4 Common Survey Question Writing Mistakes

One of the most common problems with extrapolating from your data has nothing to do with sample size or data analysis techniques. It's the quality of your data. Your survey data is only as good as the questions that you use to create it, and often the questions that survey researchers use cause unusable data that doesn't imply what you think it implies intuitively.

When you're writing your own survey questions using survey software, it's important to make sure that you're avoiding the common mistakes that seem to plague survey researchers. These include: Confusing Questions Questions must in no way be misinterpreted or misread. An example would be “which one of these items isn’t your favorite?” For one person that may mean that it’s their least favorite.

For someone else, it may simply be one they notice but do not like as much as their favorite. For others, they may not see the “isn’t” and read it as “is.” All of these could lead to answers that are incorrect, and thus data that isn’t as valuable. Biased Questions Companies that have strong feelings about their products or services are also prone to biasing their own questions.

This is when the question itself implies the answer. An overt example of bias would be “PRODUCT is the best on the market today, why do you think that is?” For someone answering, your product may not be the best on the market. You’re implying their answer, and/or changing it depending on what they see in your question.

Unclear Question Piggybacking Many question writers rely on other questions in an effort to make answering the survey easier. But in some cases the connection to the original question may not be clear. It’s important to make sure that your survey is clear as a WHOLE, not necessarily the individual questions involved, and that every question will be interpreted successfully by any applicant.

Useless Questions Finally, there is the issue of questions that do not necessarily tell you anything of value. These are questions that are unintentionally vague, like “do you like it?” (where “like” does not indicate anything of substance) or questions that are simply asking something that doesn’t have any relevance to the survey at hand. These questions can affect your data because they confuse it, turning your dataset into a lot of information with significantly less substance.

Question writing is practically an art form, because it has to be completed with a great deal of care for it to be successful for your survey. Take the time to sit and review every question otherwise, your online survey may not be as valuable as you had hoped. Read more on: How to Design Effective Online Survey Questionnaires How to Design HR Questionnaires