what stat test do i do if my independent variable is binary (Y/N, converted to 1/0) and dependent is continuous (score 1-5)? ive tried log and linear regression but neither seem right.
I'm doing it on R Studio.
the log regression was as follows:
library(readr)
CrateAvSRB2col
Hey Ellie Merson , you state in your question to that the your dependent variable is continuous but just to clarify you are implying that a value of 1.2 is possible? If not Chi-Square is what I would use.
If so you are just do linear regression. I would be happy to run your data in SPSS to tell you what I get, if that is helpful to you.
In your case it is out of scope to use regression.
Your situation is the one typical for t-test or (given the rank character of your scale) a Mann- Whitney test checking the statistical significance and the amount of separation (e.g. registered by t-statistic) of the two 0 and 1 groups.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression