Interpretation of oppositely directed effects
Hi :)
I'm currently struggling with the following situation: Im conducting two Baysian repeated measures t-Tests. The two conditions I want to compare are the same in both analyses, but for two different DV. In both cases I'm trying to quantify the evidence for the hypothesis that Measure 1> Measure 2.
Test 1 shows evidence for the hypothesis with BF+0=10
Test 2 shows very strong evidence against it with BF+0=0.0001
Can these results be combined to conclude that there is overall evidence against the hypothesis?
Can both BF be combined to quantify the total evidence, i.e. in this case as something like .001?
Do I have to perform a combined test that features both DV as a compound variable, or is none of this valid and both results have to be regarded as strictly separate?
Thank's a lot!
Comments
Hi Chribo,
These tests are based on different data (DVs) so they can't easily be combined, afaik. I'd make absolutely sure that the direction of the effect has been coded correctly -- to get BF+0 = 0.0001 the effect must go clearly in the direction opposite that the one predicted. When this happens, I think it should be mentioned (you could show the posterior under H1 using a two-sided analysis); of course you if the choice is between H0 and H+, you have what you have, but the data will speak against H0 as well, so in an exploratory sense it may be good to remark that the data actually go in the opposite direction.
Cheers,
E.J.