By elobot10 - 4/8/2018
Hello,
Is there any way to (accurately) record the response latency for each question (specifically each slider) answered on a surveypage?
I'm trying to do this for the experiment discussed here and I'm running into the following problem:- If keep the original formatting and use the <block> element to display my surveypages (rather than the <survey> element), the same response latency gets recorded for each slider (I'm assuming it's the latency for the first slider that was clicked during the trial?).
- If I use the <survey> element instead, the latency IS different for each question, but the surveydata file that gets spit out at the end of it all only has a single row of data. Poking around the forum a bit, it seems like this is because I am repeating the same surveypage over and over again.
Is there any way to trick Inquisit into thinking that I'm not running the same page over and over again? Alternatively, is there way that I can it to record multiple response latencies for a single surveypage when using the block element? It seems to be able to do that for responses, so I don't see why it can't do that for latency.
Many thanks!
|
By Dave - 4/9/2018
+xHello, Is there any way to (accurately) record the response latency for each question (specifically each slider) answered on a surveypage? I'm trying to do this for the experiment discussed here and I'm running into the following problem: - If keep the original formatting and use the <block> element to display my surveypages (rather than the <survey> element), the same response latency gets recorded for each slider (I'm assuming it's the latency for the first slider that was clicked during the trial?).
- If I use the <survey> element instead, the latency IS different for each question, but the surveydata file that gets spit out at the end of it all only has a single row of data. Poking around the forum a bit, it seems like this is because I am repeating the same surveypage over and over again.
Is there any way to trick Inquisit into thinking that I'm not running the same page over and over again? Alternatively, is there way that I can it to record multiple response latencies for a single surveypage when using the block element? It seems to be able to do that for responses, so I don't see why it can't do that for latency.
Many thanks!
No, a <surveypage> doesn't capture latencies for each individual question on it. The latency recorded reflects when the surveypage was submitted, i.e. when the "next" button was clicked.
|
By elobot10 - 4/9/2018
Ah, okay.
Is there any way at all to get Inquisit to record response latencies for sliders (or other survey questions)?
|
By Dave - 4/9/2018
+xAh, okay. Is there any way at all to get Inquisit to record response latencies for sliders (or other survey questions)?
No, not really. The only _approximate_ way would be to have only a single question on a given <surveypage> instead of multiple questions. Then, of course, the surveypage's latency most closely reflects the latency of the one question that is on it.
The basic conundrum is this: If you have multiple questions on a page, what exactly is the latency for any given question on it supposed to be? It's perfectly possible to answer questions out of order as well as revise the response to a question you've already answered. I.e. suppose you have to sliders A and B: You move slider B, then you go to slider A, move it, then you go back and move slider B some more, etc. What would be the proper response latency for slider B? Only the 1st time you moved it? Or whatever the time was when you finalized your response, which would include any time you spent on slider A. While this might be a simple example for which some different plausible answers can be given, it quickly becomes more complicated the more questions you have and the more you go back and forth between them.
|
|