Sunday, November 23, 2008
Sobel's Test, Part 2
I think I understand what the diagram on the website means conceptually, but my question is more related to reading the SPSS printout. On my SPSS output for the third regression of question 2, I have three Betas: the constant, and two Betas for the two independent variables I used in the regression. What I'm not sure of is which of these three Betas Sobel's test is asking for.
Subscribe to:
Post Comments (Atom)
2 comments:
Well, you know it is not the beta for the constant, because that refers to the value of your intercept (B0).
You must determine which variable is your mediator and which is the predictor. I could be mistaken but you should also have a beta value for your interaction, correct? Maybe not, I haven't done the assignment.
I guess I am having a difficult time understanding; if Sobel's test is only asking for two betas and you have two betas and you know which variable is your mediator, help me understand where the difficulty lies?
This is what happens when I don't send the homework assignment to the TA's. That was my mistake, and I apologize.
If you substitute the variables from your analysis into the diagram from the website, the beta you need will be obvious. I know it may seem silly, but actually DRAW the model with your variables in it.
Now that you've drawn the model, look at the output from your three regressions.
The first regression gives you predictor to outcome, yes? Does that show up as either path A or B in the diagram? (To answer my own question, no, that is the unlabeled path in the diagram.)
The second regression gives you predictor to mediator, yes? Where does that fit in the model? Figure this out, and substitute in those values in the Sobel's test calculator.
The third regression gives you both predictor to outcome and mediator to outcome. Looking at the model, which one of those do you need? Figure out which one of those betas are represented by path B in the model, and substitute in those values in the Sobel's test calculator.
Please do NOT try to do this in your head, shortcutting the step of drawing and labeling the model with your own variables. It may seem like a silly waste of time, but in fact, it should make it VERY clear which values go where.
Post a Comment