Hello researchers,
I have a statistical analysis question. Attached is a scatterplot that looks at the relationship between circularity and perimeter across six different experimental groups. I wanted to know what would be the best way to determine if there is a statistically significant difference between each of these experimental groups. At a glance, it looks like the groups represented by the darker colors are differerent from the lighter ones. Any advice would be greatly appreciated. Thank you!
Depending on your hypotheses, there are possibly 3 approaches you might try: (1) If you’re interested in comparing groups for higher/lower (e.g., black has more pixels & less circularity than white), then you could do ANOVA. (2) If you’re interested in how the relationships between perimeter and circularity may differ for each group, conduct Fisher z-score comparisons on the correlations. (3) Or if you’re interested in how the groups might cluster differently in your XY plane, then maybe do a cluster analysis and then a Chi Square to see if clusters disproportionally have members of some groups. All of these analyses tend to require large N’s, and your scatterplot looks like maybe you have a dozen cases in each group, so I’m a bit concerned you might not have the power for most stats (beyond the ANOVA). Hope this helps some, Kevin! ~ Kevin
You haven't provided enough information to suggest any procedures, other than with 15 pairwise comparisons you should say how important type 1 and type 2 errors are.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis