effect_metrics_one_grouped: Output a regression table with estimates and macro statistics

Description

The regression output comes from stats::lm. T-test is performed using stats::t.test. Normality check is performed using stats::shapiro.test. Equality of variances across groups is assessed using car::leveneTest. Cohen's d is calculated using effectsize::cohens_d.

Usage

effect_metrics_one_grouped(
  data,
  col,
  cross,
  method = "lm",
  labels = TRUE,
  clean = TRUE,
  ...
)

Value

A volker list object containing volker tables with the requested statistics.

Regression table:

estimate: Regression coefficient (unstandardized).
ci low / ci high: lower and upper bound of the 95% confidence interval.
se: Standard error of the estimate.
t: t-statistic.
p: p-value for the statistical test.
stars: Significance stars based on p-value (*, **, ***).

Macro statistics:

Adjusted R-squared: Adjusted coefficient of determination.
F: F-statistic for the overall significance of the model.
df: Degrees of freedom for the model.
residual df: Residual degrees of freedom.
p: p-value for the statistical test.
stars: Significance stars based on p-value (*, **, ***).

If method = t.test:

Shapiro-Wilk test (normality check):

W: W-statistic from the Shapiro-Wilk normality test.
p: p-value for the test.
normality: Interpretation of the Shapiro-Wilk test.

Levene test (equality of variances):

F: F-statistic from the Levene test for equality of variances between groups.
p: p-value for Levene's test.
variances: Interpretation of the Levene test.

Cohen's d (effect size):

d: Standardized mean difference between the two groups.
ci low / ci high: Lower and upper bounds of the 95% confidence interval.

t-test

method: Type of t-test performed (e.g., "Two Sample t-test").
difference: Observed difference between group means.
ci low / ci high: Lower and upper bounds of the 95% confidence interval.
se: Estimated standard error of the difference.
df: Degrees of freedom used in the t-test.
t: t-statistic.
p: p-value for the t-test.
stars: Significance stars based on p-value (*, **, ***).

Arguments

data: A tibble.
col: The column holding metric values.
cross: The column holding groups to compare.
method: A character vector of methods, e.g. c("t.test","lm"). Supported methods are t.test (only valid if the cross column contains two levels) and lm (regression results).
labels: If TRUE (default) extracts labels from the attributes, see codebook.
clean: Prepare data by data_clean.
...: Placeholder to allow calling the method with unused parameters from effect_metrics.

Examples

Run this code

library(volker)
data <- volker::chatgpt

effect_metrics_one_grouped(data, sd_age, sd_gender)

Run the code above in your browser using DataLab