In the 'EDIT' - 'Settings' - 'Respondents access' tab, you can set cookies, restrict access to the survey and set a password to access the survey.
1) A 'cookie' is a code that is placed on the respondent's computer by a server (e.g. 1KA), so that 1KA can identify the respondent when they try to complete the survey again. You can choose until when to save the cookies; the following options are available:
- 'Until the end of the browser session', where the cookie will be deleted as soon as the respondent closes their browser (Internet Explorer, Mozilla Firefox, Opera, etc.); therefore, if the respondent comes back within the same session, they can make further corrections. Caution: if a second respondent enters in the same session after the first respondent, the cookie will still be saved from the first respondent, which will cause issues with the survey!;
- 'For 1 hour' means that the cookie will be deleted after one hour from the first time the respondent clicked on the survey URL. The same warning as above (browser session) applies;
- 'For 1 month', the cookie will be deleted one month after the respondent first clicked on the survey. The same warning as above (browser session) applies.
If you choose to save cookies for longer than until the end of the survey, then you can further set the behavior of the survey when the same user returns:
- Starts from the beginning, with the respondent seeing the answers they have already given;
- Starts at the last submitted page, i.e. returns to the page where they left the survey.
You can also allow a user who has already completed your survey (a user who has completed the survey) to edit their answers afterwards, but this means that the data and analyses will also change, so in most (but not all) cases, it is better to choose that respondents can not edit their answers.
Regardless of the cookies (whether the option "until the end of the questionnaire", "until the end of the browser session", "for 1 hour" or "for 1 month" is selected), you can also specify what the respondent's options are while they are answering the survey:
- Can subsequently edit their answers: when they return to the previous page, they can change their answers. These then overwrite the previously entered values in the database when clicking on 'Next page'.
- Cannot subsequently edit their answers: even if they click on 'Previous page' or on the 'Back' button in the browser, they can only see their answers but cannot change them.
2) 1KA also enables the recognition of respondents who are registered 1KA users, but they must be logged in to the 1KA installation while completing the survey. You have two options:
- 1KA recognises the user as a respondent, i.e. a person (logged in to the 1KA installation) who answers your survey. This setting is suitable for internal administrative processes, e.g. when members of a collective fill in some data (e.g. address, contact) and can always correct their answers;
- 1KA recognises the user as a person entering other respondent's answers (data input), i.e. an interviewer entering answers, for which 1K provides the appropriate interface adjustments. This option is particularly useful in the case of mass entries, or when a field survey is carried out and then interviewers enter the collected answers, to see who has entered which answers. The same applies to telephone surveys.
The above two cases are rare; the default option is of course that 1KA does not recognise the user of the 1KA installation. Turning this option on can have serious consequences. If the 'Yes, as data input' option is enabled, the user has additional settings available for Questionnaire Entry.
3) IP address restrictions
To prevent the same person from answering your questionnaire multiple times, you also have the option of blocking the IP address of the device from which the respondent is accessing the survey. You can block it for 10 minutes, 20 minutes, 60 minutes, 12 hours or 24 hours. However, this means that you can unintentionally also block another person using the same computer from answering the survey. It only makes sense in rare cases, e.g. voting, where there is a risk that someone is submitting multiple responses with a bot or a script. Of course, respondents can also circumvent IP restrictions by using virtual IP addresses. In general, however, this is a risky option which does not solve the problem of duplicate responses, but generates the risk of preventing a large number of relevant users from completing the survey, as often whole organisations or departments have the same IP address, for which there are many computers and potential respondents. More >>
4) Password to access the survey
Access to the survey can best be restricted via a password. You can generate multiple passwords and thus create groups of respondents that can be separated in analyses or used as a basis for IF conditions. In this case, the password serves as a safeguard to ensure that only those you want to complete the survey do so. A single password can be created for all respondents, or a password for each respondent. If you want to enter several passwords at once, select the 'Password mass insert' option. The use of passwords is only available for users of the 2KA and 3KA packages for individuals, and for users of the business packages for groups
In the 'PUBLISH' - Invitations' tab, you can also set up personalised codes to access the survey, via the email invitation interface available to users of the 3KA package for individuals, and users of the business packages for groups. Usually this code is then used in the email invitation, preferably automatically (in the survey URL) so that it does not need to be entered by the respondent. Alternatively, we may ask the respondent to enter the code manually. Similarly, the code can be sent in other ways, e.g. in a letter, and the respondents can type it in. The generation and use of individual codes is described here.
The use of passwords is in principle stronger and overrides the other settings for respondent access restriction (cookie, IP number, registered user identification), but interference may occur, so we strongly recommend to use only one of the above mentioned ways to identify respondents: cookie, IP number, registered user identification or individual password.