Questionnaires and web experiments are a crucial method of epidemiology that provides crucial information on the condition of public health and disease. They are a common method of collecting data that are typically less costly and time-consuming than face-toface interviews, mail-in questionnaires, or automated phone menu systems. However, questionnaires and Web experiments have a number of limitations that need to be addressed in order to ensure valid and reliable results.
A questionnaire can be affected by response bias. This is the tendency of respondents to answer questions based on their personal opinions instead of research goals. Furthermore, the design of a questionnaire can influence responses in a variety of ways. For example the language used in the questions can affect whether respondents understand and interpret the questions in the same way (reliable) or whether you can determine the topic you’re interested in (valid), or are able to answer in a timely manner (credible).
Respondents may also experience survey fatigue or a lack of interest in the questions being asked, which reduces the likelihood of them providing honest answers. In addition, the absence of incentives or compensation can dissuade respondents from taking the time to complete the questionnaire.
Online questionnaires can also pose an issue for certain experimental designs such as studies of reaction time or position. The variability in browser settings screens, sizes, and operating systems makes it difficult to control and measure the same variables across different participants.
The bottom line is that Web-based surveys can only be accessed by those who are keyboard and Internet knowledgeable. This excludes a substantial segment of the population. It is also difficult to Web researchers to report on participants after the experiment window has closed.