Abstract

Leading questions can be a significant contributor to a non-sampling error in survey research. Leading questions can be subtle, even inadvertent. How can social scientists remove such leading questions from their own questionnaires? Tapping the wisdom of the legal profession may be part of the answer. Leading questions are often disallowed in US courtrooms. Therefore, US courts, jurists and legal scholars have devoted attention to exactly what makes a question a leading question, literally for centuries. Here, we explore what that legal debate has to teach modern-day social science researchers. By cataloguing and categorizing over two dozen legal definitions of “leading question,” dating back to the eighteen-hundreds, we find three dominant tactics, or pathways, which convert a question into a leading question. First, strategic use of facts may nudge the respondent towards a specific answer. Second, the very form of the question can suggest an answer. Last, the question’s frame (context, or questioner intonation) may steer the respondent. These concepts are explained and examples are provided. Many years of legal practice have demonstrated there are times when leading questions are allowable, when best avoided, and when there are particularly insidious. These are discussed, and guidance for social scientists is presented.

Keywords