I am analyzing a dataset where I did not participate in the survey design and I have a problem with the variables. We have a survey that collects demographic data (categorical) and a score from a scale (continuous), and separately, time to complete an online module (continuous).
If I had designed the survey, I would have implemented a way to create paired data to easily run paired t-tests and simple regressions, but the datasets are independent (I don't know whose times are on the module variable). Is there a regression model that can be used to compare the scale value to the time value, both continuous but not paired nor equal in size?
To run a regression, you need to be able to create a correlation between the independent and dependent variables. In your case, it sounds like there is no way to link the data on the dependent variable to the other variables, so it is not possible to correlate that variable with the rest of the data, which makes it impossible to do any form of regression.
The only hope would be if there is some kind of matching ID variable in both data sets.
Thank you,David L Morgan David L Morgan, that was my initial understanding.
Kareem Omran
do you think the OLS model is a workaround, since I definitely don't have a way to match the variables, as David said?- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression