The strength of evidence versus the power of belief: Are we all Bayesians?

author: Jessica Utts, University of California, Irvine
published: Aug. 9, 2010,   recorded: July 2010,   views: 3116
Categories

Slides

Related content

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.
Lecture popularity: You need to login to cast your vote.
  Bibliography

Description

Although statisticians have the job of making conclusions based on data, for many questions in science and society prior beliefs are strong and may take precedence over data when people make decisions. For other questions, there are experts who could shed light on the situation that may not be captured with available data. One of the appealing aspects of Bayesian statistics is that the methods allow prior beliefs and expert knowledge to be incorporated into the analysis along with the data. One domain where beliefs are almost sure to have a role is in the evaluation of scientific data for extrasensory perception (ESP). Experiments to test ESP often are binomial, and they have a clear null hypothesis, so they are an excellent way to illustrate hypothesis testing. Incorporating beliefs makes them an excellent example for the use of Bayesian analysis as well. In this paper, data from one type of ESP study are analyzed using both frequentist and Bayesian methods.

See Also:

Download slides icon Download slides: icots2010_utts_awab_01.pdf (1.3 MB)

Download slides icon Download slides: icots2010_utts_awab_01.ppt (1.7 MB)


Help icon Streaming Video Help

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Reviews and comments:

Comment1 Jeff, January 10, 2012 at 2:45 a.m.:

I too have to be very skeptical that there is not a file drawer problem going on here. Jessica's analysis really needs to take this into consideration and demonstrate that it is not the reason for this bias to be taken seriously. It is the obvious answer that I had been thinking about from the very first slide, and was a question brought up by the audience. The fact that it is such an obvious explanation that she is aware of, yet she didn't mention it during the speech is even more troubling.


Comment2 phayes, September 20, 2013 at 3:18 p.m.:

The file drawer effect or (a convergence on) experimental designs with unrecognised flaws or... I think the naïvety of the ESP researchers (and the statisticians who encourage them!) is quite remarkable¹. AFAICT they haven't really even begun to do what would be required in the way of sound experimental methodology for their results to be interpretable as evidence supporting the ESP hypothesis. The irony in Sean Carroll's tongue-in-cheek recommendation for an appropriately tiny parapsychology research budget² is that the research budget needed to take these silly experiments out of the realm of cargo cult science would probably be far larger than it is now.

¹ http://www-biba.inrialpes.fr/Jaynes/c...
² http://blogs.discovermagazine.com/cos...

Write your own review or comment:

make sure you have javascript enabled or clear this field: