Five Last Steps in Writing a Questionnaire
In quantitative survey research, the questionnaire affects nearly everything else that is critical to a study: the data, the analysis, and the findings. And because of this, writing surveys is the component of research that requires more on-the-job experience than any other. It requires thinking on multiple levels about all other aspects of the research process even as you are trying to formulate questions that will answer what you want to know.
To help you write better surveys, Versta Research’s fall newsletter describes Five Last Steps in Writing a Questionnaire. Beyond the standard best practices typically laid out in textbooks and methods courses (use neutral language, avoid double-barreled questions, and so on) our newsletter article offers recommended best practices for finalizing a survey before it goes to programming for design or layout. It provides an easy set of guidelines for those crucial last components that can make or break the success of a survey.
Here is a brief overview. After all your questions have been carefully formulated, revised, and approved for production, do the following:
- Review Your Don’t-Knows. It may be tempting to add a “don’t know” or “no opinion” option to every question, but research shows it dilutes the insights you can get from your data.
- Review Your Other-Specify Boxes. Your clients may ask you to include them, but good researchers know they add ambiguity and rarely yield useful information.
- Review Your Randomization. Randomization can help or hurt data quality, so never do it automatically, and check that you have specified where it is needed.
- Review Your Scales. This is where you need to think ahead to analysis and reporting, because sometimes four points will work better, sometimes five, and sometimes eleven.
- Review Your Skip Logic and Programming Instructions. There are always mistakes, and mistakes will mess up your data. You must find them now and fix them before it happens.
A final run-through (or five final run-throughs!) of your approved questionnaire that focuses on these elements will ensure you are giving careful consideration to issues in survey design that likely got pushed aside as you focused on so many other aspects of getting your questionnaire right.