Coefficient H

Please check McNeish (2018) for details. Here, I am using the NSCH dataset as an example.

#load packages
library(haven)
library(tidyverse)
library(userfriendlyscience)

#import data
data<-read_sav("nsch.sav")

#clear the current graphics frame and get ready for the next plot
plot.new()

data %>% 
  filter(across(c(RECOGBEGIN, CLEAREXP, WRITENAME, RECSHAPES), ~ . < 6)) %>% 
  select(RECOGBEGIN, CLEAREXP, WRITENAME, RECSHAPES) %>% 
  scaleStructure()

## 
## Information about this analysis:
## 
##                  Dataframe: .
##                      Items: all
##               Observations: 11648
##      Positive correlations: 6 out of 6 (100%)
## 
## Estimates assuming interval level:
## 
##              Omega (total): 0.71
##       Omega (hierarchical): 0.72
##    Revelle's omega (total): 0.75
## Greatest Lower Bound (GLB): 0.73
##              Coefficient H: 0.74
##           Cronbach's alpha: 0.7
## Confidence intervals:
##              Omega (total): [0.71, 0.72]
##           Cronbach's alpha: [0.69, 0.71]
## 
## Estimates assuming ordinal level:
## 
##      Ordinal Omega (total): 0.8
##  Ordinal Omega (hierarch.): 0.8
##   Ordinal Cronbach's alpha: 0.8
## Confidence intervals:
##      Ordinal Omega (total): [0.8, 0.81]
##   Ordinal Cronbach's alpha: [0.8, 0.81]
## 
## Note: the normal point estimate and confidence interval for omega are based on the procedure suggested by Dunn, Baguley & Brunsden (2013) using the MBESS function ci.reliability, whereas the psych package point estimate was suggested in Revelle & Zinbarg (2008). See the help ('?scaleStructure') for more information.

References

McNeish, D. (2018). Thanks coefficient alpha, we’ll take it from here. Psychological Methods, 23(3), 412–433. https://doi.org/10.1037/met0000144