For those of you who are willing to spend a few minutes on it: a little experiment, in which you’ll be asked to supply words in some semantic domain. For example: list the first 10 colors that come to your mind (without reflecting on them, mulling them over, trying to be original or clever, etc.), in the order they come to you.
There are actually two experiments. Flip a coin to choose Experiment 1 or Experiment 2 (but not both). The instructions for Experiment 1 are here, for Experiment 2 here. I’ll close the submissions in 4 days, on December 1st.
November 28, 2011 at 1:25 am |
There is no choice of experiments. There is only one experiment.
November 28, 2011 at 6:11 am |
No, there really are two different experiments. But I can’t explain the difference until submissions are over.
I *can* say that the fact that there are people like you who think the two experiments are identical is significant.
November 28, 2011 at 6:47 am |
After looking at it second time, I realized the difference( at least in the way questions have been framed; though not of their deeper significance). I am not aware of the actual purpose of the experiment, still, I think, in case, experimenter wants his subjects to choose just one experiment ( which is the case at present) –the access should be given only to one(any) experiment. When a subject views both of the questions and has excess to both of them, hesh might come up with an answer which is not as natural as when he has access to only one of the questions.
November 28, 2011 at 11:03 am |
I was aware of the pitfalls of this way of collecting responses, but chose to do things as simply as possible. The alternative would have been for me to learn to use questionnaire software that would randomize the choice between the two experiments, so that respondents wouldn’t know about other versions of the task. More work for me, and a bit more work for respondents, who would then be constrained by a template for their responses; as it is, the task is easy for the respondents and takes very little time.
December 1, 2011 at 3:32 pm |
I’m not participating since it’s too late already (and besides I’m not a native speaker of English), but, if my short-term memory isn’t deceiving me, the first phrasing initially suggested organs like heart, stomach, liver, intestines, muscles, and the like, whereas with the second I more readily thought of extremities like nose, arm, leg, belly, ear, finger, etc. But of course the latter is wholly irrelevant since I’d already spoiled the pool with the first experiment. Still, the human part feels more important to me in the second phrasing, despite being between parentheses.
November 28, 2011 at 10:12 am |
Looking at Experiment 2 after sending in my results for Experiment 1… I’d hypothesize that the second (“20 (human) body parts”) implies more that the reply be twenty SPECIFICALLY HUMAN body parts, whereas the first (“20 (human) body parts”) specifies it more as an afterthought, asking for body parts that might be human. I’d be interested in seeing how the results differ.
November 28, 2011 at 10:53 am |
Since people seem bent on looking at both experiments, here’s the difference, which is in the instructions: Experiment 1 says “the first 20 parts of the body (the human body)…” and Experiment 2 says “the first 20 (human) body parts”. You have misremembered the instructions for Experiment 1 in an interesting way, which I’ll comment on in a few days. (Responses are still streaming in, taxing my ability to code them — partly a result of some colleagues’ having forwarded my “experiment” posting on to others.)
November 30, 2011 at 4:27 pm |
[…] (yet) a report on the results of my little experiment, but an exploration of how the human body and its parts are viewed in websites that intend to be […]