I took "attitude towards mental illness" as a dependent variable and "gender" as the independent variable to find out the gender differences in attitude towards mental illness. The data was not following the normal distribution so I did the non-parametric test to find the difference. I found the result, I interpreted and communicated the paper to the journal. Now, the reviewer of the paper has asked to control the confounders like other socio-demographic variables which might influence the gender attitude like income, education and occupation etc. So, I request the experts to tell me how to control the confounders in SPSS. In case if you ask me to run ANCOVA and Multinomial Logistic Regression so I did that too. But since, I am rookie in statistics so unable to interpret the results. Please help me!
I'd recommend ordinal logistic regression. The reason for it is not because the data aren't normally distributed, but because they are ordinal data.
For guidance on estimating an ordinal logistic regression model via SPSS, see Module 5 here:
- https://www.restore.ac.uk/srme/www/fac/soc/wie/research-new/srme/modules/index.html
@Bruce Weaver will it be helpful in controlling confounding variables?
I am a bit confused, how did we come to the conclusion that the DV is ordinal in the first place?
Bruce Weaver did you mean example/module 4 (binary logistic regression) or 5 (ordinal logistic regression) in your link?
Hello Rainer Duesing. Ronán Michael Conroy understood the DV to be ordinal, and Pathak Abhijit did not contradict him. I then pointed out that Module 5 on the following page showed how to estimate an ordinal logistic regression model using SPSS:
- https://www.restore.ac.uk/srme/www/fac/soc/wie/research-new/srme/modules/index.html
Pathak Abhijit asked how it addressed the issue of controlling for other variables, and I pointed to Example 4 (in Module 5). ;-)
Dear Bruce Weaver, sorry my mistake, I mixed it up, everything fine what you said.
Regarding the ordinal scale: yes he did not contradict, but Ronan sounded so sure about it and I was not, so I thought I missed something obvious here. Let's see what Pathak Abhijit will say.
Sir, due to some chores I didn't go through it yet. I am quite happy like researchers of big stature are giving so warm response. Give me a day or two I will go through it and will give my response
@Bruce Weaver, sir, example 4 went above my head, unable to comprehend. @Rainer Duesing Ronan sir understood my question well, my DV is categorical.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude