Related Experiment Videos
Factors Influencing Response Rates in National Internet-Based Questionnaires: Analysis of a Large National Survey
Søren Birkeland1,2, Anders Haakonsson1, Sören Möller3
1Odense University Hospital, Open Patient data Exploratory Network (OPEN), J. B. Winsløws Vej 21, 3rd floor, 5000 Odense, Denmark, 45 5142 1413.
Background:
Digital mailbox platforms established by authorities for secure communication with the general populace are increasingly available for distribution of surveys. While such solutions constitute an attractive way of easily distributing surveys, nonresponse remains a critical problem, warranting measures to increase participation.
Objective:
We aimed to investigate questionnaire characteristics predicting participation and the effect of reminders among men in a survey distributed through a national digital mailbox solution.
Methods:
This study used an internet-based survey among men aged 45 to 70 years; the survey was sent out in 2 waves through a digital mailbox used by Danish authorities. Predictors for response were analyzed using logistic regression.
Results:
Of 24,000 invitations, 22,288 participants accepted, with 6756 (30.3%) completing the survey. We found some evidence suggesting a rule of thumb that reminders may add another half to the response rate achieved in a preceding invitation, with 21% and 10% of responses obtained before and after the reminder, respectively (in the first wave). Longer questionnaire variants (odds ratio [OR] 0.77, 95% CI 0.66-0.88; P<.001 [longest vs shortest variants]), questionnaires with attached informational material (OR 0.92, 95% CI 0.85-0.98; P=.02[with vs without material]), or with potentially distressful emotional content (OR 0.91, 95% CI 0.85-0.98; P=.008 [most vs least distressful content]) tended to yield lower response rates. Furthermore, older age was associated with a higher response rate when comparing the oldest (65-70 years) to the youngest (45-55 years) age groups (OR 2.49, 95% CI 2.31-2.70; P<.001).
Conclusions:
We confirmed some previous findings obtained with other modes of survey distribution on the association between response rates and questionnaire characteristics, showing that it is important to pay attention to specific characteristics when designing internet-based surveys distributed through digital mailboxes. Nonresponse may constitute a considerable and increasing challenge, and reminders may help substantially increase participation.
Related Concept Videos
Surveys
Longitudinal Research
Data Collection by Survey
Factorial Design