A good question assists in acquiring clear insights and goal-relevant information concerning the target group. I learned this in a friend’s health facility survey. One vital survey question may be: “What did you like or hate about products/services you purchased from us in the past?” This is because by answering such a question, an organization may better optimize its website, services, supplies, and marketing strategies to increase conversions and improve user satisfaction. Creating a survey is not as simple as writing down a few questions on a scrap of paper. To create a high-quality questionnaire, one should write, modify, and rewrite it several times. One should consider several things before beginning to compose a survey to ensure that one acquires the information one need to make important decisions.
Ordinal (or ranked) questions, such as those asking for an answer between “great” and “bad,” or “extremely favorable,” “slightly favorable,” “extremely unfavorable,” or “slightly unfavorable,” are not typically randomized. This is because the arrangement of the categories provides crucial context for the question being asked. In most cases, the order in which these scales are given should be such that respondents may readily locate themselves along the spectrum, but this is not always the case. Questioning half of the sample on whether abortion should be “always legal, legal in most cases, illegal in most cases, always illegal,” and the other half on the same inquiry with the available options written in reverse arrangement, beginning with “always illegal,” does not remove the recency impact but instead spreads it randomly across the target group. The wording of a question is crucial to conveying its meaning and aim to the participant and ensuring that everyone gets the same result. Even slight language changes can dramatically alter participants’ responses.