Regression analysis in ordinal data?

[kassiopi]

In my thesis I will study the relationship between two variables as well as the role of a moderator variable. All of my variables have been measured via Likert scale questionnaires and thus they are ordinal data. I was thinking of doing multiple regression analysis but I am not sure. What kind of statistical test should I use?

Thank you in advance!

 

[Karabiner]

You could have a look at ordinal logistic regression.

If you have multi-item Likert scales (not just single Likert items),
then maybe you can treat this as interval scaled.

With kind regards

Karabiner

 

[kassiopi]

Karabiner thank you a lot for your answer. Every scale consists of multiple likert-type questions. Additionally, every participant who completes the questionnaire achieves a specific score. So, could I treat the data as interval and use parametric moderator analysis?

 

[Karabiner]

I have often seen that the sum score of a multiple Likrt-type item scale
was treated as if it was interval scaled. Therefore I assume that a linear
regression approach can be justified.

With kind regards

Karabiner