I have two groups (10 participants in one and 12 in the other), each one viewed 1 stimulus (web page that present 4 products to choose between). One of the web pages has an orange label under one of the products while the other doesn't (that's the only difference). For each participant I have the number of fixations made on each product and I know which product they chose in the end.
My hypothesis is that the group that has the stimulus with the orange label under one product would more likely choose that product.
Is there a statistical test to compare these groups and see if the orange label actually has influence on the choice?
thanks a lot.
Logistic regression-see the attached. This is a simple two group comparison.
Logistic regression-see the attached. This is a simple two group comparison.
You don't even need to do logistic regression, you can do fishers exact test.
However, with so few people, these tests have very low sensitivity (i.e., they won't be able to detect anything but the most extreme effects of orange highlight). Statistical tests probably are inappropriate here. Stick with descriptive statistics (percentage choosing product in each group, and their ratio, as well as the "impact" (i.e., rate difference times group size)), and caution all of it by saying (1) it's entirely possible that simple sampling variability can explain group difference, and (2) baseline difference and tendencies between participants may also explain group differences. It's almost impossible to infer that the highlighting caused any differences
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics