How to craft surveys that maximize data integrity

Editor’s note: Roxanne Salen is an account executive at Denver research firm The Praxi Group.

When it comes to quantitative data integrity, much of the quality is derived from designing a survey that is engaging and simple for respondents to complete. All too often, in efforts to extract as much information as possible from one survey, data integrity becomes an afterthought. By crafting a respondent-friendly survey, you enhance the quality of the data that will ultimately support major business decisions for your company. Here are a few tips to uphold the quality of data you collect:

Reasonable questionnaire length
Oftentimes the demands of stakeholders drive researchers to try and gather answers to every outstanding business question in one very long survey. However, a long survey is detrimental to respondent engagement and can reduce (or even eliminate) thoughtful responses. Consider breaking the topic into two smaller studies. If that’s not an option, go over the survey with a fine-tooth comb to make sure you’re only asking the most actionable questions for your research team and stakeholders. Or break up the monotony of questions by creating engaging question formats such as a drag-and-drop task or sliding-scale graphics.

Answerable questions
Design questions that allow respondents to accurately recall their experience. When a question is based too heavily on recall, your answers (and data) may be skewed.

Use consumer-friendly language
Remove industry-speak from the questionnaire. Use clear and simple terms that reflect the way your customers think about and use your products.

Give explicit instructions for grid-format questions
Always indicate if the respondent should select an answer from a row or a column, as well as if they should select a single response or all that apply. Again, the more clarity your survey provides, the easier it will be for your respondent to provide an accurate response.

Logical order
Questions should be asked in the order that the respondent would have experienced the interaction or event; ask for the occasion trigger first and then in chronological order to ensure accurate recall of perceptions regarding each component.

Craft questions that are not open to too much interpretation
Consider that each respondent holds unique viewpoints. A question that uses vague language or even one that inadvertently triggers multiple emotional responses could result in an answer that does not meet the original goal of the question crafted by the researcher. Reread your question and ask yourself if a respondent could interpret it differently than you are intending.

Market research lends itself to the support of many company and product decisions. However, the data is only as good as the responses it is comprised of; be mindful of the respondent experience and you’ll maximize your data integrity.

This entry was posted in Consumer Research, Interviewing, Market Research Best Practices, Quantitative Research. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>