My study has 3 treatments. the values measured are only on day 0 and the last day of the experiment.
That depends on what your question is, and how your outcome is measured!
Often in three-group studies there is one group that gets a control treatment, which could be no treatment at all, or a placebo, or 'treatment as usual', meaning the best current treatment. This acts as a baseline for comparisons.
The second issue is how the other groups are made up. It could be that they are completely distinct treatments, but the other common design is that they are two versions of the same treatment (for example, different doses of the same drug).
These factors will determine which groups are compared to which.
Then there is the question of how the outcome is measured. Continuous variable, ordered categories, or a binary variable.
So tell us more about your research question and your methods please.
Assuming that your dependent variable is interval-level, the most common format for analyzing what you describe is a repeated measures ANOVA. If your results are significant, you can use "post-hoc" tests to compare the three categories.
It depends on data type, distribution, and research question.
Assuming you have continuous data and paired measurements (the same subjects measured on day 0 and the last day), a common approach would be:
Paired Sample T-Test: This test compares the means of two related groups to determine if there is a statistically significant difference between these means. In your case, you would compare day 0 and the last day for each treatment group separately.
If you want to compare the changes across all three treatment groups simultaneously, you could use a repeated measures ANOVA.
Martha Nina Mugot Batican's brief description of the study makes me wonder if the 3 treatments are 3 independent groups, with the day 0 measurement being a baseline measure taken before any treatments have been administered. If so, and especially if there has been random allocation to groups, and if means and SDs are sensible for describing the DV, I would be considering ANCOVA with the Day 0 value as the covariate and the last day value as the DV. But as others have said, more info is needed!
CHOOSING THE CORRECT STATISTICAL TEST IN SAS, STATA, SPSS AND R
https://stats.oarc.ucla.edu/other/mult-pkg/whatstat/
we are not an oracle... we do not look for the answers by looking into a crystal ball,
Please always add as much background information as possible to guide the answer.
- Badges
- Science topic
- Similar topics
- Psychology