Hi,
I am very much interested in conducting factor analysis on binary data. In my data, several brands (8) are assessed with 19 different statements (items). Respondents are asked to assess whether or not certain statement characterize a brand.
I have seen that optimal scaling can be applied to non-metric data, however, it does not work with binary data. Can you suggest me any practical statistical method for binary data ?
I would much appreciate your quick response
Best regards,
Davit
Try latent class analysis (LCA) - respondents can be classified into mutually exclusive and exhaustive types, or latent classes, based on their pattern of answers on a set of categorical variables.
Basically there are 4 types of latent variable measurement models depending on whether the observed variables are categorical or continuous and whether the latent variable is categorical or continuous.
1. factor analysis - composed of continuous latent and observed variables
2. latent class analysis - composed of a categorical latent variable and observed variable
3. latent trait analysis
4. latent profile analysis.
Hope this helps.
Dear Carmen,
Thank you for your response
Can I implement LCA in Stata or in SPSS?
Best regards,
Davit
Most people will use MPLUS to implement latent class models. SAS can be used too. Not in SPSS unfortunately.
LCA is new in STATA 15. Check this out:
https://www.stata.com/new-in-stata/latent-class-analysis/
Dear Alejandro,
Thank you for your suggestions
Best regards,
Davit
- Badges
- Science topic