The Great Pharyngulation Experiment

The Urban Dictionary defines the verb “Pharyngulate” as

To crash an online poll by publishing its link so that large numbers of blog readers will make their opinions known there, often shifting the results dramatically.

The term comes from the name of P. Z. Myers’ popular blog Pharyngula.

As someone interested both professionally and personally in the relationship between religion and rationality in general, and the state of popular opinion on matters related to subjects such as religion and science education in particular, I try to keep informed about efforts to survey public opinion and gather data on these questions. And so the possibility that online polls may be thoroughly meaningless is worth being aware of.

A Matrix Squiddie in PZ's Honor

Pharyngulation is itself a puzzling phenomenon. On the one hand, it seems on the surface like it might significantly skew data gathered using internet surveys. On the other hand, it would seems that in theory it shouldn’t be possible to pharyngulate an online poll unless either the view of the pharyngulators also reflects a majority viewpoint, or there simply aren’t as many interested readers of a web site as there are readers of Pharyngula, or the vast majority of readers of Pharyngula have enough time on their hands to spend large amounts of time answering a poll over and over again (in those cases when that is even possible).

So is Pharyngulation real or a myth? How does it impact attempts to gather data about public opinion? It strikes me that this blog may be ideally poised to try to answer that question, given the range of readers who represent a wide spectrum and variety of atheist, Christian, and other viewpoints.

And so I thought I would try to do an experiment – to my knowledge, the first of its kind – to gauge the reality and effects of pharyngulation. And so I am setting up a “poll,” which follows below. Rather than use an embedded poll, since most will not work on non self-hosted WordPress blogs, and limit the maximum number of responses, not to mention that sometimes the data gets corrupted, I am trying a simpler variation. I am posting two comments, “Yes” and “No,” and am asking you to vote by leaving a REPLY comment on one or the other. Since Disqus counts replies to comments and lets you hide them, this will give an easy tally of the number of comments, and thus the number of votes on each.

What I need you to do is to vote, and more importantly, spread the word. If you have a blog, a Facebook page, a Twitter account, or any other way of getting a message out, please spread the word about this poll. Feel free to encourage your readers to vote a certain way. I’m hopeful that maybe we’ll even have the honor of P. Z. Myers telling his many readers to participate. What I’m curious to find out is whether a concerted effort by many bloggers will result in a balanced outcome, or whether the experiment will in fact confirm the reality – and the power – of pharyngulation.

Enough introduction. Let’s get started.

Is pharyngulation a real phenomenon that significantly affects poll outcomes?

Please vote by leaving a reply comment on either my “YES” or “NO” comment below. Thanks!

Stay in touch! Like Exploring Our Matrix on Facebook:

Footnoting Blogs in Academic Publications
The Gist of Jesus
Do You Want To Know The Truth?
How the Internet Works
  • James F. McGrath


    • Hsagisman


    • Tom Verenna


    • Chris Weimer

      Yes. The poll shows over 90% not believing in god.

    • Darren Pardee


    • Gary

      Yes. Obviously. People interested in a particular blog site represent a narrow slice of population, those interested in that blog (duh). Anything that opens up the response to a larger number of people, with a more diverse background, will represent an influence on the results. And actually be a more representative result with a wider base of responders. Even if there was no bias, a larger number of responders should reduce the calculated error of the results. This assumes that “Pharyngulate” is a random effect, not just directed toward one side or the other. Since both sides of a question have an opportunity to affect the results, I assume the result is more representative of a general population result.

    • Jessica Harmon


    • Lauren Blackwelder

      Yes… and apparently people don’t know how to follow instructions. lol. :-)

    • Just Sayin’

      Yes, a thousand times yes!

    • Erlend


    • Peter Kirk

      Yes. Ph. is effective when a poll is publicised on a blog frequented by people committed to one side of a question. For example, if a poll “is there a God?” gets mentioned on a popular atheist blog, or Christian one.

    • westley

      I vote yes

    • JL


    • cb

      Yes. (You should follow up with a poll on reading poll directions ;0

    • Diversandsundry

      yes, only i’m not sure it works in comment threads 😉

    • skm9


    • Stringman

      Most certainly.  But that’s not where I found out about this survey two months too late.

  • James F. McGrath


  • Anumma

    “or there simply aren’t as many interested readers of a web site as there are readers of Pharyngula” I think this is usually the case. The horde is…well, a pretty big horde.

  • Diogenes of Sinope

    Yes, of course it is. I’ve participated in pharyngulating polls!

  • David


  • Troy Britain

    Yes, of course it is a real phenomenon, however it only has a narrow effect. That is unless “Pharyngulation” were to become a widespread practice among popular bloggers. If that were to happen it might become statistically more significant.

  • Retired Survey Researcher

    ANY poll with self-selected respondents is unscientific.

  • Retired Survey Researcher

    Whoops, I meant to add: ANY poll with self-selected respondents is unscientific, and therefore the results are unreliable.  Look up “selection bias” for the many problems with this kind of polling.

  • Scott Bailey


  • Gregory Jeffers