I wanted assistance on interpreting SPSS output that I made for my Group Project regarding Eating Habits. We basically made a survey related to eating habits and wrote down the score of each participant in SPSS. We also compared these scores to another group's survey that was related to stress and their scores. We have to analyze betas and p values, basically multiple regression.

Here's what I have written:

From our multiple regression analysis, we derived a beta of -0.260 for the predictor variable called stress scores. This beta of -0.260 implies that stress scores is negatively associated with eating habits scores, such that higher stress scores predicts lower eating habits scores, when controlling for other predictor variables like age and gender. A more meaningful interpretation would be eating habits scores decreased by 0.260 standard deviations for each 1 standard deviation change in stress scores... In conjunction to beta and b, we observed a p value of 0.277 which determines no statistical significance because its value exceeds the significance threshold of 0.05.

I have attached my SPSS output at the very bottom.

My question was: Can I say that...

The relationship between stress scores and eating habits is not significant when controlling for age or gender. OR... The relationship between stress scores and eating habits scores goes away when age and gender are held constant.

                  Standardized Coefficients Beta             Significance 
Stress Scores     -0.260                                       0.277 
Age                0.055                                       0.811
Gender             0.325                                       0.182

Dependent variable = Eating Habits Scores

What would be a better way to explain the relationship between stress scores and eating habits scores with respect to other predictor variables such as age and gender?


Of the two choices you give in your final paragraph before the numbers, only the first is supported by the numbers you show. To show the second, you would also have to run a regression without the control variables.

  • $\begingroup$ Thank you so much Mr. Flom. Do you know how to run the regression without the control variables in SPSS software? $\endgroup$ – N.Nidhi Mar 10 '18 at 22:29
  • $\begingroup$ Just remove them from the model. $\endgroup$ – Peter Flom Mar 10 '18 at 22:32
  • 1
    $\begingroup$ That makes more sense. Thank you for time and help, I really appreciate it. I was really stressed about these nonsignificant betas. You made my day! $\endgroup$ – N.Nidhi Mar 10 '18 at 22:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.