[solved] Logger/Variable relationship

Comments

• lvanderlinden

  September 2012 edited 10:25PM

  ITA

  Hi Kyle,

  I'll get back to you soon!

  Regard

  Did you like my answer? Feel free to
• lvanderlinden

  September 2012 edited September 2012

  Hi Kyle,

  With the second version of your script you're almost there! The only thing is that you should place a logger item at the end of every sequence instead of only at the very end of your experiment. By doing this, every trial/question looks as follows:

  • question + response
  • logger

  The problem with your structure is that the variable "response" is overwritten by the response on every next question. But if you restructure your second version slightly by appending a logger item to every sequence (demo_sequence, question_sequence, and map_sequence), the general variable "response" will always contain the appropriate response. And, in case of your rating questions, this will indeed be a number between 1 and 7.

  Note that you need to append the same logger item to all three sequences. So, for example, use

  • append new item → logger in the demo_sequence, and then
  • append existing item → logger (or however you renamed it) in the other two sequences

  I hope this helps!

  Good luck and feel free to post any further questions!

  Lotje

  Did you like my answer? Feel free to
• Kyle

  September 2012 edited 10:25PM

  Lotje,

  That is just what was needed. Still learning the curves of the program, but it's coming together. Many thanks to you!

  Kyle
• plastic_student

  September 2012 edited 10:25PM

  Hello,

  I am struggling with the same problem here. The thing is, that I do not only need the responses logged, but also avialable for later conditions on showing items.

  I tried to save the responses after each questionaire-item into another global variable via inline-script, but from the script item I do not seem to get access to the response variable.

  Does anyone have an idea?

  Thanks.
• Edwin

  September 2012 edited September 2012

  If I understand correctly, you basically want to log the responses for every item and access them later on? I'm going to assume you're going to present your questionnaire first and present items later on (based on a participant's responses on the questionnaire).

  You could use an inline script to log the responses into a global variable (a dictionary) and another inline script to access this global variable and set an entry to a variable used in the current sequence (if that sentence makes no sense, don't worry, just look at the example bellow).

  Basically, what were going to do is make a library of a participant's responses to particular questions. For this to work in random order of presentation too (which is the most flexible), the easiest way to do so is probably to add a new variable to your loop-item to hold question numbers in. In this example, we'll call it 'qnr' and we'll set to '1' in the first cycle, to '2' in the second, '3' in the third etc (N.B.: do not include the apostrophes)

  Your experimental sequence could look like this:

  main sequence
  - inline_script_0
  - questionnaire loop (with [qnr]!)
  -- questionnaire sequence
  --- questionnaire item
  --- inline_script_1 (see below)
  - item presentation loop
  -- second sequence
  --- inline_script_2
  --- item for answer dependent showing
  

  inline_script_0

  global respcontainer
  
  # create an empty dictionary
  respcontainer = {}
  

  inline_script_1

  global respcontainer
  
  # add the current response to the dictionary, with the proper question number
  respcontainer[self.get("qnr")] = self.get("response")
  

  inline_script_2

  global respcontainer
  
  # here you implement how you select the question number,
  # this depends on what you want to do with the experiment
  qnr = 
  
  # set it so that OpenSesame items can see it
  self.experiment.set("relevant_resp", respcontainer[qnr])
  

  After this, in your item for answer dependent showing, you can use [relevant_resp] to access the answer that a participant gave on the questionnaire at 'qnr'.

  Hope this helps! In case you're doing anything different from what I assumed here and you can't figure out how to fix that, just let me know!

  Edwin