Why oh why? Chi-square p-value <0.001, but GLMM p-value 0.998!

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Rose_Anne
Posts: 1
Joined: Fri Jul 28, 2017 10:43 am

Why oh why? Chi-square p-value <0.001, but GLMM p-value 0.998!

Postby Rose_Anne » Fri Jul 28, 2017 11:09 am

Hello SPSS experts!

I have one very short statistical question regarding my research on the quality of care for people with mild or severe dementia in three countries, specifically with the descriptives. I do hope you can help me!

Data about patient care were collected from different physicians, so we want to account for this clustering at the level of physicians in our analysis. I checked if there is any difference in the longest place of residence in the last year of life (home, care home, elsewhere) for people with severe dementia between three countries. With the usual Chi-square test, the p-value was significant (<0.001), but when I conducted the Generalized Linear Mixed Model to account for clustering at the level of physicians, the p-value was almost 1 (0.998).

Extra info
Screen Shot 2017-07-28 at 13.08.01.png
1. Sample size : Country 1=312 Country 2=128 Country 3 = 49
2. The differences between the p-values from chi-square and from the GLMM were good in other descriptive variables (age, gender, and cause of death). The same goes with the differences between the p-values from chi-square and GLMM were good in all descriptive variables for people with mild dementia.

What do you think is potential problem in this model? If something is unclear or you would want more info, please let me know! I am more than happy to hear from you! :)

All the best
Rose_Anne
You do not have the required permissions to view the files attached to this post.

Who is online

Users browsing this forum: No registered users and 2 guests

cron