1
$\begingroup$

Based on the plots attached. Do i have any issues with the homoscedasticity assumption? It looks like the dots on the scatterplot are spread but there is some sort of downward trend. Is this causing any problems? If yes, how do i solve this in SPSS?

enter image description here

enter image description here

$\endgroup$
1
$\begingroup$

Homogeneity of variance (an easier thing to spell for the thing you're looking for) is best examined by looking at the residuals of your model. Your second plot shows the residuals, but I think you have a much larger problem. Your residuals appear in lines, indicating you are modelling some sort of tiered response (likely the results of a survey).

The assumptions of linear regression, including homogeneity of variance, are most certainly violated in this case and you should consider using a different model which more closely resembles how the data were generated.

$\endgroup$
1
  • 1
    $\begingroup$ You have just 5 distinct observed values of the response.Is a linear regression suitable for your data? But there isn't a trend. That is just an illusion arising from each distinct observed value lying on a line residual = observed $-$ predicted. $\endgroup$ – Nick Cox Nov 20 '20 at 0:32

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.