in the planned contrast the error term is smaller than it is with the standard t test because it is based on all the cases (source).

The effect size and confidence interval are calculated via bootES::bootES, and correct for contrasts but not for covariates and other predictors. Because this method uses bootstrapping, it is recommended to set a seed before using for reproducibility reasons (e.g., sed.seet(100)).

Does not for the moment support nested comparisons for marginal means, only a comparison of all groups. For nested comparisons, please use emmeans::contrast() directly, or for the easystats equivalent, modelbased::estimate_contrasts().

When using nice_lm_contrasts(), please use as.factor() outside the lm() formula, or it will lead to an error.

Value

A dataframe, with the selected dependent variable(s), comparisons of interest, degrees of freedom, t-values, p-values, Cohen's d, and the lower and upper 95% confidence intervals of the effect size (i.e., dR).

See Also

nice_lm_contrasts, Tutorial: https://rempsyc.remi-theriault.com/articles/contrasts

Examples

# Basic example
set.seed(100)
nice_contrasts(
  data = mtcars,
  response = "mpg",
  group = "cyl",
  bootstraps = 200
)

set.seed(100)
nice_contrasts(
  data = mtcars,
  response = "disp",
  group = "gear"
)

# Multiple dependent variables
set.seed(100)
nice_contrasts(
  data = mtcars,
  response = c("mpg", "disp", "hp"),
  group = "cyl"
)

# Adding covariates
set.seed(100)
nice_contrasts(
  data = mtcars,
  response = "mpg",
  group = "cyl",
  covariates = c("disp", "hp")
)

# Now supports more than 3 levels
mtcars2 <- mtcars
mtcars2$carb <- as.factor(mtcars2$carb)
set.seed(100)
nice_contrasts(
  data = mtcars,
  response = "mpg",
  group = "carb",
  bootstraps = 200
)

Easy density plots

Description

Make nice density plots easily. Internally, uses na.rm = TRUE.

Usage

nice_density(
  data,
  variable,
  group = NULL,
  colours,
  ytitle = "Density",
  xtitle = variable,
  groups.labels = NULL,
  grid = TRUE,
  shapiro = FALSE,
  title = variable,
  histogram = FALSE,
  breaks.auto = FALSE,
  bins = 30
)

Arguments

Value

A density plot of class ggplot, by group (if provided), along a reference line representing a matched normal distribution.

See Also

Other functions useful in assumption testing: nice_assumptions, nice_normality, nice_qq, nice_varplot, nice_var. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic plot
nice_density(
  data = iris,
  variable = "Sepal.Length",
  group = "Species"
)

# Further customization
nice_density(
  data = iris,
  variable = "Sepal.Length",
  group = "Species",
  colours = c("#00BA38", "#619CFF", "#F8766D"),
  xtitle = "Sepal Length",
  ytitle = "Density (vs. Normal Distribution)",
  groups.labels = c(
    "(a) Setosa",
    "(b) Versicolor",
    "(c) Virginica"
  ),
  grid = FALSE,
  shapiro = TRUE,
  title = "Density (Sepal Length)",
  histogram = TRUE
)

Nice formatting of lm models

Description

Formats output of lm() model object for a publication-ready format.

Usage

nice_lm(
  model,
  b.label = "b",
  standardize = FALSE,
  mod.id = TRUE,
  ci.alternative = "two.sided",
  ...
)

Arguments

Details

The effect size, sr2 (semi-partial correlation squared, also known as delta R2), is computed through effectsize::r2_semipartial. Please read the documentation for that function, especially regarding the interpretation of the confidence interval. In rempsyc, instead of using the default one-sided alternative ("greater"), we use the two-sided alternative.

To interpret the sr2, use effectsize::interpret_r2_semipartial().

For the easystats equivalent, use report::report() on the lm() model object.

Value

A formatted dataframe of the specified lm model, with DV, IV, degrees of freedom, regression coefficient, t-value, p-value, and the effect size, the semi-partial correlation squared, and its confidence interval.

See Also

Checking simple slopes after testing for moderation: nice_lm_slopes, nice_mod, nice_slopes. Tutorial: https://rempsyc.remi-theriault.com/articles/moderation

Examples

# Make and format model
model <- lm(mpg ~ cyl + wt * hp, mtcars)
nice_lm(model)

# Make and format multiple models
model2 <- lm(qsec ~ disp + drat * carb, mtcars)
my.models <- list(model, model2)
x <- nice_lm(my.models)
x


# Get interpretations
cbind(x, Interpretation = effectsize::interpret_r2_semipartial(x$sr2))

Easy planned contrasts using lm models

Description

Easily compute planned contrast analyses (pairwise comparisons similar to t-tests but more powerful when more than 2 groups), and format in publication-ready format. In this particular case, the confidence intervals are bootstraped on chosen effect size (default to Cohen's d).

Usage

nice_lm_contrasts(
  model,
  group,
  data,
  p_adjust = "none",
  effect.type = "cohens.d",
  bootstraps = 2000,
  ...
)

Arguments

Details

Statistical power is lower with the standard t test compared than it is with the planned contrast version for two reasons: a) the sample size is smaller with the t test, because only the cases in the two groups are selected; and b) in the planned contrast the error term is smaller than it is with the standard t test because it is based on all the cases (source).

The effect size and confidence interval are calculated via bootES::bootES, and correct for contrasts but not for covariates and other predictors. Because this method uses bootstrapping, it is recommended to set a seed before using for reproducibility reasons (e.g., sed.seet(100)).

Does not for the moment support nested comparisons for marginal means, only a comparison of all groups. For nested comparisons, please use emmeans::contrast() directly, or for the easystats equivalent, modelbased::estimate_contrasts().

When using nice_lm_contrasts(), please use as.factor() outside the lm() formula, or it will lead to an error.

Value

A dataframe, with the selected dependent variable(s), comparisons of interest, degrees of freedom, t-values, p-values, Cohen's d, and the lower and upper 95% confidence intervals of the effect size (i.e., dR).

See Also

nice_contrasts, Tutorial: https://rempsyc.remi-theriault.com/articles/contrasts

Examples

# Make and format model (group need to be a factor)
mtcars2 <- mtcars
mtcars2$cyl <- as.factor(mtcars2$cyl)
model <- lm(mpg ~ cyl + wt * hp, mtcars2)
set.seed(100)
nice_lm_contrasts(model, group = "cyl", data = mtcars, bootstraps = 500)

# Several models at once
mtcars2$gear <- as.factor(mtcars2$gear)
model2 <- lm(qsec ~ cyl, data = mtcars2)
my.models <- list(model, model2)
set.seed(100)
nice_lm_contrasts(my.models, group = "cyl", data = mtcars, bootstraps = 500)

# Now supports more than 3 levels
mtcars2$carb <- as.factor(mtcars2$carb)
model <- lm(mpg ~ carb + wt * hp, mtcars2)
set.seed(100)
nice_lm_contrasts(model, group = "carb", data = mtcars2, bootstraps = 500)

Nice formatting of simple slopes for lm models

Description

Extracts simple slopes from lm() model object and format for a publication-ready format.

Usage

nice_lm_slopes(
  model,
  predictor,
  moderator,
  b.label = "b",
  standardize = FALSE,
  mod.id = TRUE,
  ci.alternative = "two.sided",
  ...
)

Arguments

Details

The effect size, sr2 (semi-partial correlation squared, also known as delta R2), is computed through effectsize::r2_semipartial. Please read the documentation for that function, especially regarding the interpretation of the confidence interval. In rempsyc, instead of using the default one-sided alternative ("greater"), we use the two-sided alternative.

To interpret the sr2, use effectsize::interpret_r2_semipartial().

For the easystats equivalent, use report::report() on the lm() model object.

Value

A formatted dataframe of the simple slopes of the specified lm model, with DV, levels of IV, degrees of freedom, regression coefficient, t-value, p-value, and the effect size, the semi-partial correlation squared, and its confidence interval.

See Also

Checking for moderation before checking simple slopes: nice_lm, nice_mod, nice_slopes. Tutorial: https://rempsyc.remi-theriault.com/articles/moderation

Examples

# Make and format model
model <- lm(mpg ~ gear * wt, mtcars)
nice_lm_slopes(model, predictor = "gear", moderator = "wt")

# Make and format multiple models
model2 <- lm(qsec ~ gear * wt, mtcars)
my.models <- list(model, model2)
x <- nice_lm_slopes(my.models, predictor = "gear", moderator = "wt")
x


# Get interpretations
cbind(x, Interpretation = effectsize::interpret_r2_semipartial(x$sr2))

Easy moderations

Description

Easily compute moderation analyses, with effect sizes, and format in publication-ready format.

Usage

nice_mod(
  data,
  response,
  predictor,
  moderator,
  moderator2 = NULL,
  covariates = NULL,
  b.label = "b",
  standardize = TRUE,
  mod.id = TRUE,
  ci.alternative = "two.sided",
  ...
)

Arguments

Details

The effect size, sr2 (semi-partial correlation squared, also known as delta R2), is computed through effectsize::r2_semipartial. Please read the documentation for that function, especially regarding the interpretation of the confidence interval. In rempsyc, instead of using the default one-sided alternative ("greater"), we use the two-sided alternative.

To interpret the sr2, use effectsize::interpret_r2_semipartial().

For the easystats equivalent, use report::report() on the lm() model object.

Value

A formatted dataframe of the specified lm model, with DV, IV, degrees of freedom, regression coefficient, t-value, p-value, and the effect size, the semi-partial correlation squared, and its confidence interval.

See Also

Checking simple slopes after testing for moderation: nice_slopes, nice_lm, nice_lm_slopes. Tutorial: https://rempsyc.remi-theriault.com/articles/moderation

Examples

# Make the basic table
nice_mod(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt"
)

# Multiple dependent variables at once
nice_mod(
  data = mtcars,
  response = c("mpg", "disp", "hp"),
  predictor = "gear",
  moderator = "wt"
)

# Add covariates
nice_mod(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt",
  covariates = c("am", "vs")
)

# Three-way interaction
x <- nice_mod(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt",
  moderator2 = "am"
)
x


# Get interpretations
cbind(x, Interpretation = effectsize::interpret_r2_semipartial(x$sr2))

Report missing values according to guidelines

Description

Nicely reports NA values according to existing guidelines. This function reports both absolute and percentage values of specified column lists. Some authors recommend reporting item-level missing item per scale, as well as participant’s maximum number of missing items by scale. For example, Parent (2013) writes:

I recommend that authors (a) state their tolerance level for missing data by scale or subscale (e.g., “We calculated means for all subscales on which participants gave at least 75% complete data”) and then (b) report the individual missingness rates by scale per data point (i.e., the number of missing values out of all data points on that scale for all participants) and the maximum by participant (e.g., “For Attachment Anxiety, a total of 4 missing data points out of 100 were observed, with no participant missing more than a single data point”).

Usage

nice_na(data, vars = NULL, scales = NULL)

Arguments

Value

A dataframe, with:

• var: variables selected

• items: number of items for selected variables

• na: number of missing cell values for those variables (e.g., 2 missing values for first participant + 2 missing values for second participant = total of 4 missing values)

• cells: total number of cells (i.e., number of participants multiplied by number of variables, items)

• na_percent: the percentage of missing values (number of missing cells, na, divided by total number of cells, cells)

• na_max: The amount of missing values for the participant with the most missing values for the selected variables

• na_max_percent: The amount of missing values for the participant with the most missing values for the selected variables, in percentage (i.e., na_max divided by the number of selected variables, items)

• all_na: the number of participants missing 100% of items for that scale (the selected variables)

References

Parent, M. C. (2013). Handling item-level missing data: Simpler is just as good. The Counseling Psychologist, 41(4), 568-600. https://doi.org/10.1177%2F0011000012445176

Examples

# Use whole data frame
nice_na(airquality)

# Use selected columns explicitly
nice_na(airquality,
  vars = list(
    c("Ozone", "Solar.R", "Wind"),
    c("Temp", "Month", "Day")
  )
)

# If the questionnaire items start with the same name, e.g.,
set.seed(15)
fun <- function() {
  c(sample(c(NA, 1:10), replace = TRUE), NA, NA, NA)
}
df <- data.frame(
  ID = c("idz", NA),
  open_1 = fun(), open_2 = fun(), open_3 = fun(),
  extrovert_1 = fun(), extrovert_2 = fun(), extrovert_3 = fun(),
  agreeable_1 = fun(), agreeable_2 = fun(), agreeable_3 = fun()
)

# One can list the scale names directly:
nice_na(df, scales = c("ID", "open", "extrovert", "agreeable"))

Easy normality check per group

Description

Easily make nice per-group density and QQ plots through a wrapper around the ggplot2 and qqplotr packages.

Usage

nice_normality(
  data,
  variable,
  group = NULL,
  colours,
  groups.labels,
  grid = TRUE,
  shapiro = FALSE,
  title = NULL,
  histogram = FALSE,
  breaks.auto = FALSE,
  ...
)

Arguments

Value

A plot of classes patchwork and ggplot, containing two plots, resulting from nice_density and nice_qq.

See Also

Other functions useful in assumption testing: nice_assumptions, nice_density, nice_qq, nice_var, nice_varplot. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic plot
nice_normality(
  data = iris,
  variable = "Sepal.Length",
  group = "Species"
)

# Further customization
nice_normality(
  data = iris,
  variable = "Sepal.Length",
  group = "Species",
  colours = c(
    "#00BA38",
    "#619CFF",
    "#F8766D"
  ),
  groups.labels = c(
    "(a) Setosa",
    "(b) Versicolor",
    "(c) Virginica"
  ),
  grid = FALSE,
  shapiro = TRUE
)

Easy QQ plots per group

Description

Easily make nice per-group QQ plots through a wrapper around the ggplot2 and qqplotr packages.

Usage

nice_qq(
  data,
  variable,
  group = NULL,
  colours,
  groups.labels = NULL,
  grid = TRUE,
  shapiro = FALSE,
  title = variable
)

Arguments

Value

A qq plot of class ggplot, by group (if provided), along a reference interpretation helper, the 95% confidence band.

See Also

Other functions useful in assumption testing: nice_assumptions, nice_density, nice_normality, nice_var, nice_varplot. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic plot
nice_qq(
  data = iris,
  variable = "Sepal.Length",
  group = "Species"
)

# Further customization
nice_qq(
  data = iris,
  variable = "Sepal.Length",
  group = "Species",
  colours = c("#00BA38", "#619CFF", "#F8766D"),
  groups.labels = c("(a) Setosa", "(b) Versicolor", "(c) Virginica"),
  grid = FALSE,
  shapiro = TRUE,
  title = NULL
)

Easily randomization

Description

Randomize easily with different designs.

Usage

nice_randomize(
  design = "between",
  Ncondition = 3,
  n = 9,
  condition.names = c("a", "b", "c"),
  col.names = c("id", "Condition")
)

Arguments

Value

A dataframe, with participant ID and randomized condition, based on selected design.

See Also

Tutorial: https://rempsyc.remi-theriault.com/articles/randomize

Examples

# Specify design, number of conditions, number of
# participants, and names of conditions:
nice_randomize(
  design = "between", Ncondition = 4, n = 8,
  condition.names = c("BP", "CX", "PZ", "ZL")
)

# Within-Group Design
nice_randomize(
  design = "within", Ncondition = 4, n = 6,
  condition.names = c("SV", "AV", "ST", "AT")
)

# Make a quick runsheet
randomized <- nice_randomize(
  design = "within", Ncondition = 4, n = 128,
  condition.names = c("SV", "AV", "ST", "AT"),
  col.names = c(
    "id", "Condition", "Date/Time",
    "SONA ID", "Age/Gd.", "Handedness",
    "Tester", "Notes"
  )
)
head(randomized)

Easily recode scores

Description

Easily recode scores (reverse-score), typically for questionnaire answers.

For the easystats equivalent, see: datawizard::reverse().

Usage

nice_reverse(x, max, min = 1)

Arguments

Value

A numeric vector, of reversed scores.

Examples

# Reverse score of 5 with a maximum score of 5
nice_reverse(5, 5)

# Reverse several scores at once
nice_reverse(1:5, 5)

# Reverse scores with maximum = 4 and minimum = 0
nice_reverse(1:4, 4, min = 0)

# Reverse scores with maximum = 3 and minimum = -3
nice_reverse(-3:3, 3, min = -3)

Easy scatter plots

Description

Make nice scatter plots easily.

Usage

nice_scatter(
  data,
  predictor,
  response,
  xtitle = predictor,
  ytitle = response,
  has.points = TRUE,
  has.jitter = FALSE,
  alpha = 0.7,
  has.line = TRUE,
  method = "lm",
  has.confband = FALSE,
  has.fullrange = FALSE,
  has.linetype = FALSE,
  has.shape = FALSE,
  xmin,
  xmax,
  xby = 1,
  ymin,
  ymax,
  yby = 1,
  has.legend = FALSE,
  legend.title = "",
  group = NULL,
  colours = "#619CFF",
  groups.order = "none",
  groups.labels = NULL,
  groups.alpha = NULL,
  has.r = FALSE,
  r.x = Inf,
  r.y = -Inf,
  has.p = FALSE,
  p.x = Inf,
  p.y = -Inf
)

Arguments

Value

A scatter plot of class ggplot.

See Also

Visualize group differences via violin plots: nice_violin. Tutorial: https://rempsyc.remi-theriault.com/articles/scatter

Examples

# Make the basic plot
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg"
)


# Save a high-resolution image file to specified directory
ggplot2::ggsave("nicescatterplothere.pdf", width = 7,
  height = 7, unit = "in", dpi = 300
) # change for your own desired path

# Change x- and y- axis labels
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  ytitle = "Miles/(US) gallon",
  xtitle = "Weight (1000 lbs)"
)

# Have points "jittered", loess method
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  has.jitter = TRUE,
  method = "loess"
)

# Change the transparency of the points
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  alpha = 1
)

# Remove points
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  has.points = FALSE,
  has.jitter = FALSE
)

# Add confidence band
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  has.confband = TRUE
)

# Set x- and y- scales manually
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  xmin = 1,
  xmax = 6,
  xby = 1,
  ymin = 10,
  ymax = 35,
  yby = 5
)

# Change plot colour
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  colours = "blueviolet"
)

# Add correlation coefficient to plot and p-value
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  has.r = TRUE,
  has.p = TRUE
)

# Change location of correlation coefficient or p-value
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  has.r = TRUE,
  r.x = 4,
  r.y = 25,
  has.p = TRUE,
  p.x = 5,
  p.y = 20
)

# Plot by group
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl"
)

# Use full range on the slope/confidence band
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  has.fullrange = TRUE
)

# Remove lines
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  has.line = FALSE
)

# Change order of labels on the legend
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  groups.order = c(8, 4, 6)
)

# Change legend labels
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  groups.labels = c("Weak", "Average", "Powerful")
)
# Warning: This applies after changing order of level

# Add a title to legend
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  legend.title = "cylinders"
)

# Plot by group + manually specify colours
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  colours = c("burlywood", "darkgoldenrod", "chocolate")
)

# Plot by group + use different line types for each group
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  has.linetype = TRUE
)

# Plot by group + use different point shapes for each group
nice_scatter(
  data = mtcars,
  predictor = "wt",
  response = "mpg",
  group = "cyl",
  has.shape = TRUE
)

Easy simple slopes

Description

Easily compute simple slopes in moderation analysis, with effect sizes, and format in publication-ready format.

Usage

nice_slopes(
  data,
  response,
  predictor,
  moderator,
  moderator2 = NULL,
  covariates = NULL,
  b.label = "b",
  standardize = TRUE,
  mod.id = TRUE,
  ci.alternative = "two.sided",
  ...
)

Arguments

Details

The effect size, sr2 (semi-partial correlation squared, also known as delta R2), is computed through effectsize::r2_semipartial. Please read the documentation for that function, especially regarding the interpretation of the confidence interval. In rempsyc, instead of using the default one-sided alternative ("greater"), we use the two-sided alternative.

To interpret the sr2, use effectsize::interpret_r2_semipartial().

For the easystats equivalent, use report::report() on the lm() model object.

Value

A formatted dataframe of the simple slopes of the specified lm model, with DV, levels of IV, degrees of freedom, regression coefficient, t-value, p-value, and the effect size, the semi-partial correlation squared, and its confidence interval.

See Also

Checking for moderation before checking simple slopes: nice_mod, nice_lm, nice_lm_slopes. Tutorial: https://rempsyc.remi-theriault.com/articles/moderation

Examples

# Make the basic table
nice_slopes(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt"
)

# Multiple dependent variables at once
nice_slopes(
  data = mtcars,
  response = c("mpg", "disp", "hp"),
  predictor = "gear",
  moderator = "wt"
)

# Add covariates
nice_slopes(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt",
  covariates = c("am", "vs")
)

# Three-way interaction (continuous moderator and binary
# second moderator required)
x <- nice_slopes(
  data = mtcars,
  response = "mpg",
  predictor = "gear",
  moderator = "wt",
  moderator2 = "am"
)
x


# Get interpretations
cbind(x, Interpretation = effectsize::interpret_r2_semipartial(x$sr2))

Easy t-tests

Description

Easily compute t-test analyses, with effect sizes, and format in publication-ready format. The 95% confidence interval is for the effect size, Cohen's d, both provided by the effectsize package.

Usage

nice_t_test(
  data,
  response,
  group = NULL,
  correction = "none",
  paired = FALSE,
  verbose = TRUE,
  ...
)

Arguments

Details

This function relies on the base R t.test() function, which uses the Welch t-test per default (see why here: https://daniellakens.blogspot.com/2015/01/always-use-welchs-t-test-instead-of.html). To use the Student t-test, simply add the following argument: var.equal = TRUE.

Note that for paired t tests, you need to use paired = TRUE, and you also need data in "long" format rather than wide format (like for the ToothGrowth data set). In this case, the group argument refers to the participant ID for example, so the same group/participant is measured several times, and thus has several rows. Note also that R >= 4.4.0 has stopped supporting the paired argument for the formula method used internally here.

For the easystats equivalent, use: report::report() on the t.test() object.

Value

A formatted dataframe of the specified model, with DV, degrees of freedom, t-value, p-value, the effect size, Cohen's d, and its 95% confidence interval lower and upper bounds.

See Also

Tutorial: https://rempsyc.remi-theriault.com/articles/t-test

Examples

# Make the basic table
nice_t_test(
  data = mtcars,
  response = "mpg",
  group = "am"
)

# Multiple dependent variables at once
nice_t_test(
  data = mtcars,
  response = names(mtcars)[1:7],
  group = "am"
)

# Can be passed some of the regular arguments
# of base [t.test()]

# Student t-test (instead of Welch)
nice_t_test(
  data = mtcars,
  response = "mpg",
  group = "am",
  var.equal = TRUE
)

# One-sided instead of two-sided
nice_t_test(
  data = mtcars,
  response = "mpg",
  group = "am",
  alternative = "less"
)

# One-sample t-test
nice_t_test(
  data = mtcars,
  response = "mpg",
  mu = 10
)

# Make sure cases appear in the same order for
# both levels of the grouping factor

Easily make nice APA tables

Description

Make nice APA tables easily through a wrapper around the flextable package with sensical defaults and automatic formatting features.

Usage

nice_table(
  data,
  highlight = FALSE,
  stars = TRUE,
  italics,
  col.format.p,
  col.format.r,
  col.format.ci,
  format.custom,
  col.format.custom,
  width = NULL,
  spacing = 2,
  broom = NULL,
  report = NULL,
  short = FALSE,
  title,
  note,
  separate.header
)

Arguments

Details

The resulting flextable objects can be opened in Word with print(table, preview ="docx"), or saved to Word with the flextable::save_as_docx() function.

Value

An APA-formatted table of class "flextable"

See Also

Tutorial: https://rempsyc.remi-theriault.com/articles/table

Examples

# Make the basic table
my_table <- nice_table(
  mtcars[1:3, ],
  title = c("Table 1", "Motor Trend Car Road Tests"),
  note = c(
    "The data was extracted from the 1974 Motor Trend US magazine.",
    "* p < .05, ** p < .01, *** p < .001"
  )
)
my_table


# Save table to word
mypath <- tempfile(fileext = ".docx")
flextable::save_as_docx(my_table, path = mypath)


# Publication-ready tables
mtcars.std <- lapply(mtcars, scale)
model <- lm(mpg ~ cyl + wt * hp, mtcars.std)
stats.table <- as.data.frame(summary(model)$coefficients)
CI <- confint(model)
stats.table <- cbind(
  row.names(stats.table),
  stats.table, CI
)
names(stats.table) <- c(
  "Term", "B", "SE", "t", "p",
  "CI_lower", "CI_upper"
)
nice_table(stats.table, highlight = TRUE)

# Test different column names
test <- head(mtcars)
names(test) <- c(
  "dR", "N", "M", "SD", "b", "np2",
  "ges", "p", "r", "R2", "sr2"
)
test[, 10:11] <- test[, 10:11] / 10
nice_table(test)

# Custom cell formatting (such as p or r)
nice_table(test[8:11], col.format.p = 2:4, highlight = .001)

nice_table(test[8:11], col.format.r = 1:4)

# Apply custom functions to cells
fun <- function(x) {
  x + 11.1
}
nice_table(test[8:11], col.format.custom = 2:4, format.custom = "fun")

fun <- function(x) {
  paste("x", x)
}
nice_table(test[8:11], col.format.custom = 2:4, format.custom = "fun")

# Separate headers based on periods
header.data <- structure(
  list(
    Variable = c(
      "Sepal.Length",
      "Sepal.Width", "Petal.Length"
    ), setosa.M = c(
      5.01, 3.43,
      1.46
    ), setosa.SD = c(0.35, 0.38, 0.17), versicolor.M =
      c(5.94, 2.77, 4.26), versicolor.SD = c(0.52, 0.31, 0.47)
  ),
  row.names = c(NA, -3L), class = "data.frame"
)
nice_table(header.data,
  separate.header = TRUE,
  italics = 2:4
)

Obtain variance per group

Description

Obtain variance per group as well as check for the rule of thumb of one group having variance four times bigger than any of the other groups. Variance ratio is calculated as Max / Min.

Usage

nice_var(data, variable, group, criteria = 4)

Arguments

Value

A dataframe, with the values of the selected variables for each group, their max variance ratio (maximum variance divided by the minimum variance), the selected decision criterion, and whether the data are considered heteroscedastic according to the decision criterion.

See Also

Other functions useful in assumption testing: nice_assumptions, nice_density, nice_normality, nice_qq, nice_varplot. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic table
nice_var(
  data = iris,
  variable = "Sepal.Length",
  group = "Species"
)

# Try on multiple variables
nice_var(
  data = iris,
  variable = names(iris[1:4]),
  group = "Species"
)

Attempt to visualize variance per group

Description

Attempt to visualize variance per group.

Usage

nice_varplot(
  data,
  variable,
  group,
  colours,
  groups.labels,
  grid = TRUE,
  shapiro = FALSE,
  ytitle = variable
)

Arguments

Value

A scatter plot of class ggplot attempting to display the group variances. Also includes the max variance ratio (maximum variance divided by the minimum variance).

See Also

Other functions useful in assumption testing: nice_assumptions, nice_density, nice_normality, nice_qq, nice_var. Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic plot
nice_varplot(
  data = iris,
  variable = "Sepal.Length",
  group = "Species"
)

# Further customization
nice_varplot(
  data = iris,
  variable = "Sepal.Length",
  group = "Species",
  colours = c(
    "#00BA38",
    "#619CFF",
    "#F8766D"
  ),
  ytitle = "Sepal Length",
  groups.labels = c(
    "(a) Setosa",
    "(b) Versicolor",
    "(c) Virginica"
  )
)

Easy violin plots

Description

Make nice violin plots easily with 95% (possibly bootstrapped) confidence intervals.

Usage

nice_violin(
  data,
  response,
  group = NULL,
  boot = FALSE,
  bootstraps = 2000,
  colours,
  xlabels = NULL,
  ytitle = response,
  xtitle = NULL,
  has.ylabels = TRUE,
  has.xlabels = TRUE,
  comp1 = 1,
  comp2 = 2,
  signif_annotation = NULL,
  signif_yposition = NULL,
  signif_xmin = NULL,
  signif_xmax = NULL,
  ymin,
  ymax,
  yby = 1,
  CIcap.width = 0.1,
  obs = FALSE,
  alpha = 1,
  border.colour = "black",
  border.size = 2,
  has.d = FALSE,
  d.x = mean(c(comp1, comp2)) * 1.1,
  d.y = mean(data[[response]]) * 1.3,
  groups.order = "none",
  xlabels.angle = 0
)

Arguments

Details

Using boot = TRUE uses bootstrapping (for the confidence intervals only) with the BCa method, using the rcompanion_groupwiseMean function.

For the easystats equivalent, see: see::geom_violindot().

Value

A violin plot of class ggplot, by group.

See Also

Visualize group differences via scatter plots: nice_scatter. Tutorial: https://rempsyc.remi-theriault.com/articles/violin

Examples

# Make the basic plot
nice_violin(
  data = ToothGrowth,
  response = "len"
)


# Save a high-resolution image file to specified directory
ggplot2::ggsave("niceviolinplothere.pdf", width = 7,
  height = 7, unit = "in", dpi = 300
) # change for your own desired path

# Change x- and y- axes labels
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  ytitle = "Length of Tooth",
  xtitle = "Vitamin C Dosage"
)

# See difference between two groups
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  comp1 = "0.5",
  comp2 = "2"
)

nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  comp1 = 2,
  comp2 = 3
)

# Compare all three groups
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  signif_annotation = c("*", "**", "***"),
  # manually enter the number of stars
  signif_yposition = c(30, 35, 40),
  # What height (y) should the stars appear
  signif_xmin = c(1, 2, 1),
  # Where should the left-sided brackets start (x)
  signif_xmax = c(2, 3, 3)
)
# Where should the right-sided brackets end (x)

# Set the colours manually
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  colours = c("darkseagreen", "cadetblue", "darkslateblue")
)

# Changing the names of the x-axis labels
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  xlabels = c("Low", "Medium", "High")
)

# Removing the x-axis or y-axis titles
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  ytitle = NULL,
  xtitle = NULL
)

# Removing the x-axis or y-axis labels (for whatever purpose)
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  has.ylabels = FALSE,
  has.xlabels = FALSE
)

# Set y-scale manually
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  ymin = 5,
  ymax = 35,
  yby = 5
)

# Plotting individual observations
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  obs = TRUE
)

# Micro-customizations
nice_violin(
  data = ToothGrowth,
  group = "dose",
  response = "len",
  CIcap.width = 0,
  alpha = .70,
  border.size = 1,
  border.colour = "white",
  comp1 = 1,
  comp2 = 2,
  has.d = TRUE
)

Interpolate the Inclusion of the Other in the Self Scale

Description

Interpolating the Inclusion of the Other in the Self Scale (IOS; self-other merging) easily. The user provides the IOS score, from 1 to 7, and the function will provide a percentage of actual area overlap between the two circles (i.e., not linear overlap), so it is possible to say, e.g., that experimental group 1 had an average overlap of X% with the other person, whereas experimental group 2 had an average overlap of X% with the other person.

Usage

overlap_circle(response, categories = c("Self", "Other"), scoring = "IOS")

Arguments

Details

The circles are generated through the VennDiagram::draw.pairwise.venn() function and the desired percentage overlap is passed to its cross.area argument ("The size of the intersection between the sets"). The percentage overlap values are interpolated from this reference grid: Score of 1 = 0%, 2 = 10%, 3 = 20%, 4 = 30%, 5 = 55%, 6 = 65%, 7 = 85%.

Value

A plot of class gList, displaying overlapping circles relative to the selected score.

See Also

Tutorial: https://rempsyc.remi-theriault.com/articles/circles

For a javascript web plugin of a continuous version of the Inclusion of Other in the Self (IOS) task (instead of the pen and paper version), for experiments during data collection, rather than data analysis, please see: https://github.com/jspsych/jspsych-contrib/tree/main/packages/plugin-ios

Examples

# Score of 1 (0% overlap)
overlap_circle(1)

# Score of 3.5 (25% overlap)
overlap_circle(3.5)

# Score of 6.84 (81.8% overlap)
overlap_circle(6.84)

# Changing labels
overlap_circle(3.12, categories = c("Humans", "Animals"))


# Saving to file (PDF or PNG)
plot <- overlap_circle(3.5)
ggplot2::ggsave(plot,
  file = tempfile(fileext = ".pdf"), width = 7,
  height = 7, unit = "in", dpi = 300
)
# Change for your own desired path

Easy scatter plots over multiple times (T1, T2, T3)

Description

Make nice scatter plots over multiple times (T1, T2, T3) easily.

Usage

plot_means_over_time(
  data,
  response,
  group,
  groups.order = "none",
  error_bars = TRUE,
  ytitle = NULL,
  legend.title = "",
  significance_stars,
  significance_stars_x,
  significance_stars_y,
  significance_bars_x,
  print_table = FALSE,
  verbose = FALSE
)

Arguments

Details

Error bars are calculated using the method of Morey (2008) through Rmisc::summarySEwithin(), but raw means are plotted instead of the normed means. For more information, visit: http://www.cookbook-r.com/Graphs/Plotting_means_and_error_bars_(ggplot2).

Value

A scatter plot of class ggplot.

References

Morey, R. D. (2008). Confidence intervals from normalized data: A correction to Cousineau (2005). Tutorials in Quantitative Methods for Psychology, 4(2), 61-64. doi:10.20982/tqmp.04.2.p061

Examples

data <- mtcars
names(data)[6:3] <- paste0("T", 1:4, "_var")
plot_means_over_time(
  data = data,
  response = names(data)[6:3],
  group = "cyl",
  groups.order = "decreasing"
)

# Add significance stars/bars
plot_means_over_time(
  data = data,
  response = names(data)[6:3],
  group = "cyl",
  significance_bars_x = c(3.15, 4.15),
  significance_stars = c("*", "***"),
  significance_stars_x = c(3.25, 4.5),
  significance_stars_y = list(
    c("4", "8", time = 3),
    c("4", "8", time = 4)
  )
)
# significance_stars_y: List with structure: list(c("group1", "group2", time))

Visually check outliers (dot plot)

Description

Easily and visually check outliers through a dot plot with accompanying reference lines at +/- 3 MAD or SD. When providing a group, data are group-mean centered and standardized (based on MAD or SD); if no group is provided, data are simply standardized.

Usage

plot_outliers(
  data,
  group = NULL,
  response,
  method = "mad",
  criteria = 3,
  colours,
  xlabels = NULL,
  ytitle = NULL,
  xtitle = NULL,
  has.ylabels = TRUE,
  has.xlabels = TRUE,
  ymin,
  ymax,
  yby = 1,
  ...
)

Arguments

Value

A dot plot of class ggplot, by group.

See Also

Other functions useful in assumption testing: Tutorial: https://rempsyc.remi-theriault.com/articles/assumptions

Examples

# Make the basic plot
plot_outliers(
  airquality,
  group = "Month",
  response = "Ozone"
)

plot_outliers(
  airquality,
  response = "Ozone",
  method = "sd"
)

Get group means and CIs (rcompanion::groupwiseMean)

Description

Get group means and bootstrapped effect sizes from the rcompanion::groupwiseMean function. The function had to be taken separately from the package as the dependency is failing upon install of the current package.

From the original documentation: "Calculates means and confidence intervals for groups."

From: https://rcompanion.org/handbook/C_03.html

"For routine use, I recommend using bootstrapped confidence intervals, particularly the BCa or percentile methods (but...) by default, the function reports confidence intervals by the traditional method."

Usage

rcompanion_groupwiseMean(
  formula = NULL,
  data = NULL,
  var = NULL,
  group = NULL,
  trim = 0,
  na.rm = FALSE,
  conf = 0.95,
  R = 5000,
  boot = FALSE,
  traditional = TRUE,
  normal = FALSE,
  basic = FALSE,
  percentile = FALSE,
  bca = FALSE,
  digits = 3,
  ...
)

Arguments

Details

The input should include either formula and data; or data, var, and group. (See examples).

     Results for ungrouped (one-sample) data can be obtained by either
     setting the right side of the formula to 1, e.g.  y ~ 1, or by
     setting \code{group=NULL} when using \code{var}.

Value

A data frame of requested statistics by group.

Note

The parsing of the formula is simplistic. The first variable on the left side is used as the measurement variable. The variables on the right side are used for the grouping variables.

     In general, it is advisable to handle \code{NA} values before
     using this function.
     With some options, the function may not handle missing values well,
     or in the manner desired by the user.
     In particular, if \code{bca=TRUE} and there are \code{NA} values,
     the function may fail.

     For a traditional method to calculate confidence intervals
     on trimmed means,
     see Rand Wilcox, Introduction to Robust Estimation and
     Hypothesis Testing.

Author(s)

Salvatore Mangiafico, mangiafico@njaes.rutgers.edu

References

https://rcompanion.org/handbook/C_03.html

Examples

### Example with formula notation
data(mtcars)
rcompanion_groupwiseMean(mpg ~ factor(cyl),
  data         = mtcars,
  traditional  = FALSE,
  percentile   = TRUE
)

# Example with variable notation
data(mtcars)
rcompanion_groupwiseMean(
  data = mtcars,
  var = "mpg",
  group = c("cyl", "am"),
  traditional = FALSE,
  percentile = TRUE
)

Standardize based on the absolute median deviation

Description

Scale and center ("standardize") data based on the median absolute deviation (MAD).

Usage

scale_mad(x)

Arguments

Details

The function subtracts the median to each observation, and then divides the outcome by the MAD. This is analogous to regular standardization which subtracts the mean to each observaion, and then divides the outcome by the standard deviation.

For the easystats equivalent, use: datawizard::standardize(x, robust = TRUE).

Value

A numeric vector of standardized data.

References

Leys, C., Ley, C., Klein, O., Bernard, P., & Licata, L. (2013). Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. Journal of Experimental Social Psychology, 49(4), 764–766. https://doi.org/10.1016/j.jesp.2013.03.013

Examples

scale_mad(mtcars$mpg)

Winsorize based on the absolute median deviation

Description

Winsorize (bring extreme observations to usually +/- 3 standard deviations) data based on median absolute deviations instead of standard deviations.

Usage

winsorize_mad(x, criteria = 3)

Arguments

Details

For the easystats equivalent, use: datawizard::winsorize(x, method = "zscore", threshold = 3, robust = TRUE).

Value

A numeric vector of winsorized data.

References

Leys, C., Ley, C., Klein, O., Bernard, P., & Licata, L. (2013). Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. Journal of Experimental Social Psychology, 49(4), 764–766. https://doi.org/10.1016/j.jesp.2013.03.013

Examples

winsorize_mad(mtcars$qsec, criteria = 2)