SurveyFactory Documentation Documentation Restricting multiple responses per respondent
Restricting multiple responses per respondent
Revision as of 23:04, 24 March 2013 by Matt (talk | contribs) (1 revision)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Each publication of your survey will include options that allow you to limit the number of responses that each respondent may submit. Many of these controls are only best efforts, and cannot completely stop a respondent from submitting more than one response.

The following is a list of the ways in which you can limit the number of responses per respondent and also some notes on exceptions to the rule, where extra responses may be accepted.

HTTP Cookies

This restriction is available for the link publication and will be automatically used unless you specify otherwise.

One Response per Respondent

When you have set one response per respondent, the default behavior is to force the respondent's browser to have cookies enabled and then limit that respondent to submitting only one response by setting cookies in their browser. You can always change the usage of cookies to optional, or disabled, by checking the HTTP Cookies access restriction in your publication's Respondent Access and Submission Controls section.

Multiple Responses per Respondent

When you have set multiple responses per respondent for the publication, an unlimited number of responses will be allowed by default. If you check the HTTP Cookies access restriction in your publication's Respondent Access and Submission Controls section, a Responses per Browser field will appear. This field gives you further control over the frequency and total number of responses you would like to allow.

The first option controls how many total responses can be accepted or completed over the life of the survey publication. The second option sets how many responses can be accepted or completed over a specific period of time. The two options can be used together, and in the case of conflicting settings between the two, the more restrictive setting of the two will be enforced.

Note that if you delete a response, that will allow the respondent to retake the survey. Similarly, if the respondent has completed a response, but it is left pending and not accepted and charged to your account, it may be deleted after 30 days. At that point, the respondent would be able to complete the survey once again. Also note that cookies are not entirely enforceable as a respondent can always manually remove them from their web browser or turn cookies off entirely.

IP Addresses

By default, SurveyFactory will make no attempt to block mutiple responses by a respondent using the IP address. IP addresses are not a perfect one to one match on your respondents. Some services will put thousands of users behind a single IP address, while other services will assign a different IP address to a user on each request.

If you want to restrict the number of responses submitted by the same IP address, you may do so by checking the IP Address access restriction in your publication's Respondent Access and Submission Controls section. When you do so, you will be able to control the total number of submissions per IP address over the life of the survey, or over a specified period of time.

Email Addresses

The link publication type will allow you to enforce email verification for your respondents. If you have set only one response per respondent, they will be blocked from submitted more than one response with the same email address. You should read the exceptions section of this page for some important caveats to this rule. If limiting the number of responses per respondent is important, you should always perform the email verification prior to the start of the survey, or only accept responses once they are completed.

When you have selected to allow multiple responses per respondent, two additional options will appear allowing you to limit the number of responses over the life of the survey, or over a specific period of time. If the two options conflict, the stricter of the two will prevail.

Invitations

Invitations are by far the most effective way of ensuring you get only the number of responses per respondent that you desire. Using a data set publication type, you will create invitations for each user and then use these custom links to invite the participants to your survey. If you have selected one response per respondent, that invitation is only good for a one time submission.

However, you can also set multiple responses per respondent, in which case additional fields will appear allowing you to set the total number of responses over the life of the survey and the number of responses a participant may submit over a specified period of time. Therefore, you could allow up to 12 submissions total, but only one every 30 days so that the respondent can submit up to one response per month for a year, but never 2 responses in a 30 day period.

Because data set publications only allow responses to come through an invitation you have created, this is the most certain way to track who is responding to your surveys and limiting their responses. The only downside is that you must know who your potential respondents are before they may take your survey.

Exceptions

Responses will begin to block as duplicates once they have been accepted and charged to your account or submitted as complete by the respondent. Once a response has reached that stage, it can no longer be marked as duplicate, but may be used to block additional responses. This can occasionally cause more responses to be accepted than you had initially intended. The following scenarios help highlight this inconsistency.

Email Addresses

Let's assume that you have set email verification for your publication and also have set only one response per respondent. Normally, this will block any attempts to submit more than one response using the same email address. However, let's also assume that you have chosen to do email verification at the end of the survey, but you want to have responses automatically accepted and charged to your account after the first couple of questions have been completed.

The problem of blocking duplicate responses here, is that the email address does not even get added to the response until long after it has been accepted and charged to your account. Since duplicates will never be blocked once they are accepted, no responses for this publication will get blocked as duplicates due to the verified email address.

IP Addresses

IP addresses are not always static for a user and can change from request to request, or the user could even move from their home computer to an office computer while taking the survey. Therefore, there are situations where it is possible that a response will be accepted and charged to your account, but then later accessed via a different IP address which had already submitted a response to the survey.

For example, assume a user completes the response at both home and work, which have two different IP addresses. You have also setup the survey to allow respondents to modify their response after submission. If the user then returns home and decides to edit their completed response from work, their home IP address will be assigned to the response. That will cause a duplicate, but no error will be thrown, as both submissions have already been "completed" by the respondent.