linear regression with plot that shows linearity is not adhered to

Hello. I have the following research question "does quality predict satisfaction"? Quality has 3 subdimensions : perivallon (environment), allhlepidrash(communication) and ofelos (benefit).
I figured I should do many linear regressions to check it out. E.g. compare ofelos with satisfaction (ikanopoihsh), allhlepidrash with satisfaction and perivallon with satisfaction. But when i plot the data I see the graph I' ll try to attach... So is it correct to do linear regressions or is something else needed?

Do you only have 10 data points or is there over plotting of points? Are the data ratings so that the distance between "numbers" is not necessarily uniform?

Maybe you have to deal with categorical variables in which case you might find this discussion useful:

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.