nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

813589_en-US

liliconcom

Matched pair analysis for categorical survey responses before/after a treatment

Hi I have the results of a survey given both Before and After a treatment.

For the continuous outcomes the Specialized Modeling -> Matched Pairs is straightforward to obtain paired t-tests, with the N = # of matched pairs.

But I am struggling with the best way to test the categorical variables, esp. since there are unequal proportion for the Before and After survey responses, ex. 50 Before and 200 After.

I have the data arranged in both Before/after vertically (long) and paired (wide) formats.

Thanks!

Basic Data Analysis and Modeling

Re: Matched pair analysis for categorical survey responses before/after a treatment

@StPhun wrote:
Hi I have the results of a survey given both Before and After a treatment.

For the continuous outcomes the Specialized Modeling -> Matched Pairs is straightforward to obtain paired t-tests, with the N = # of matched pairs.

But I am struggling with the best way to test the categorical variables, esp. since there are unequal proportion for the Before and After survey responses, ex. 50 Before and 200 After.

I have the data arranged in both Before/after vertically (long) and paired (wide) formats.

Thanks!

To analyze categorical variables in your Before and After survey with unequal proportions, you can use McNemar's test for paired data or a Chi-square test for independent proportions if responses are not paired.

Comments

There are no comments yet