Question wording is the translation of the desired question content and structure into words that respondents can clearly and easily understand, Deciding on question wording is perhaps the most critical and difficult task in developing a questionnaire, as illustrated by the Census 2000 questionnaire in the opening example, If a question is worded poorly, respondents may refuse to answer it or may answer it incorrectly, The first condition, known as item non response , can increase the complexity of data analysis.25 The second condition leads to response error, Unless the respondents and the researcher assign exactly the same meaning to the question, the results will be seriously biased.26 To avoid these problems, we offer the following guidelines: (I) define the issue, (2) use ordinary words, (3) use unambiguous words, (4) avoid leading questions, (5) avoid implicit alternatives, (6) avoid implicit assumptions, (7) avoid generalizations and estimates, and (8) use positive and negative statements.
Define the Issue
A question should clearly define the issue being addressed. Beginning journalists are admonished to define the issue in terms of who, what, when, where, why, and way (the six Ws).27 These can also serve as guidelines for defining the issue in a question, for an application of these guidelines to descriptive research.) Consider the following question:
Which brand of shampoo do you use? (Incorrect)
On the surface, this may seem to be a well-defined question, but we may reach a different conclusion when we examine it under the microscope of who, what, when, and where, “Who” in this question refers to the respondent, It is not clear, though, whether the researcher is referring to the brand the respondent uses personally or the brand used by the household. “What” is the brand of shampoo. However, what if more than one brand of shampoo is being used? Should the respondent mention the most preferred brand, the brand used most often, the brand used most recently, or the brand that comes to mind first? “When” is not clear; does the researcher mean last time, last week, last month, last year, or ever? As for “where,” it is implied that the shampoo is used at home, but this is not stated clearly. A better wording for this question would be:
Which brand or brands of shampoo have you .personally used at home during the last month? In case of more than one brand, please list all the brands that apply . (Correct)
Use Ordinary Words
Ordinary words should be used in a questionnaire and they should match the vocabulary level of the respondents.P When choosing words, keep in mind that the average person in the United States has a high school, not a college, education, For certain respondent groups, the education level is even lower, For example, the author did a project for a major telecommunications firm that operates primarily in rural areas, The average educational level in these areas is less than high school, and many respondents had only fourth- to sixth-grade education, Technical jargon should also be avoided. Most respondents do not understand technical marketing words, For example, instead of asking,
“00 you think the distribution of soft drinks is adequate?” (Incorrect)
“00 you think soft drinks are readily available when you want to buy them?” (Correct)
Use Unambiguous Words
The words used in a questionnaire should have a single meaning that is known to the respondents, A number of words that appear to be unambiguous have different meanings to different people.29 These include “usually,” “normally,” “frequently,” “often,” “regularly,” “occasionally,” and “sometimes.” Consider the following question:
In a typical month, how often do you shop in department stores?
The answers to this question are fraught with response bias, because the words used to describe category labels have different meanings for different respondents, Three respondents who shop once a month may check three different categories: occasionally, sometimes, and often. A much better wording for this question would be the following:
In a typical month, how often do you shop in department stores?
____Less than once
____1 or 2 times
____3 or 4 times
____More than 4 times
Note that this question provides a consistent frame of reference for all respondents, Response categories have been objectively defined, and respondents are no longer free to interpret them in their own way. _ Additionally, all-inclusive or all-exclusive words may be understood differently by different people. Some examples of such words are “all,” “always,” “any,” “anybody,” “ever,” and “every.” Such words should be avoided. To illustrate, “any” could mean “every,” “some,” or “one only” to different respondents, depending on how they look at it In deciding on the choice of words, researchers should consult a dictionary and thesaurus and ask the following questions of each word used:
1. Does it mean what we intended?
2. Does it have any other meanings?
3. If so, does the context make the intended meaning clear?
4. Does the word have more than one pronunciation?
5. Is there any word of similar pronunciation that might be confused with this word?
6. Is a simpler word or phrase suggested?
The U.S. Census Bureau took great pains to use ordinary and unambiguous words in the Census 2000 questionnaires, which not only improved the response rate but also resulted in more accurate data (see opening example).
Avoid leading or Biasing Questions
A leading question is one that clues the respondent to what answer is desired or leads the respondent to answer in a certain way, Some respondents have a tendency to agree with whatever way the question is leading them to answer. This tendency is known as yea-saying and results in a bias called acquiescence bias. Consider the following question:
Do you think that patriotic Americans should buy imported automobiles when that would put American labor out of work?
This question would lead respondents to a “No” answer. After all, how can patriotic Americans put American labor out of work? Therefore, this question would not help determine the preferences of Americans for imported versus domestic automobiles. A better question would be:
Do you think that Americans should buy imported automobiles?
Bias may also arise when respondents are given clues about the sponsor of the project, Respondents tend to respond favorably toward the sponsor. The question, “Is Colgate your favorite toothpaste?” is likely to bias the responses in favor of Colgate, A more unbiased way of obtaining this information would be to ask, “What is your favorite toothpaste brand?” Likewise, the mention of a prestigious or non prestigious name can bias the response, as in, “Do you agree with the American Dental Association that Colgate is effective in preventing cavities?” An unbiased-question would be to ask, “Is Colgate effective in preventing cavities?”30
Avoid Implicit Alternatives
An alternative that is not explicitly expressed in the options is an implicit alternative, Making an implied alternative explicit may increase the percentage of people selecting that alternative, as in the two following questions.
1. Do you like to fly when traveling short distances? (Incorrect)
2_ Do you like to fly when traveling short distances, (Correct)
or would you rather drive?
In the first question, the alternative of driving is only implicit, but in the second question, it is explicit. The first question is likely to yield a greater preference for flying than the second question.
Questions with implicit alternatives should be avoided unless there are specific reasons for including them When the alternatives are close in preference or large in number, the alternatives at the end of the list have a greater chance of being selected, To overcome this bias, the split ballot technique should be used to rotate the order in which the alternatives appear.
Implicit in question I arc the consequences that will arise as a result of a balanced budget, There might be a cut in defense expenditures, increase in personal income tax, cut in social programs, and so on, Question 2 is a better way to word this question. Question is failure to make its assumptions explicit would result in overestimating the respondents support for a balanced budget.
Avoid Generalizations and Estimates
Questions should be specific, not general. Moreover, questions should be worded so that the respondent does not have to make generalizations or compute estimates. Suppose we were interested in households’ annual per capital expenditure on groceries. If we asked respondents.
“What is the annual per capital expenditure on groceries in your household?” (incorrect)
they would first have to determine the annual expenditure on groceries by multiplying the monthly expenditure on groceries by 12 or the weekly expenditure by 52, Then they would have to divide the annual amount b) the number of persons in the household,0 Most respondents would be unwilling or unable to perform these calculations. A better way of obtaining the required information would be to ask the respondents two simple questions:
“What is the monthly (or weekly) expenditure on groceries in your household?”
“How many members are there in your household?” (Correct)
The researcher can then perform the necessary calculations.