Data Protection in Surveys

Anonymity of survey participants

Maintaining anonymity is a very important issue for the Unit for Quality Assurance. The software used to carry out the surveys (Evasys by Electric Paper) allows surveys to be carried out anonymously or not. In both cases, participants are informed of the survey's anonymity, or lack thereof, in the accompanying notification. At the University of Vienna, Evasys has so far only been used to carry out anonymous surveys, where each survey participant receives an automatically generated code that authorises him or her to take part in the survey exactly once.

The Unit for Quality Assurance has no knowledge either of which codes are assigned to which survey participants, or of which answers were given by which survey participants. Unfortunately, this does at times mean that we are unable to resend a code to a participant who has lost his or her original code (other than via the reminder function to all survey participants). A new code cannot be issued either, as it is not known whether the old code was in fact used. In general, we have no possibility of viewing or manipulating the original database.

Furthermore, it is not possible to identify one person in the data set by analysing specific identifying characteristics (e.g. gender, age, semester, etc.) until only one person per combination of characteristics remains, as doing so would require access to the raw data, which is why the Unit for Quality Assurance does not release the raw data on principle. In this way, course instructors can view the reports via their system access, but are unable to download the raw data.

The identification of individual persons is only possible in paper and pencil surveys, where the participants' hand-written commentaries on the questionnaires themselves could be compared to other hand-written texts (e.g. exams, registration sheets, etc.). Theoretically, one other possibility for identifying individual persons exists; by ascertaining the use of specific word sequences, expressions, arguments or even spelling mistakes attributable exclusively to one person, it would be possible to identify that participant's survey responses.

Evasys data protection guidelines

Online surveys are carried out using the software Evasys by Electric Paper (www.electricpaper.de). Electric Paper has issued the following guidelines concerning data and user protection:

Viewing the survey results

It is in conjunction with course evaluations in particular that the question repeatedly arises as to who is permitted to view the results and/or to who the results are sent:

• The results are automatically sent exclusively to the course instructors. Instructors are able to view available course evaluation results for their own courses on ongoing basis via their personal accounts.
• At the end of each semester the directors of studies receive a list of courses that were not evaluated despite the requirement to do so. The more, they receive a summarised version of course evaluation results, where the average value for the question "All in all I think the course is…" appears for each course. Moreover, directors of studies are able to view all course evaluation results for the respective directorate of studies via its account.
• Upon request, deans and members of the Rectorate also receive the data on course evaluations.

In all other matters (e.g. for professorial appointment procedures or in the context of contract extensions) the Unit for Quality Assurance provides information on the availability of course evaluation results. The results, however, are only passed on with the approval of the respective course instructor.