Question about real effort tasks

[Chris @ oTree]

For those of you who run real effort tasks such as counting zeroes or text transcription, what kind of granularity of data do you need?

(a) having the full data of each trial, such as what matrix was shown, what answer the user gave, how long it took to count that matrix, etc.? 

or:

(b) just simply how many trials were solved in the given time?

Thanks,

Chris

[Alejandro Hirmas]

Hi Chris!

In my case I would always prefer (a). You can always aggregate it yourself if needed. In some cases, if there is a strong effect of the tasks themselves, then this can be needed. 

[Chris @ oTree]

Thanks!