Double Negative

A double negative refers to the use of two negatives in one statement or question. In questionnaire design, this is almost always a situation to be avoided. A double-negative usually creates an unnecessary amount of confusion in the mind of the respondent and makes it nearly impossible for the researcher to accurately determine what respondents were agreeing or disagreeing to.

Such a question can increase item nonresponse by increasing the percentage of respondents unable to understand the question. A more insidious problem is an increase in the number of responses from people who have misunderstood the question and responded based on that misunderstanding. Both item nonresponse and misunderstanding contribute to the type of survey error know as measurement error.

Sometimes the questions sound sufficiently confusing that no serious researcher would include them in a questionnaire. For example, Are you likely to vote for or against a candidate who is opposed to the proposed ordinance to prohibit smoking in public places?

However, the second of the two negations often appears in the answer options. It is in these cases when a double negative question may seem perfectly clear to researchers, particularly those who are caught up in issues of the moment. For example, members of the media who support release of the names of jury members might wish to ask:

Perhaps the most infamous example of a double-negative occurred in November 1992 in a survey conducted by the prestigious Roper Organization, a respected survey research center founded in 1947. Roper was commissioned by the American Jewish Committee to conduct a survey of adults in the United States to measure public attitudes and beliefs about Jews. The following question slipped through the usual quality control steps:

The published results of the survey implied that one third of adults in the United States felt it was possible that the Holocaust never occurred. The outrage and confusion resulting from the release of the study results prompted several other studies conducted with the specific aim of avoiding the double-negative problem. The other studies worded their survey questions to avoid double negatives. Results from these studies revealed that fewer than 10% of the population felt it was possible the Holocaust had never occurred.

The double-negative problem is difficult to avoid in questionnaire design for attitude and opinion surveys because researchers are often presenting choices to respondents that represent positives and negatives. One approach to avoiding double negations is to make every effort to present the issue without using any negative statement. If the double negative must be used, cognitive interviewing should be employed during pretesting to ensure that respondents have a clear understanding of the question.