|
Post by Florian on Jul 22, 2020 18:01:56 GMT
Hello Gijsbert,
I just noticed that there may be another problem with "var in" in the new PsyToolkit version 3.1.1. An experiment using variables from the survey worked fine in version 3.1.0, but does not seem to work in version 3.1.1. Specifically, I am using the answer to a radio quesion of the survey in the experiment (using var in ...), but the imported variable &screenreader seems to be always 0 in the experiment (it should be either 1 or 2, see the survey code below). I also just re-compiled the survey in version 3.1.0, and it works just fine again. So I'll use the older version for the purpose of that study, but I thought it might be good to fix the issue in the new version. Thanks a lot!
Best, Florian
---survey code---
l: screenreader t: radio q: Are you using a screen reader (e.g., JAWS or VoiceOver)? - yes - no
---experiment code---
var in screenreader ... task soundcheck show text &screenreader 0 -400 255 255 255 ...
|
|
|
Post by PsyToolkit on Jul 23, 2020 6:54:49 GMT
Thanks for letting me know. I will try this out and let you know!
|
|
|
Post by PsyToolkit on Jul 23, 2020 7:29:59 GMT
It works, I have just checked it carefully. I think I know what your problem is. If you run the survey from the start, it works fine.
But if you run the experiment and then go back to the first question and choose a different one, it will not work.
This is because PsyToolkit (by default) will always take the first value people score in a survey.
If you click the option "allow back" it will work.
This is a problem people run into sometimes during testing.
|
|
|
Post by Florian on Jul 23, 2020 8:10:31 GMT
Thanks Gijsbert, I did not go back in the survey though. The question was just clicked once, but the &screenreader variable submitted to the experiment was 0. The strange thing is that the same code works in version 3.1.0. Best, Florian
|
|
|
Post by PsyToolkit on Jul 23, 2020 8:56:55 GMT
Okay, that is strange. I cannot replicate the problem. Maybe best to contact me via psytoolkit@gmx.com if it still does not work. I want to make sure this works.
|
|