Multiple Regression


I am trying to do my analysis for my work but getting very confused.

I am using multiple regression and have my dependent variable, which is fine. However, my two predictor variables are attachment and basic need satisfaction.

I have used the Experiences in Close Relationships questionnaire (ECR) and there are two scores for each participant; anxiety and avoidance. I was told to work out the mean for each, but I have no idea what I need to enter as my predictor variables, do I enter the mean score for anxiety and the mean score for avoidant into the same model? Or do I do one for avoidant and one for anxious?

Also, for basic need satisfaction, I have three traits, autonomy, relatedness and competence and was told to work out the average score for each. Do I enter the average scores for each?

Please help!


Ummm I'm not sure why you would need to work out the average score for your variables for a regression. Perhaps they are telling you to work them out for some descriptives of the data?

Or did you mean that they were asking you to get the mean score for each participant for each construct e.g. mean(basic needs)= (autonomy + relatedness + competence)/3? I think that would be a bad idea because the scales are designed to measure different aspects of the overall construct and it would essentially negate the usefulness of the regression, and the usefulness of the measure by using mean scores (unless that is what the test is designed to do).

Does the measure give you a total score (e.g. a total basic needs score based on the different subscales?). You could put this in if you liked, but I would use the subscales if you have them for more specificity (are the subscales reliable though?).

Either way, putting raw scores in is what would be most common (I have not heard of mean score of a subscale for the sample being put in before). Others comments welcome!