Interpreting Dunn-Bonferroni test: All three differ significantly?

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

Posts: 1
Joined: Sat Aug 12, 2017 3:28 pm

Interpreting Dunn-Bonferroni test: All three differ significantly?

Postby StefanieK » Sat Aug 12, 2017 3:29 pm

I have this huge knot in my head while writing my thesis atm. I am totally untalented when it comes to statistics, so it could be that my question is easy to answer for some of you.

I did a Kruskal-Wallis test to test the influence of three different linking styles of a Facebook fan page on the engagement rate. I also did a Dunn-Bonferroni post hoc test and the result was that all three groups differed significantly from each other. The mean ranks are 120.85, 168.17 and 213.11 and p being 0,000 twice and 0,013 once.

Do I now just report that they all differ significantly from each other? And what would that mean? That incorporating a link (mean rank=120.85) is less effective when it comes to a high engagement rate than not using any link (mean rank=168.17) or sharing a post (mean rank=213.11)?

I hope somebody can clarify this for me.

Thanks, Steffi

Who is online

Users browsing this forum: No registered users and 1 guest