search

coschroeder / mental_health_analysis

GNU General Public License v3.0
3 stars 0 forks source link

Ordinal Logistic Regression? #1

Open phtaheri opened 3 months ago

phtaheri commented 3 months ago

To authors of this publication,

For a final group project in my statistics course, we chose to replicate your regression analyses. I am curious as to why you have an ordered logistic regression included in the data analysis script (at the very bottom), but do not mention its inclusion in the paper or supplements. I am confused as to how to interpret your output. Can you please provide some clarification on its inclusion, and results, including the motives behind including it in the script?

Thanks, Parsa

coschroeder commented 2 months ago

Hi Parsa,

thanks for your interest in our study and reproducing the analysis. We would be very interested in your experience with the data and if you could reproduce the results.

Regarding the ordinal logistic regression:

In the first part of the linear regression we investigate GAD7, which is the mean over 7 items. Therefore we decided to look at it as a "continuous" outcome variable, as it can take a lot of different values. To be comparable with this analysis, we also looked at PHQ2 as a "continuous" variable, although it can only take seven values. But we performed the ordinal logistic regression as a control experiment, and the results were qualitatively the same.

Please let me know if you have further questions.

Best, Cornelius