Pause and Continue Questionnaire Completion

When respondents access the link to a questionnaire, SoSci Survey starts a new interview – linked to a CASE number in the data set. As soon as answers are entered in the online questionnaire, SoSci Survey saves them in the interview.

This guide addresses how respondents can pause and resume completing the survey later.

Important: Respondents' answers are only saved (for the vast majority of question types) when they click “Next” or “Back”. So if a person answers questions, does not click “Next”, and closes the browser window, the answers from that page are not saved anywhere.

To continue an interrupted interview, SoSci Survey must assign the questionnaire call to an interview. There are different ways to do this:

• The link can directly contain a so-called “token” for a single interview, recognisable by an i=… in the link.
• It can be a link personalised by means of a participation code with s=….
• It can be a personalised link from a serial mail or a participation URL, these contain a d=…
• Or SoSci Survey can store a cookie with the interview token in the respondent's browser.

The cookie option can be activated in the Project settings in the Privacy tab. It is deactivated by default because cookies have a bad reputation in terms of data protection due to their use in the advertising and tracking context.

Where does an interviewee get one of the links described above if they do not want to work with cookies?

The interview-specific link with an i=… appears in the address bar of the browser when “Next” is clicked for the first time in the questionnaire.

Note: If the questionnaire is completed on a publicly accessible device, you should deactivate this function in the Project settings in the Data protection tab. Otherwise, later participants may be able to access the answers of previous participants. This is because the interview-specific link is saved in the browser history (chronicle).

• Under Compile questionnaire in the Settings tab, you can activate a button to pause the interview. This button leads to a function that sends the interview-specific link to the interviewee by e-mail or explains how to save it in the bookmarks.
• The mailResume() function also sends the interview-specific link by e-mail. However, calling this function only works in conjunction with serial mails or an opt-in question.
• You can also use the placeholder %caseToken% to display the interview-specific link in the questionnaire, e.g. with the following HTML code.

<p>
Please save this link:
<a href="https://www.soscisurvey.de/PROJEKT/?i=%caseToken%">www.soscisurvey.de/PROJEKT/?i=%caseToken%</a>
</p>

SoSci Survey offers various functions to restrict the group of participants for a survey. These include personalised links (see also Access Restrictions and Prevent Multiple Submissions by the Same Respondent). Personalised links can be created in three ways:

• Access Codes
• Serial mails, see Send Mailings
• Participation links, which can be created in the required quantity under Invitation Mailings → Access URLs.

Privacy: Serial mails and participation URLs technically utilise the same functionality. With regard to the assignability of addresses (see Privacy Mode for Address Entries), participation links function like anonymous address entries. However, if the assignment of participation links to address entries is stored somewhere, it cannot be ruled out that an assignment to an address entry can be made by merging this list with server log files.

• When a personalised link is called up, SoSci Survey first checks whether an interview already exists for the link. If no interview exists, a new interview is started.
• For serial mail and participation links, SoSci Survey checks in the next step whether an existing interview may be continued or whether a new interview should start each time the link is called up. This setting can be defined for serial mail in the Questionnaire link tab with the Multiple participation option.
• In the third step, the setting of the questionnaire for Continue interview is checked, to be set under Compose Questionnaire → tab Settings.
• If the questionnaire allows a continuation, the variable MISSING may be checked in the existing data set. This value indicates the percentage of items and questions that have already been asked but not answered (see Additional Variables in the Data Set). If the value is below the limit set in the questionnaire, the interview is continued at the point at which the browser window was last closed.

If MISSING is higher than the limit set in the questionnaire, the behaviour differs slightly depending on the type of personalisation.

• For participation codes, a new, empty interview is started so that several cases with the same SERIAL appear in the data record.
• For serial mail/participation links, the existing interview is opened but jumps back to the first page. The answers from the first participation are pre-filled in this case.