Questionnaires can be self-administrated, i.e. administrated to the potential responders by mail, E-mail or Internet. They fill in the questionnaire without the help of a member of the investigation team.
Self-administrated questionnaires have the advantages to be cheap and easy to administer. They preserve confidentiality and can be completed at the respondent's convenience without the influence of the interviewer. However, self-administrated questionnaires can result in low response rates, because people feel less motivated to respond. Questions can be misunderstood easily without the help of an interviewer. In addition, there can be a considerable time lag between the first sending of the questionnaire and the collection of all questionnaires, particularly if questionnaires are sent out by mail.
In recent years, web-based questionnaires have become increasingly popular among epidemiologists. Advantages include the increased timeliness and potentially higher response rate compared to mailed questionnaires. A web-based questionnaire is easily set up and the corresponding link can be sent to a large number of persons at low or no expense. Also, investigators using web-based applications can collect data without having to do the time-consuming data entering themselves. Web-based questionnaires are particularly useful in popluations with high internet literacy such as young people or company employees. It is less useful for surveys in the general population as persons with little computer literacy are unlikely to respond. Some web-based applications are free while others need to be purchased [1, 2, 3, 4].
Interviewer-administrated questionnaires can be used in face to face or telephone interviews. They can be used easily to interview less literate or illiterate people. Interviewers can help to clarify ambiguous questions and the answers are available more quickly than in a mailed questionnaire. The most important disadvantage is the bias which can be introduced by different interviewers' perceptions and interpretations of the answers (interviewer bias). Also, in large surveys, more than one interviewer is needed to carry out all interviews, thus resulting in an increase in needed resources. Questionnaires need to be short (up to a maximum of 10 min), especially for telephone interviews. Also, telephone interviews are not the optimal setting to ask about sensitive issues such as sexual behaviour.
1.Survey Monkey http://www.surveymonkey.com/
2. LimeSurvey http://www.limesurvey.org/
3. QuestionPro http://www.questionpro.com/web-based-survey-software.html
4. QuestBack http://www.questback.com/