The Danger of Online Surveys

At Research First we are often asked by clients about online surveys. Our advice is always the same – if you want to survey a group of people who are regular online users, online surveys are a great way to get to them quickly and cost-effectively. But if you want to reach a general population, online surveys are highly likely to leave you with a very inaccurate picture of that population. This is because of a problem researchers call 'self selection bias', and occurs where the people who complete a survey differ in important ways from the population as a whole.

As a result, the survey results are not an accurate representation of the population’s attitudes because only those motivated to participate completed the survey questionnaire. Researchers call errors like self selection bias 'systematic' because they are a product of how the research is designed.

To demonstrate just how much bias there can be (and how much the choice of research method can affect the results you get), consider this example. Recently Research First conducted a project that matched (i) a conventional mail survey with (ii)

Research First, Level 1 / 320 Manchester Street, PO Box 5596, Papanui, Christchurch 8542
Phone: (508) 4RESEARCH (508) 473 732 info@researchfirst.co.nz