Dear all,
I want to compare the percentage of "proportion of time in a specific arrhythmia" between two independent groups of different sizes (group A - 16 patients and group B - 61 patients). Can this be done with a t-test?
Regards
Chi-square won't work because the data aren't a table of counts. That is, it's not like you can count the number of seconds of arhythmia and the number of seconds of not-arhythmia.
Something like a t-test may work, or nonparametric tests like Wilcoxon-Mann-Whitney or a median test.
Because the dependent variable is percent/proportion, I would be tempted to use beta regression, with a simple dummy variable as the independent variable to represent the groups.
You may use chi square. You may refer the following link of research gate
https://www.researchgate.net/post/comparing_frequency_counts_between_two_groups_of_different_sample_size_Chi-square
Chi-square won't work because the data aren't a table of counts. That is, it's not like you can count the number of seconds of arhythmia and the number of seconds of not-arhythmia.
Something like a t-test may work, or nonparametric tests like Wilcoxon-Mann-Whitney or a median test.
Because the dependent variable is percent/proportion, I would be tempted to use beta regression, with a simple dummy variable as the independent variable to represent the groups.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS