Howdy, Stranger!

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

Supported by

Unequal sample size

edited May 2016 in JASP & BayesFactor

Hi,

Does Bayesian statistics in Jasp accounts for unequal sample sizes when considering the strength of evidence?
If so, how? If not, what is than the best way to deal with unequal sample sizes?

I have used JASP to compare the following groups and studies on one outcome measure using a group*study design.
Study 1: Ngroup1 = 89, Ngroup2 = 26
Study 2: Ngroup1 = 38, Ngroup2 = 41
Note: Variances were equal in these groups.

Thanks in advance!

Marieke van der Schaaf

Comments

  • EJEJ
    edited 11:44PM

    Hi Marieke,

    Richard may have more to say about this, but the Bayesian test considers the probability of the data under two competing hypotheses about effect size. This automatically takes into account unequal sample sizes. Note that this is also apparent from Rouder's web-app: http://pcl.missouri.edu/bayesfactor

    Cheers,
    E.J.

Sign In or Register to comment.