Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links

Categories

Supported by
Bestel Een wereld vol denkers door Sebastiaan Mathôt via bol.com

Bayesian Log-Linear Regression

msprmspr
in JASP & BayesFactor

Hello,

I am looking into using Bayesian Log Linear Regression and I have 2 quick questions:

1) When I run this on my data, the BF10 = 1. I did come across a thread about this (https://github.com/jasp-stats/jasp-desktop/issues/1348) but this is still occurring even when I have it set to display the 2 best models. Is this suppose to be the case?

2) Is there any way to do post-hocs on this? I have a 2(group)x4(condition) design and I would like to see where the group differences are. Would it be possible to look at 2x2 contingency tables as post-hocs?

Thank you for your help!

Comments

  • EJEJ

    Hi mspr,

    1. The BF10 =1 because the "H0" in the comparison is the model on the first row -- in other words, the model is compared against itself.
    2. We have some post-hoc correction for pairwise comparisons in the ANOVA. It is based on an adjustment of the prior model odds. Something similar could be worked out here, I guess. So yes, you can look at 2x2 tables post-hoc, but you would need to correct by adjusting the prior model odds. This can be based on subjective considerations (how likely was this particular effect a priori?) or on a consideration of how many post-hoc tests you have available. But this has not (yet) been implemented in JASP.
      Cheers,
      E.J.
  • msprmspr

    Hi EJ,

    Thank you for your help!

Sign In or Register to comment.
Sebastiaan Mathôt » v2.0.18 » Copyright 2011 — 2019 » Vanilla flavored » License