Back to top

Creating web surveys

Web surveys are definitely one of the cheapest and most powerful ways of surveying. However it often happens that questionnaire is not well prepared and therefore we can’t get answers to the questions we look for or we can’t get enough responses. Below you can find some tips on how to make your questionnaire even better.

  1. First step of any (also web) research is “offline” work. Don’t start by just writing questions, but think what are the goals of the research. Ask yourself what interests you. Write down goals and form questions from them. Put yourself in respondent’s shoes and ask questions in a way understandable to anyone; think about what respondents find interesting and which questions will be successful. Then contact some individuals from the target population and ask them to comment on the survey. It is always welcome to find out how questions are perceived by people you want to research.
  2. Although it seems easy to just sit behind the computer, write questions and activate survey, don’t do that! Because of this simplicity it is easy to think that you don’t need clearly defined goals before start of the research. But that is far away from the truth. Invite you colleagues and initiate brainstorming!
  3. Imagine your survey as a conversation. Questions should run logically – questions from the same thematic section should be put together. Questions should be divided into groups according to theme. You can also use title before these questions. (eg. “Demographic question” before demographic questions etc.). Start with simple, broad questions and later go to more specific ones. Don’t forget: It is best to put demographic questions at the end.
  4. Before starting surveying, test the survey. It can be tested on some colleagues or a focus group can be performed.
  5. Access to the survey should be simple. Don’t require registration for filling out survey. It is best not to require from respondents to install new programs or plugins.
  6. Survey should be short and focussed, because that increases the number and quality of the responses. Short surveys have higher response rate, because respondents often loose interest when filling out a survey or just stop filling it out. Ask only question that are important for the goals of your research. Never include questions that would only be inserted because you find them interesting. It is better to focus on one single goal of the research than to perform long, broad survey covering multiple themes. Research show that ideal length of a survey is shorter that minutes. Surveys between 6 and 10 minutes are acceptable. If survey is longer than 10 minutes, we start to witness mass termination of responding.