Blog Coordinator

X-Phi Grad Programs

« Welcome to the new Experimental Philosophy blog! | Main | More on whether top tier journals in ethics and political philosophy are less likely to publish papers with experimental content »

08/05/2013

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Thomas Nadelhoffer

You should post something in the thread about not looking in the comment thread before taking the survey so that people can post worries or concerns without polluting the data.

For instance, the questions are a bit odd. I, for one, don't attribute any pain to the avatar. So, I answered "not at all" to the first question. But then this makes the second question a bit odd since it starts with a conditional predicated on my having agreed with the first statement--namely, "if so..." So, I thought "not at all" was the most appropriate answer to question two (even though it's not quite right). But then how should I have answered question three: It asks yet another "if so..." conditional, but this time none of the answers make sense for me. Making matters worse, these are forced choice answers.

It would be much better, I think, to present people with direct statements not conditionals and ask them to agree or disagree with these direct statements.

We ran into similar difficulties in developing our new scale for measuring beliefs about free will (which is a story for another day). For now, I just wanted to point out what I take to be a problem with the survey.

An unrelated difficulty is that it appears that people can take it more than once. That's not ideal either. It was nevertheless an interesting study. So, thanks for posting it! Hopefully, he finds some interesting results.

p.s. If Hyo-eun Kim would like to be a contributor to the blog, she should send me an email!

Hyo-eun Kim

Thank you very much for your feedback!! Your comments are tremendously helpful. As for the problem with re-taking surveys, please somebody give me some tips to resolve the issue. There are a few ways of having mTurk and Qualtrics check on completed turkers, but this survey is not for lay people, but only for philosophers. I will appreciate further suggestions from others.

Thomas Nadelhoffer

Hyo-eun Kim,

Thanks again for sharing your study. I will send you an invitation to be an official contributor to the blog later on today. For now, I just wanted to tell you that Qualtrics allows you to set it up so that a person cannot take the survey twice from the same url. First, open your survey. Then hit the Survey Options tab. Then, under Survey Protection, there is a box labeled "Prevent Ballot Stuffing." You need to check this box. There is a caveat, however. If you're running a study on campus in a lab and you have multiple Ps taking the study at the same computer, you have to make sure that the Ballot Box Stuffing prevention is turned off--otherwise, after the first P completes the survey, each additional P will just get forwarded to the end! I hope that helps. Good luck with the study! Please post something about the results once you're done collecting data.

Thomas Nadelhoffer

So, I just took another look at the survey to see whether I could make some more concrete suggestions and I noticed you already tried to address some of the issues I mentioned earlier. As things stand, this is how you have worded the first three statements/items:


* Do you agree that your avatar would feel pain if he was stabbed in the stomach by the opponent character in the game?

* How unpleasant do you feel about your avatar's being stabbed by the opponent?

* When you say that your avatar feels pain, how similar is the feeling of pain to that of a normal person?

Instead of asking questions, I would have people agree or disagree with explicit statements along the following lines:

* When my avatar in the game was stabbed, the avatar experienced pain.

* When my avatar in the game was stabbed, I experienced pain.

* Avatars experience pain in the same way that normal humans experience pain.

* Avatars experience pain, in some sense, but not in the same way that normal humans experience pain.

If you don't like using this kinds of statements, you could also use Qualtrics Survey Flow to set your study up so that people are directed to new questions depending on their previous answer.

For instance, first you would give them an item like:

* Do you agree that your avatar would feel pain if he was stabbed in the stomach by the opponent character in the game?

Definitely not Definitely yes

The survey would than branch differently for someone who answered No than someone who answered Yes. This would also eliminate the oddness of presenting Ps with conditions even though they may have already denied the antecedent condition in prior questions.

Well, that's it for now. Let us know how it turns out!

Hyo-eun Kim

Thank you very much again for your kind suggestion and encouragement!

John Yates

An interesting idea for a survey. Mind reading and precognition are said by some people to be variants of the same phenomena, and the recent comments on the work of Bem give some idea of the possibilities of considerable philosophical complexity. Personally I doubt if the work of Heisenberg and other later workers on quantum theory will turn out to be relevant for various reasons, but it is at well to be as open minded as possible.

The comments to this entry are closed.