Contrast tests of interaction hypothesis

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Contrast tests of interaction hypothesis

UnderOver
I am trying to run a test of the residuals with my planned contrasts the (C + R approach) to see if the model is a good fit, however, I don't see to be able to figure exactly how to do so inside SPSS. Any assistance with this would be appreciated.
Reply | Threaded
Open this post in threaded view
|

Re: Contrast tests of interaction hypothesis

Bruce Weaver
Administrator
I guess this is what you're talking about, is it?

Psychol Methods. 2016 Mar;21(1):112-20. doi: 10.1037/met0000044. Epub 2015 Aug 3.
Residual tests in the analysis of planned contrasts: Problems and solutions.
Richter M.

Abstract

It is current practice that researchers testing specific, theory-driven predictions do not only use a planned contrast to model and test their hypotheses, but also test the residual variance (the C+R approach). This analysis strategy relies on work by Abelson and Prentice (1997), who suggested that the result of a planned contrast needs to be interpreted in light of the variance that is left after the variance explained by the contrast has been subtracted from the variance explained by the factors of the statistical model. Unfortunately, the C + R approach leads to 6 fundamental problems. In particular, the C + R approach (a) relies on the interpretation of a nonsignificant result as evidence for no effect, (b) neglects the impact of sample size, (c) creates problems for a priori power analyses, (d) may lead to significant effects that lack a meaningful interpretation, (e) may give rise to misinterpretations, and (f) is inconsistent with the interpretation of other statistical analyses. Given these flaws, researchers should refrain from testing the residual variance when conducting planned contrasts. Single contrasts, Bayes factors, and likelihood ratios provide reasonable alternatives that are less problematic.

Source:  https://www.ncbi.nlm.nih.gov/pubmed/26237506


UnderOver wrote
I am trying to run a test of the residuals with my planned contrasts the (C + R approach) to see if the model is a good fit, however, I don't see to be able to figure exactly how to do so inside SPSS. Any assistance with this would be appreciated.
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: Contrast tests of interaction hypothesis

UnderOver
I believe we are speaking about the same thing. Flawed or not, it is still being used in my discipline and someone asked me to run it as part of my analysis. I can't, for the life of me, figure out how to plug it into my contrast test in SPSS though.
Reply | Threaded
Open this post in threaded view
|

Re: Contrast tests of interaction hypothesis

Bruce Weaver
Administrator
The Nabble archive currently shows that neither of your posts has been accepted by the mailing list yet.  Have you joined the actual mailing list?  If not, you can do so here:

http://spssx-discussion.1045642.n5.nabble.com/mailing_list/MailingListOptions.jtp?forum=1068821


UnderOver wrote
I believe we are speaking about the same thing. Flawed or not, it is still being used in my discipline and someone asked me to run it as part of my analysis. I can't, for the life of me, figure out how to plug it into my contrast test in SPSS though.
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: Contrast tests of interaction hypothesis

UnderOver
I thought I had, but I ran through the subscribe function again. Hopefully it takes this time around.
Reply | Threaded
Open this post in threaded view
|

Re: Contrast tests of interaction hypothesis

UnderOver
In reply to this post by Bruce Weaver
Another way to phrase what I am looking for is the residual between cells variance and then be able to plug that into an ANOVA with my planned contrast test.