I have different groups (2 or 3 depending on the analysis) which I suppose is an independent variable. Each group contains between 20 to 100 people. I also have binary repeated measures of a variable (e.g. if smoker or not) before and after, which I guess is the dependent variable. I would think of something as the McNemar test, however, that test does not differentiate between the 2 or 3 groups, only within-subjects.
I could do a chi-square test on the measures before and another chi-square on the measures after and then, compare qualitatively if the frequencies have changed or not, but it does not seem appropriate.
Thank you in advance, any comment would be appreciated.
Hi,
Articles that examines this issue:
Williamson DS, Bangdiwala SI, Marshall SW, Waller AE. Repeated measures analysis of binary outcomes: applications to injury research. Accid Anal Prev. 1996 Sep;28(5):571-9. doi: 10.1016/0001-4575(96)00023-1
Faculty of Health Sciences Models for binary data Analysis of repeated measurements 2015 Julie Lyng Forman & Lene Theil Skovgaard Department of Biostatistics, University of Copenhagan.
Article Dealing With Binary Repeated Measures Data
Go through this article:
Sample size for a two-group comparison of repeated binary measurements using GEE
Sin-Ho Jung 1, Chul W Ahn
Affiliations expand
- PMID: 16118812
- DOI: 10.1002/sim.2136
Watch this video on “How to choose the correct statistical test when we are comparing the groups for statistical difference” https://youtu.be/aUp8sZXDzc8
A video on “Choosing the Correct Statistical Test (For Relationship between variables) with examples” https://youtu.be/1M5DQphChOg might be helpful.
- Badges
- Science topic
- Similar topics
- Correction