Skip to content

Social science and the replication crisis (my talk this Thurs 8 Oct)

My talk at the WZB Berlin Social Science Center 3pm (Central European Time):

Social science and the replication crisis

The replication crisis is typically discussed in the context of particular silly claims, or in terms of the sociology of science, or with regard to controversies in statistical practice. Here we discuss the content of unreplicated or otherwise shaky empirical claims in social science, which often seem to be associated with a model in which important attitudes and behavior can be easily manipulated using irrelevant stimuli. This set of theories, if true, would have important implications for politics and society, supporting certain views held on the left, right, and technocratic center of the political spectrum. Conversely, the lack of empirical support for the manipulable-human model has social and political implications which are worth considering: if people are not so easily swayed in this way, this suggests that we should try to more carefully understand their direct motivations.


  1. Psyoskeptic says:

    I’d love to be able to get slides or transcript or recording or…. something of this.

    • Mike Liveright says:

      I’d like to add a few other POSSIBLE options that might be implemented to encurrage the talk sponsor.

      1) A single day option. There are many of these talks that would be intersting to people around the world who would like the see the talk with a day, but not at the original time. I dont know if ZOOM, or whoever is in control of the real time talk, might have or implement a new option to record the talk and then retain it for a day or two. Then use delayed viewers could see it with the same URL.

      2) Minimal fee option. If a minimual fee, a few dollars were charged, through an online cridit card or paypall system, for delayed viewing for a while, it might give everyone a chance to support the delayed playback.

      In summary, I think that it would be real nice if many more lectures would be assumed to be available later.

  2. John Williams says:

    ” if people are not so easily swayed in this way, this suggests that we should try to more carefully understand their direct motivations.” Trump’s approval rating has been remarkably steady at ~42% through impeachment, Woodward’s book and recordings, Covid mismanagement, etc., etc., so his supports’ motivation might be a good place to start.

    • confused says:

      I don’t find this particularly surprising, because we are in a highly polarized era of US politics, plus the “echo chamber” effect of modern media/social media.

      To someone who’s already committed, the impeachment is just (expected) Democratic opposition. And lots of people apparently believe COVID is insignificant (I’m already hearing “Trump was only sick for a couple days despite being 74 and obese, clearly there’s nothing to it”); these probably wouldn’t agree Trump has mismanaged it .

      Also, I wonder about the way people are interpreting the poll question. If people are reading it as “would you prefer Trump to Biden” people who don’t actually *like* Trump might still say yes. There are generally a lot of people who will vote for one party regardless of who the candidate is, sometimes for emotional/party identity reasons, sometimes for ‘political calculation’ reasons (eg Supreme Court judges).

      • Andrew says:


        Sure, but until recently, presidential approval varied a lot during the presidential term. See the graphs for Bush 1 and 2, Clinton, Reagan, etc. So, sure, in the modern age of polarization, Obama’s approval was stable for most of his presidency, and Trump’s has been even more stable—but this is a new phenomenon.

        • confused says:

          Oh yes. I was talking specifically about this year’s approval rating being not-terribly-surprising in the context of recent years — not saying it is normal compared to a longer time scale.

          IMO we are likely seeing the beginning of a realignment of the political parties, which is why things are weird right now (or perhaps vice versa: the current political situation is too weird to be stable).

          • Mitchell Raful says:

            Humans aren’t rational. There’s no reason to believe Trump’s numbers should keep tanking regardless of his competence or lack thereof.

            • confused says:

              Well, saying that people “aren’t rational” in this broader sense is a much weaker claim than the sort of “easily manipulated using irrelevant stimuli” claims being critiqued – like being more afraid of hurricanes with male names.

              But what do you mean by “rational” anyway? I don’t think it’s really helpful to use that word when it boils down to “acting according to values which [I believe] are wrong”.

              I am not sure how much of the high polarization is “irrational” in the sense of a purely emotional/unthinking response based only on whether there’s an R or D by the candidate’s name, and how much of it is just a result of cultural/social/etc. divides (in which case it may be “rational” in the sense that voting for that party will nearly always promote the policies one wants to see, even if it’s unwise in terms of larger/longer-term implications for the nation as a whole).

  3. Anonymous says:

    Hi Andrew. Is a video recording of your talk available?

Leave a Reply