Why I'm facing inconsistent Beta values?

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Posts: 1
Joined: Fri Jul 20, 2012 6:37 am

Why I'm facing inconsistent Beta values?

Postby Cyrus » Fri Jul 20, 2012 8:45 am

I did hierarchical regression analysis on my data due to having interaction effects in my research model.
R2 increased from .695 in model1 (main effect only) to .734 in model2 (main &interaction effects)(sig. F change = .000). All the assumptions for the regression analysis have been met. I have two problems with the "coefficients" table:
regression coefficients.jpg
1. As u can see in the table, the insignificant beta value of ZSC in model1 became significant in model2! Is it ok? I'm confused! Which value should i consider to reject/accept the related hypothesis? B of model1 (which rejects the hypothesis) or 2 (which confirms it!!)?

2. Although the Beta value for ZSC_X_CS is significant, its positive sign is against the hypothesis! it's supposed to have a negative sign according to the literature & also logic! How should i treat this hypothesis? Accept? Reject? Partially accept?!!!

You do not have the required permissions to view the files attached to this post.
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: Why I'm facing inconsistent Beta values?

Postby GerineL » Thu Jul 26, 2012 8:45 am

1. It is possible for a relation to seem ns. if you don't take into account an important moderator. Suppose you want to know the relation between breast size and self-confidence. If you run a regression without any moderators it will probably not reach significance. However, if you add gender as a moderator it might: For males there is a negative relation between breast size and self-confidence, for women there is a positive relation between the two. They sort of canceled each other out in the earlier analysis, thus it did not reach significance. Thus, it is no problem that there was no sig. relation in the first analysis.

2. It is hard to explain this without knowing anything about what you tested and how. It happens sometimes that something turns out to be different than you expected, that’s why we do research, to test our hypotheses. It could be the case that there is an important third variable you did not take into account (like with the breast size gender situation). It could be that you did something wrong. It could be all sorts of things. But this is what the test tells you, so it would make no sense to reject your hypothesis based on literature even though empirical evidence points the other way. However, statistics by no means promises to find true answers to anything, so being cautious when interpreting it is always a good idea.

Who is online

Users browsing this forum: No registered users and 1 guest