#### lulzyeahokbro

##### New Member
Just 2 quick questions that I really hope someone can help with.

I’m using SPSS to analyze data collected from a survey that was distributed to 110 participants. I don’t need help setting up the data; I just need help knowing which tests are appropriate.

One things I am looking at is whether or not the time they spend at work during the week (options were: 5-10 hours a week, 10-20, 20-30, 30-40, or 40+ so I have numerical values of 1, 2, 3, 4, and 5 to represent each response) has an effect on whether or not they are involved with going to the gym (options were: either they go to the gym or not so I have numerical values of 1 and 2 to represent each response). Which test would I run to see if there’s a significant effect based on the data collected from these two questions?

Second question:

Since I have two groups (the ones who go to the gym and the ones who don’t), how can I compare their levels of satisfaction with regard to their fitness? Their satisfaction responses are either that they agree or disagree with being satisfied with their level of exercise (options are either being satisfied or not being satisfied). Which test would I run to see whether going to the gym impacts their level of satisfaction?
Silly survey I know, but I would really appreciate any responses. Thanks guys. :tup:

#### trinker

##### ggplot2orBust
Caveat: you aren't explicit on what info you've given so my answers may reflect your lack of explicitness...

One things I am looking at is whether or not the time they spend at work during the week (options were: 5-10 hours a week, 10-20, 20-30, 30-40, or 40+ so I have numerical values of 1, 2, 3, 4, and 5 to represent each response) has an effect on whether or not they are involved with going to the gym (options were: either they go to the gym or not so I have numerical values of 1 and 2 to represent each response). Which test would I run to see if there’s a significant effect based on the data collected from these two questions?
You have a binary outcome and categorical predictor. My first approach would likely be logistic regression. You could also do this on the second question as well but use going to the gym as a predictor and satisfaction with the outcome. There are other approaches using chi squared or fishers exact test that may also be a viable solution.

#### trinker

##### ggplot2orBust
PS I would have put this under the statistics forum as it's not really specific to SPSS