Questionnaires are often used for data collection in epidemiological studies and are a relatively inexpensive way to collect information for research purposes. The aim of a questionnaire is to generalize replies from a small sample to a larger population. If people do not reply to the questionnaires (so called “non-responders”), the research results will tend to be less accurate. Non-responders reduce the effective sample size and can introduce bias. Thus, when performing questionnaires a crucial point is to obtain a sufficiently high response rate. Some ways to obtain that are presented here.
First, the odds of response are almost doubled if the respondents are paid for their participation. However, some more inexpensive factors can be easily adjusted to attract the respondents’ interest.
A Cochrane review from 2009 [Edwards 2009] revealed that the odds of response increased when styling questionnaires in a personal manner, such as signing letters by hand. If the response were to be anonymous, it was important that an assurance of confidentiality was declared. Odds of response were also higher when a simple header and pictures were included in both the cover letter and the questionnaire. The pictures should be placed early to interest the respondents. One or more reminders should be sent to people who do not reply, emphasising the importance of a reply and including a statement that many other participants have already responded to the questionnaire.
Distribution of questionnaires by e-mail has become popular. Electronic questionnaires have some advantages over postal surveys. First, it is economical. Once the software is bought, all costs related to distribution are eliminated. For postal questionnaires, the researcher will have to consider expenses for printed paper, envelopes and stamps – both for the questionnaire and all reminders, which is likely to exceed the price of the questionnaire software. Second, data collection is simpler; the respondents enter all data themselves, consequently eliminating all bias related to data input and accordingly also a substantial part of the workload of the researcher. Third, electronic questionnaires can embed interactive elements like hyperlinks, pictures and movie clips in the form. And finally, electronic questionnaire software can administrate the distribution of reminders only to e-mail addresses that have not yet replied, without revealing the identity of the respondent.
Today, electronic questionnaires seem to obtain response rates superior to what used to be achieved by postal questionnaires. However, electronic questionnaires have become increasingly popular and it might be just a matter of time before respondents consider questionnaires as spam and reject to reply to them.
Edwards PJ, Roberts I, Clarke MJ, DiGuiseppi C, Wentz R, Kwan I, Cooper R, Felix LM, Pratap S. Methods to increase response to postal and electronic questionnaires. Cochrane Database of Systematic Rev 2009.
Download NIOM Newsletter June 2015