1
$\begingroup$

I have an independent variable hours/day sitting and a dependent variable test scores, I think all of the assumptions of linear regression are violated, as in the pictures that I added

So I did one way anova instead and I grouped the independent variable into two catagories (6-11 hours, and 12-17 hours). And I used Mann Whitney test instead of t test following the chart that I attached from this site: https://www.spss-tutorials.com/spss-independent-samples-t-test/ I added the results of Mann Whitney test I also added the questionnaire that I used for the two dependent variables.

enter image description here

enter image description here

enter image description here

I'm not sure if this is the correct method because many authors say that grouping continuous independent variables is not a good method, which method should I use?

I have 6 more independent variables and I used one-way anova or t test or kruskal wallis test and Mann Whitney test separately according to violation of the assumptions.

Are these methods that I conducted correct? Sample size is 100 Thank you

enter image description here

enter image description here

enter image description here

enter image description here

enter image description here

enter image description here

Improve this question
$\endgroup$
8
  • 2
    $\begingroup$ Can you tell us more about the data? You've recorded how long individuals sit in hours, and they have these test scores. Are the test scores on a 1-5 scale or something? How is that test score measured? $\endgroup$
    – Demetri Pananos
    Commented Mar 22, 2022 at 18:27
  • 1
    $\begingroup$ @DemetriPananos thank you for your response, my research is about carpal tunnel syndrome I want to determine the associated factors with carpal tunnel syndrome in college students, I have collected data from 100 students. Yes the test scores are scaled from 1-5 for each question. I have two dependent variables symptom severity scale and functional status scale. For Symptoms severity scale I took the total score, the same for functional status scale. I added the questionnaire which I used for the two dependent variables symptom severity scale and functional status scale. $\endgroup$
    – Stats34
    Commented Mar 22, 2022 at 18:36
  • 4
    $\begingroup$ ANOVA is linear regression, so cannot be a solution for violating assumptions of linear regression. $\endgroup$
    – Tim
    Commented Mar 22, 2022 at 19:13
  • 2
    $\begingroup$ @Stats34 If your outcome is an ordinal outcome then you need to perform ordinal regression. $\endgroup$
    – Demetri Pananos
    Commented Mar 22, 2022 at 21:33
  • 2
    $\begingroup$ The variable you actually analyzed, though, is not continuous: when you grouped the outcome, you reduced it to (at best) an ordinal variable. $\endgroup$
    – whuber ♦
    Commented Mar 22, 2022 at 22:55

1 Answer 1

Reset to default
3
$\begingroup$

I have 6 more independent variables and I used one-way anova or t test or kruskal wallis test and Mann Whitney test separately according to violation of the assumptions.

Separate regressions or tests with each of the independent variables is not the best way to proceed. If you omit any independent variable that is associated with outcome and is correlated with variables in the model, then the regression coefficients you get risk being biased. With 100 cases you then have more than 14 cases per independent variable, so you should be able to fit them all together in a multiple regression without overfitting your data.

So far, what you show doesn't suggest there is an association between time sitting and your Boston score, although there are problems with each of the tests you show (as noted in comments). That might improve if you take other independent variables into account in a multiple regression. I wouldn't worry much yet about residual plots and so forth for this single-predictor regression; those might also improve when you take all of your predictors into account together.

If your outcome is the sum of 19 items, each with a 1-5 scale, then you should be able to treat that as continuous. If your full multiple regression model still shows problems you could consider ordinal logistic regression, which models the ordering of outcomes without requiring assumptions related to residuals.

Improve this answer
$\endgroup$
1
  • $\begingroup$ thank you very much for taking the time, I will try both methods. Many thanks $\endgroup$
    – Stats34
    Commented Mar 24, 2022 at 20:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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