Post by Lea on Jul 11, 2020 15:01:22 GMT
Hello all!
I have read every answer to the FAQs and every article about randomization I could find, but still I sadly haven't found what I was looking for. I am not sure if PsyToolkit is suitable for what we are planning to do (I am doing an internship at a German university for my bachelor degree in Psychology) but before giving up I decided to ask you directly.
We are doing a perception experiment and already have recorded 111 people saying 3 different sentences (1 to 3) in 3 different ways (a to c) meaning we have a total of 999 stimuli. However, we want our participants to only rate 333 stimuli because we don't want to have one participant hear the same speaker say the same sentence in more than one way. Meaning if Speaker 1 is selected saying sentence one in way a, we want the stimuli of "speaker one/saying sentence one/way b and c" to be excluded from the following trials for this specific participants as well as "speaker one/saying sentence two and three/in way a". This is supposed to be for every speaker and every sentence, so if speaker one comes up again after randomly selected other speaker/sentence/way combinations, for example with "sentence three/in way b", we need to add "speaker one/sentence three/way c" (since "speaker one/sentence three/way a" is already excluded as per the first selection of speaker one) as well as "speaker one/sentence two/way b" to the excluded stimuli, only leaving "speaker one/sentence two/way c" left to choose from when speaker one is selected again for the last time.
I could find info on all the other randomizations (for example block randomization of gender of the speaker, randomization of the stimuli selection per se and randomly assigning a participant to one of three rater conditions (between factor)) on your site/FAQ but can't seem to find anything about excluding stimuli in a "if X is selected, exclude YZ" or "don't repeat speaker/sentence combination and don't repeat speaker/way combination" kind of way.
I know that our experiment isn't an easy one and that it is even harder to understand from scratch, so I, of course, understand if you do not have the time to answer.
However, I would be really grateful to hear from you!
Best wishes,
Lea
I have read every answer to the FAQs and every article about randomization I could find, but still I sadly haven't found what I was looking for. I am not sure if PsyToolkit is suitable for what we are planning to do (I am doing an internship at a German university for my bachelor degree in Psychology) but before giving up I decided to ask you directly.
We are doing a perception experiment and already have recorded 111 people saying 3 different sentences (1 to 3) in 3 different ways (a to c) meaning we have a total of 999 stimuli. However, we want our participants to only rate 333 stimuli because we don't want to have one participant hear the same speaker say the same sentence in more than one way. Meaning if Speaker 1 is selected saying sentence one in way a, we want the stimuli of "speaker one/saying sentence one/way b and c" to be excluded from the following trials for this specific participants as well as "speaker one/saying sentence two and three/in way a". This is supposed to be for every speaker and every sentence, so if speaker one comes up again after randomly selected other speaker/sentence/way combinations, for example with "sentence three/in way b", we need to add "speaker one/sentence three/way c" (since "speaker one/sentence three/way a" is already excluded as per the first selection of speaker one) as well as "speaker one/sentence two/way b" to the excluded stimuli, only leaving "speaker one/sentence two/way c" left to choose from when speaker one is selected again for the last time.
Per condition (way of saying the sentence), every participant (listener) hears one randomly selected sentence for each speaker. Per speaker, the same sentence must not be presented in more than one condition.
I could find info on all the other randomizations (for example block randomization of gender of the speaker, randomization of the stimuli selection per se and randomly assigning a participant to one of three rater conditions (between factor)) on your site/FAQ but can't seem to find anything about excluding stimuli in a "if X is selected, exclude YZ" or "don't repeat speaker/sentence combination and don't repeat speaker/way combination" kind of way.
I know that our experiment isn't an easy one and that it is even harder to understand from scratch, so I, of course, understand if you do not have the time to answer.
However, I would be really grateful to hear from you!
Best wishes,
Lea
