Time Duration-Dependent Skipto? (case 3134)

Hi--

I am trying to piggyback a short study onto another. The problem is that we only want people to be able to take the study if there is sufficient time left in the session for participants to take it.

I was thinking that this could be accomplished via a complex skip using responses.xls, if it can record the amount of time that people take to go from the second screen to the last screen. But, I am unsure how to actually do it.

So, ideally, my study would be presented to Ps only if they take < 18 min to complete the initial study. Is there a way to program this rule?

I am trying to get this coded and ready for Monday morning at 8am, as the materials are time-sensitive (we are measuring political attitudes and voting intentions ahead of the Tuesday election). Any immediate advice would be extremely welcome and appreciated.

Thanks much for your help,
-- geoff d. (in the Petty lab @ OSU)

This is obviously too late to help Geoff (sorry!), but if anyone else is curious, here's one way to accomplish this:

This requires creating a complex skip with the responses.xls feature. If you haven't yet, best to first read the "Advanced Features" section of the manual.

The responses.xls (excel spreadsheet) can capture data as the experimental session is running. So in your experiment, turn on the "write reaction time" setting for every single question in your first study. Then, in the "variables" column of the responses.xls, create rows for all the time data. That way, as MediaLab is recording the time data, it will also write it to responses.xls.

Now you can create a calculated variable in responses.xls from the reaction time data. Using Excel's "sum" function, create a variable (let's call it "duration") that sums ALL the reaction time data.

Finally, enter the variable name of the last question of your first study (or perhaps better, a space-filling question that comes after the last question) into the responses.xls. Now you're going to create a conditional skip using excel's "IF" function. It will look something like this: IF(b32>1080000, "debrief", "study2")

Excel understands that as: "If cell b32 (i.e. duration) is more than 1080000ms (i.e. 18 minutes), then the value of this cell will be "debrief". Anything else, it will be "study2". Now when MediaLab completes that question, it will either skip to a question named "debrief" (if more 18 minutes has been recorded in the reaction time data), or skip to a question named "study2".

Now, this won't be precise, because the reaction time sum is going to miss some of the time that elapses during the experiment (e.g., time loading questions). Depending on how long your experiment is, and how much time it takes different components to load, this could add up. You could try running a pilot participant or two and time them yourselves. Compare your time to the duration calculation, and based on the discrepancy, add a constant to the excel formula that calculates the duration.

Little bit of a kludge, but that should get the job done.