What is it about?
Analyzing data from surveys using statistical models is based on assumptions. We show that the analysis suffers substantially when the assumptions are broken. Specifically, the current methods used for analysis of data given on 5-point or 7-point scales (from disagree to agree) are not robust when the underlying assumptions are not met.
Photo by Celpax on Unsplash
Why is it important?
Tens of thousands of studies are each year published based on crude psychological data measures. Analysis of this kind of data must be preceded by careful investigation of the assumptions that underlie the statistical procedures. We demonstrate that the procedures are not working when the assumptions are violated.
Read the Original
This page is a summary of: The sensitivity of structural equation modeling with ordinal data to underlying non-normality and observed distributional forms., Psychological Methods, April 2021, American Psychological Association (APA), DOI: 10.1037/met0000385.
You can read the full text:
The following have contributed to this page