Hi Everyone, I am new to STATA. I am currently working on a research project that needs to compare the proportion (Prevalence) of one disease between six age groups, and to the best of my knowledge z test can compare the proportion for one or two groups. Any ideas?
Hi Abdullah.
You can compare proportions by regression analyisis. Specifically binomial regression with log link function.
Use the command:
binreg var_dep i.var_group, rr
You'll see that stata calls the exponentiated coefficent "Risk Ratio". However, this regression model works for proportions (prevalence and cumulative incidence).
Additionally you can add more covariates in order to adjust your estimations. Confounding analysis is a must in observational studies.
respected researcher
1-In such a case, you must follow several steps:
Extract the comparison in the prevalence of the disease in each age group separately.
2-One-way analysis of variance was extracted for each variable to obtain the differences between the age groups.
3-Extraction of two-way analysis of variance to compare the research variables according to the six age groups.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS