There are two ways that questions can influence subsequent questions: order bias and order misunderstanding.
Order bias results when we order questions such that the first question influences the response to the second question and thereby leads to inaccurate responses to that second question. A classic example is to be found in a Gallup poll conducted in September 1997 (Moore, 2002). The poll asked the following two questions:
“Do you generally think Al Gore is honest and trustworthy?”
“Do you generally think Bill Clinton is honest and trustworthy?”
When asked about Al Gore first, 68 percent of respondents indicated that Al Gore was honest and trustworthy. However, when asked about Bill Clinton first, only 60 percent of respondents indicated that Al Gore was honest and trustworthy.
Understanding the question was not the problem. The problem was that the question about Bill Clinton affected responses to the question about Al Gore. The effect was not random but directional. It is a clear case of order bias.
Order misunderstanding occurs when respondents misunderstand a question because they bring forward the parameters of a previous question or set of questions. The transition from one question to the next, in other words, is not clear.
Consider the following two questions:
Q1: Did you, yourself, take an aspirin within the past 7 days?
Q2: How many vitamin bottles are in your household?
The first question asked about “you, yourself,” while the second question did not make clear whether the question was still about you only or whether it referred to everybody in the household. Some respondents will answer for everyone in the household, while others will carry forward the mindset from the previous question and think only about themselves. Some respondents will misunderstand the second question because it does not clearly indicate who is being addressed.
The second question, in other words, did not do enough to establish its own mindset. The wording did not make it clear enough that the components of the question had changed. This is a clear case of order misunderstanding.
In cases of order bias, respondents understand the question but allow the first question to influence their response to the next question. In cases of order misunderstanding, some respondents don’t understand the second question. The transition between the first and second questions is not clear enough.