Home > methodology > Bayesianism — a dangerous scientific cul-de-sac

Bayesianism — a dangerous scientific cul-de-sac

from Lars Syll

The bias toward the superficial and the response to extraneous influences on research are both examples of real harm done in contemporary social science by a roughly Bayesian paradigm of statistical inference as the epitome of empirical argument. For instance the dominant attitude toward the sources of black-white differential in United States unemployment rates (routinely the rates are in a two to one ratio) is “phenomenological.” The employment differences are traced to correlates in education, locale, occupational structure, and family background. The attitude toward further, underlying causes of those correlations is agnostic … Yet on reflection, common sense dictates that racist attitudes and institutional racism must play an important causal role. People do have beliefs that blacks are inferior in intelligence and morality, and they are surely influenced by these beliefs in hiring decisions … Thus, an overemphasis on Bayesian success in statistical inference discourages the elaboration of a type of account of racial disadavantages that almost certainly provides a large part of their explanation.

For all scholars seriously interested in questions on what makes up a good scientific explanation, Richard Miller’s Fact and Method is a must read. His incisive critique of Bayesianism is still unsurpassed.


  1. April 28, 2015 at 8:20 am

    I am not a fan of ‘Bayesianism’ which is simply a rearrangement of a basic probability identity and is now a sort of sect (sort of like rearrnging the golden to get various religious sects) but he has identified the wrong problem. One can easily use ‘Bayesian’ methods to relate attitudes about race to racial income or job disparities. Its just that in conventional practice, economists do what he describes as ‘phenomenoloigical’ while sociologists do the latter (usually with many fewer equations, often just words) , and this convention has worked well for the past 50 or 100 years.

  2. April 28, 2015 at 9:38 am

    I am unable to see what your argument has to do with Bayesian inference. It is a theory, based on the famous formula of Bayes, regarding the modification of prior beliefs in the light of new evidence. Your argumentation is in fact quite Bayesian. You evidently have a strong, and in my view reasonable, prior regarding the role of racism. The evidence that you discuss is insufficient to substantially alter your prior.

  3. Franklin Chiemeka Agukwe
    April 30, 2015 at 11:52 pm

    Certainly Bayesianism has identified wrong problem which makes it of little value to others

  4. Paul Schächterle
    May 1, 2015 at 12:11 pm

    I would say that Bayesian inference is a calculus over a set of beliefs.

    If the belief is wrong, the results of the calculus are irrelevant even if they are derived in a logically correct manner.

    On a more basic level one can also ask if the notion of a calculus over a “probability” that is not frequentist has any merit.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s