https://github.com/mayer79/r-confidence-intervals-cramers-v

Function to compute confidence intervals for Cramér's V measure of association
https://github.com/mayer79/r-confidence-intervals-cramers-v

Last synced: about 1 month ago
JSON representation

Function to compute confidence intervals for Cramér's V measure of association

• Host: GitHub
• URL: https://github.com/mayer79/r-confidence-intervals-cramers-v
• Owner: mayer79
• Created: 2017-02-27T13:32:45.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2018-07-13T12:42:40.000Z (over 6 years ago)
• Last Synced: 2024-10-04T12:56:52.549Z (3 months ago)
• Language: R
• Size: 1.95 KB
• Stars: 3
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# R-confidence-intervals-cramers-V

Function to compute confidence intervals for Cramér's V measure of association, see [1] (with correction). Useful to express how **strong** the relation between the two categorical variables truely is with specified (approximate) certainty. A significant result of the chi-squared test of accociation goes along with a positive lower bound of the one-sided confidence interval.

## Example in R

```

library(MBESS)

pw <- iris$Petal.Width > 1

ct <- chisq.test(pw, iris$Species)

confintCramersV(ct, alternative = "greater") # Estimate 0.91, lower c.i. 0.77

confintCramersV(ct, conf.level = 0.9)        # [0.77, 1]

```

[1] Liebetrau, A. M. (1983). Quantitative Applications in the Social Sciences: Measures of association Thousand Oaks, CA: SAGE Publications Ltd