Anonymization of open questions | Experience Community
Skip to main content

Anonymization of open questions

  • October 11, 2022
  • 3 replies
  • 60 views
A
Forum|alt.badge.img+1

Hi all,
we uploaded some name lists as json files in the data policy section of the adim tab. We want to make sure that no personal reference can be made in the survey responses. Unfortunately, the survey taking experience is quite bad due to long loading times of the survey pages. We assume that this is caused by an overload of lists/words (the list that are enabled for the survey contain about 500k words).
Has anyone ever experienced something similar and is there any possibility to reduce loading times by having so many flagged words or would the best soultion be to reduce the number of words in these lists?
Thanks for your help! 

Kind regards,
Alex

    3 replies

    MattiasM
    Level 5 ●●●●●
    Forum|alt.badge.img+63
    • MattiasM
    • Level 5 ●●●●●
    • October 11, 2022

    Good morning from a sunny Sweden!
    Are you trying to have certain words trigger a message saying "no personal references please" or what is the intended functionality you're after?
    I'm just thinking if the root cause is that you don't want your reports containing potential personal references (what ever that actually means in this case), perhaps you could add flags in text Iq and have the responses flagged after they've been submitted?

    But some additional understanding would be helpful, from my perspective at least. :)
    Best regards
    -Mattias

    A
    Forum|alt.badge.img+1
    • alexschmitt
    • Author
    • October 11, 2022

    Hello Mattias,
    thank you for your fast reply!
    We would like to prevent the names of employees from being displayed in the evaluations. 
    With the help of the sensitive data policy, we can set that names are displayed in the following format as soon as they are entered by a participant: ***** Thus, no conclusions can be drawn about what the participant had typed.
    Since the list of these names that can potentially be flagged is very long, and responses are always checked by the system immediately after they are entered, it takes a very long time to continue with the survey after a text response.
    I hope this explanation helps to clarify :)
    Best regards,
    Alex

    N
    Forum|alt.badge.img

    Hi,

    We are battling with exactly the same problem currently. We have a .json file with with over 900000 names that we would like to use for anonymisation.

    Qualtrics support confirmed these hard limits for the .json files and the topics

    • Topics: You may use a maximum of 25 topics.
    • File Size: The allowed file size is capped at 400 KB per file.

    When we used this approach the there was some performance issue on Qualtrics side and our survey was disabled and we had to remove the topics.

    Has anyone found a solution how to anonymise long list of names from open ended questions?

    Best regards, Nicke

    Terms & ConditionsAccessibility statement