WORK SHEET 10

Publish apparently insignificant results

What is a good and what a bad result in a study? Does insignificance truly mean the result is boring and the research question irrelevant? Or could an insignificant result just indicate that the correlation posited between phenomena does not exist? Okay, this is asking rhetorically. Regard all findings – be they significant or insignificant – as a relevant result that deserves to be explained. Scientific evidence results from methodologically correct studies and the interpretation of the findings resulting from them. Scientific relevance does not result from “searching” for pre-fabricated findings. Save yourself time, money and energy with a preregistration of your study.

Duration 120 min



Try submitting your research plan as a pre-study to a journal before collecting data. Here is a list of such Registered Reports.

This is the idea behind it: reviewers will assess whether the proposed methods, the theoretical basis, the statistical meaningfulness etc. are appropriate. The journal then offers an “in principle acceptance” (IPA), which means: “We will publish your paper whatever the results (as long as you stick to the agreed plan)”. Thus you as an author have a guarantee that your results will be published even if they are insignificant, and exaggerated interpretations of findings can be avoided.

Here you can find a checklist for such a registered report.

The alternative is to publish at “Experimental Results”, a journal for research data that presents insignificant or inconclusive findings. Here is an online seminar on “Pre-results Review in Economics: Lessons Learned from Setting Up Registered Reports.” by the Center for Open Science.

Checklist:

  • Watch online seminar (1h)
  • Scroll list of registered studies
  • Print out checklist for registered studies
  • Submit registered study

Congratulations!

Date: September 2020
Questions, comments and notes are welcome at open-science@zbw.eu



to Open Science Magazine