Science, confidential

We've talked here before about the crazy things you can find when you read the "Methods" section of a scientific research paper. (Ostensibly, that's the boring part.)

If you want a quick laugh this morning — or if you want to get a peek at how the sausages are made — check out the Twitter hashtag #overlyhonestmethods, where scientists are talking about the backstory behind seemingly dry statements like "A population of male rats was chosen for this study".


  1. My favorite statement in a paper was “The only variable that correlated with mortality was age.”

    I wish I’d had that framed.

  2. Wow, this is dangerous – I expect most, if not all of these lines to turn up in speeches by ignorant politicians within months, if not days.

  3. As someone who isn’t in the science field, these are still absolutely hilarious and mostly accessible. Humor transcends educational boundaries.

  4. Oh, I thought they were going to be serious methodological issues and I was getting all angry at the bastardization of science.  Or hashtagangry as the kids say these days. In fact, they were just people venting and are quite funny.

  5. Interestingly many tweets are directly about reproducibility of results, meaning if the ‘next’ researcher is able to come to the same conclusions when using your data. Ideally all journals would publish replication data, but that’s often not the case. We have to trust scientists that the results are ‘true’ – but we can’t check it.

    I did a little analysis of the tweets mentioned above. Since the hash tag #overlyhonestmethods went viral, I have been checking tweets for reproducibility issues. Many tweets are in some way connected: every time a researcher admits being ‘creative’ about getting significant results or slightly ‘polishing’ tables and figures, it will be harder to check or replicate their results. Some tweets, however, directly refer reproducible results.

    “you can’t reproduce my results because you don’t know where the samples come from”

    “our published code might or might not reproduce our results”

    “I can’t reproduce my own results”

    I included all tweets that came up when searching for reproduc* among tweets between Day 1 (January 7, first tweet by dr_leigh) to January 9, 5.17pm (the method and the exact tweets are on my blog on political science replication )

Comments are closed.