volker (version 3.2.0)

effect_metrics_items_cor_items: Output correlation coefficients for multiple items

Description

The correlation is calculated using stats::cor.test.

Usage

effect_metrics_items_cor_items(
  data,
  cols,
  cross,
  method = "pearson",
  adjust = "fdr",
  labels = TRUE,
  clean = TRUE,
  ...
)

Value

A volker table containing correlations.

If method = "pearson":

• R-squared: Coefficient of determination.

• n: Number of cases the calculation is based on.

• Pearson's r: Correlation coefficient.

• ci low / ci high: Lower and upper bounds of the 95% confidence interval.

• df: Degrees of freedom.

• t: t-statistic.

• p: p-value for the statistical test, indicating whether the correlation differs from zero.

• stars: Significance stars based on the p-value (*, **, ***).

If method = "spearman":

• Spearman's rho is displayed instead of Pearson's r.

• S-statistic is used instead of the t-statistic.

Arguments

data

A tibble containing item measures.

cols

Tidyselect item variables (e.g. starts_with...).

cross

Tidyselect item variables (e.g. starts_with...).

method

The output metrics, pearson = Pearson's R, spearman = Spearman's rho.

adjust

Performing multiple significance tests inflates the alpha error. Thus, p values need to be adjusted according to the number of tests. Set a method supported by stats::p.adjust, e.g. "fdr" (the default) or "bonferroni". Disable adjustment with FALSE.

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

library(volker)
data <- volker::chatgpt

effect_metrics_items_cor_items(
  data,
  starts_with("cg_adoption_adv"),
  starts_with("use"),
  metric = TRUE
)