|
Post by gian96 on Nov 25, 2020 9:46:14 GMT
Dear Psytoolkit's experts , I'm a student and i'm working on my thesis' project by setting up an experiment with Psytoolkit, which is perfect in this period. However, i have some questions about counterbalancing and merging data files.
The experiment consists of 4 experimental blocks of trials. These blocks are subjected to different manipulations. Since participants could rapidly rely on a learning effect, which could heavily bias the results, I need to present these blocks in all possible combinations of order. I created all possible block orders (24) and I used the function "blockorder" to let the software to randomly select one among them. I wonder, then: how I can be sure that all combinations are presented at least to one participant? Is there any scripting syntax line I could use, besides the "blockorder" function?
When I look to the experiment's data I cannot find within the file the subject's number (that is the random number associated with the subject). How can I get the subject's number into the datafile? Or how can I merge the "s" files with their corresponding "task" files?
I hope that someone could clarify these aspects to me. i'm basically new to Psytoolkit's use and i'm trying to get the hang of it. Thank to you all in advance!
|
|
|
Post by matia on Nov 26, 2020 14:12:04 GMT
Dear gian96, I will try to answer, at least some of your questions. 1. I am certain that there is a function that does not require you to do 24 blocks but to tell Psytoolkit "I would like to counterbalance the experimental blocks" however, I would do it in the same way as you, since I do not know how to do it differently. 2. Intuitively, it seems there should be an option to have at least one participant in each condition, but I would not be surprised if there is not any, because as I recall some other post prof. Stoet said that function random cannot guarantee that each condition will be represented with identical no. of participants in them, which is expected and not too big of a deal. Since you have 24 different blocks, to err on the safe side, I would have at least 5 participants per block (cca 125) to be fairly certain that at least one will be assigned to each condition, however, if I must have at least one in each condition - I would do experiment as many times as possible to capture one participant per condition. 3. You can find instructions you are looking for merging survey and experiment data in the following link: www.psytoolkit.org/lessons/project.html#_step_8_analyze_dataHope (some of) this helps, best, Matia
|
|
|
Post by gian96 on Nov 26, 2020 18:35:51 GMT
Thank you very much for your suggestions!
|
|