surveywritingThe quality of the data you collect through quantitative research directly reflects the quality of your questionnaire. Investing the appropriate time to construct a well-designed, effective survey will ensure that you can confidently count on your data to make sound business decisions.  However, if you do not invest time and thought in drafting your survey, regardless of the number of respondents, your data may not be reliable.

Below are some guidelines to help you construct a well-designed survey:


    1. What are you trying to answer with this survey?
    2. Clearly articulate 2-4 broad objectives the survey should address, and use them to guide you as you develop specific questions.

    1. Who should be answering the survey and providing data?
      1. To determine your survey’s audience, consider your study objectives, category, product, advertising and behaviors you are interested in learning about.
      2. Similarly, consider the population of respondents you are looking to target for your survey – what demographic and behavioral profiles do they need to meet? As a note: narrower, more targeted audiences can result in higher respondent cost and longer field time.

    1. When writing each question, be mindful of the survey length, consider how its data will be used and and if it will contribute to your research objectives.
      1. A good exercise is to draft the survey with all possible questions and trim the survey down to 20-25 questions that best address your goals.
    2. Two elements to think about while constructing your questions:
      1. Types of questions to include: There are many ways to ask any question, and different ways can provide you with different results. Choose the most appropriate structure based on the objective of that question. The two most widely used types of questions are:
        1. Closed-end questions:
          1. These questions provide participants with answer options or scales for their responses. There are several types of close-ended questions — multiple response, single response, scales, ranking, etc. — and each provides different results, and supports different objectives.
        2. Open-end questions:
          1. Respondents type in their own words to respond to a question. This is often used to avoid exposing respondents to options that might guide their responses.
      2. Question bias: Ensure that the way you phrase your questions does not inadvertently guide respondents’ answers. For example, asking respondents to give their opinion on a product and providing them only a list of positive attributes will result in a falsely-positive assessment of the product. Similarly, using a scale that is unbalanced, leaning either positive or negative, will result in data that does not represent how consumers really feel.

    1. The survey flow is as important as the questions themselves. As a rule of thumb:
      1. Separate your survey into different sections, with each section addressing a specific topic. This provides a flow that is easier for respondents to follow and understand.
      2. Start broad and then narrow: ask broad opinion questions – such as overall satisfaction or overall opinion – at the beginning of the survey to get respondents’ true top-of-mind reactions before they are exposed to more specific questions and options that might influence their later responses.

Writing a quality, unbiased survey is one of the most important steps in gathering useful data. It’s OK to ask for help or leave the survey writing to the experts. At Lab42, we are skilled in all aspects of survey writing and can provide consultation on a survey you already have or construct your survey from scratch. Just remember a well written, neutral survey will provide your business with the highest quality data and allow you to make the most informed business decisions.

Comments are closed.