Participants are randomly assigned to two groups- experimental (internet based) and control (physical visit). BMI Measurements will be conducted at 3, 6, and 12th month. In the end, I want to assess if the treatment is effective in changing the BMI category of patient (from obese to normal; overweight to normal).
I have looked into several statistical tests and I'm still a bit confused. I know you have to test first for a normal distribution and then proceed with the appropriate test. I wish to narrow down my choices for this design.
Check https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6913727/
Also check https://www.ahajournals.org/doi/full/10.1161/CIRCULATIONAHA.111.075424
Hi,
If you have 2 groups, and 3 longitudinal points (and enough participants), you can check the normality of your data using Shapiro-Wilk test, and then perform a 2 way repeated measure ANOVA. you will have the factor of time, the factor of group, and the factor of interaction (time x group).
Thanks for contacting us. It seems to me that your research is of repeated measures, taking into account that you measure the same indicator at three different times (t (initial) = 0, t = 1, t = 2 and t = 3 (end), for the same participants , right? Where you should interact with other variables of interest to your research. In these aspects, you can use a robust method, such as GLM Repeated Measures (GEE or GMM), depending on how your data is organized. You can consult this source: (https://www.ibm.com/support/knowledgecenter/SSLVMB_24.0.0/spss/advanced/idh_glmr.html).