Most efficient way to remove bad respondents? | Experience Community
Skip to main content
Question

Most efficient way to remove bad respondents?

  • August 2, 2021
  • 4 replies
  • 388 views
D
Forum|alt.badge.img

I have a survey that's fielding with commercial sample, and based on reviewing open end responses, I have 50+ respondents that are fraudulent / not giving good responses, and I want to set a flag to identify them in the data. I have been setting up a big filter: "if any of these are true: userID=..." and then putting all 50+ IDs into the filter, each on a separate line, then editing the filtered cases.
Setting up that filter is a pain, is there a way to do it as a batch? I think I could download the dataset, add a variable that identifies bad cases and then re-upload, is that true? Are there any other faster / more efficient ways to approach this?

    4 replies

    A
    Level 7 ●●●●●●●
    Forum|alt.badge.img+22
    • ahmedA
    • Level 7 ●●●●●●●
    • August 2, 2021

    Create a new field in the data and analysis tab, use the filter function to filter out the relevant ones and then add a value to them.

    D
    Forum|alt.badge.img
    • Dan_314159
    • Author
    • August 6, 2021

    https://community.qualtrics.com/XMcommunity/discussion/comment/39490#Comment_39490Hmm, I see this is showing as an "accepted answer," but in fact it does not answer my question. The problem is the filter requires one statement / one line per respondent, because I'm filtering by unique user IDs. This is heavily manual -- I need to copy/paste one ID per line. Is there any way to filter all my bad IDs in a batch?

    A
    Level 7 ●●●●●●●
    Forum|alt.badge.img+22
    • ahmedA
    • Level 7 ●●●●●●●
    • August 7, 2021

    Have you tried this: https://www.qualtrics.com/support/survey-platform/data-and-analysis-module/data/add-new-fields/manual-variables/#TaggingMultipleResponses

    D
    Forum|alt.badge.img
    • Dan_314159
    • Author
    • August 9, 2021

    Hmm, thank you, but I guess what I still don't understand is: if I have identified several dozen cases out of 1000 or so that gave bad open ends, what's the best way to filter down to just those bad cases? I know how to tag them all once I have them selected, but I don't know a good way to select them other than by creating a big filter with one line for each respondent ID. How can I select just those fraudulent respondents so I can apply tags?

    Terms & ConditionsAccessibility statement