You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Problem:
Interviewer behaviour is expected to change as they familiarise themselves with the questionnaire, Survey Solutions, etc. Interviews created earlier during a survey may differ somewhat to those towards the end.
We have explored the following features:
days since start of survey (take minimum answer set)
sequential number of interview (sorted by time) by interviewer
We have not found any conclusive patterns yet. In our experiment, all artificially created fakes were produced after the real interviews.
Solution:
Investigate further once more testing data is available.
Maybe allow user to specify the expected number of days of the survey, so days since start can be expressed relatively.
The text was updated successfully, but these errors were encountered:
Problem:
Interviewer behaviour is expected to change as they familiarise themselves with the questionnaire, Survey Solutions, etc. Interviews created earlier during a survey may differ somewhat to those towards the end.
We have explored the following features:
We have not found any conclusive patterns yet. In our experiment, all artificially created fakes were produced after the real interviews.
Solution:
Investigate further once more testing data is available.
Maybe allow user to specify the expected number of days of the survey, so days since start can be expressed relatively.
The text was updated successfully, but these errors were encountered: