I ma trying to run a regression analysis. I have two variables that are related. (eg. DVs: 1) do you exercise? and 2) if you workout, how many hours per week?). I am trying to run a suitable model to analyze this types of regression using single regression analysis. Which analysis should I use? Thanks in advance for your suggestions.
I hope you have other variables. In your set up, question #2 perfectly predicts #1 because they are conditional. If the respondent does not exercise, then workout hours is necessarily zero. You need more information to explain the variation of those who have non-zero hours of working out.
Richard John Smith,
I have others variables to use as independent variables. I am trying to use these two variables as dependent variables against several other independent variables. I was thinking Tobit model but not sure if that is suitable. I understand the relation of two variables (if no, 0 hours), but I am not very sure if I understand what you are trying to say with your last sentence.
I agree with Richard. What you've said about the study so far suggests computing the proportion who exercise (with 95% CI) for question 1, and descriptive statistics (both graphical and numeric) for weekly exercise time (using only those who do exercise). The exact nature of the descriptive stats will depend on what the distribution looks like. HTH.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis