I am a incredibly confused so am deferring to the wisdom of those more knowledgeable. I have scoured the web for the answer to my question and I really can't find it. For my post-grad coursework, I have a dataset with 1566 respondents on which I am conducting a hierarchical regression analysis to assess the impact of grandparents on adolescent total difficulties. Some of my control and predictor variables are categorical and some are continuous...
Block 1: I am controlling for:
- 'sex' [f/m - dummy variable coded 0/1 - catagorical]
- 'age' [11-16 - dummy variable coded 0/1/2/3/4/5 - catagorical]
Block 2: From the previous theory/literature I am entering:
- socioeconomoic means ['free school meals' - y/n - dummy variable coded 0/1 - catagorical]
- Family Adversity [continuous and scale data]
Block 3: From the previous theory/literature I am entering:
- Distance to grandparent [same town, 10 miles, diff. city, diff. country - dummy variable coded 0/1/2/3 - catagorical]
- Living with grandparent [y/n - dummy variable coded 0/1 - catagorical]
- Closeness to sig. nif. grandparent [continuous and scale data]
- Closest grandparent involvement [continuous and scale data]
As you see, I have dummy coded the categorical predictors, but how do you include and present them in a table of correlations? :'( - I can only find examples of this with all continuous/scale data. The correlation tables don't make much sense when including each dummy variable on its own merit. Would recreating the original predictor variables, removing the missing values, etc. and making them all scale/continuous be acceptable if I then used and present those reconstituted variables in a correlation matrix, prior to using the dummy variables in the regression?
Thank you so so much to anyone who can reply and help me out with this, my head is about to pop!
------------------------------------------- EDIT -------------------------------------------
As a follow-up, I have been reading more these past few hours, and am I correct to assume that I do not need to report correlations, due to one of my variables being assigned 5 dummy variables? If I placed this on a scale rather than 5 dummy variables, would I then be able to assess and present a correlation matrix before my regression table?
Hello Alistair,
I'd just list the dummy variates along with the continuous variates and report the correlations among all. I would neither omit any of the dummy variates nor recombine them into an ersatz scale (if a scale had made sense in the first place, you obviously wouldn't have needed dummy variates!).
Good luck with your reporting!
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression