
Effect size and test for comparing multiple variables by multiple grouping variables
Source:R/effects.R
effect_counts_items_grouped_items.Rd
Effect size and test for comparing multiple variables by multiple grouping variables
Usage
effect_counts_items_grouped_items(
data,
cols,
cross,
method = "cramer",
labels = TRUE,
clean = TRUE,
...
)
Arguments
- data
A tibble containing item measures and grouping variable.
- cols
Tidyselect item variables (e.g. starts_with...).
- cross
The columns holding groups to compare.
- method
The output metrics: cramer = Cramer's V, f1 = F1-value (only for variable sets with the same labels).
- 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_counts.
Examples
library(volker)
data <- volker::chatgpt
effect_counts(
data,
starts_with("cg_adoption_adv"),
starts_with("use_")
)
#>
#>
#> |Expectations | Usage| Cramer's V| Chi-squared| n| df| p| stars|
#> |:-----------------------------------------------------------|-----------------------:|----------:|-----------:|--:|--:|-----:|-----:|
#> |ChatGPT has clear advantages compared to similar offerings. | in private context| 0.32| 41.27| 99| | 0.001| ***|
#> |ChatGPT has clear advantages compared to similar offerings. | in professional context| 0.24| 23.46| 99| | 0.097| .|
#> |Using ChatGPT brings financial benefits. | in private context| 0.24| 22.48| 99| | 0.128| |
#> |Using ChatGPT brings financial benefits. | in professional context| 0.37| 53.99| 99| | 0.000| ***|
#> |Using ChatGPT is advantageous in many tasks. | in private context| 0.25| 24.15| 99| | 0.083| .|
#> |Using ChatGPT is advantageous in many tasks. | in professional context| 0.30| 34.57| 99| | 0.004| **|
#> |Compared to other systems, using ChatGPT is more fun. | in private context| 0.30| 34.62| 99| | 0.011| *|
#> |Compared to other systems, using ChatGPT is more fun. | in professional context| 0.20| 16.23| 99| | 0.449| |
#>
#> 2 missing case(s) omitted.
#>