Many qucs.ions. particularly those measuring attitudes and lifestyles, arc worded as statements to
\\ hich respondents indicate their degree of agreement or disagreement. Evidence indicates that the
response obtained is influenced by the directionality of the statements: whether they are stated
positively or negatively. In these cases. it is better to use dual statements, some of which are positive
and the others negative. Two different questionnaires could be prepared. One questionnaire
would contain half negative and half positive statements in an interspersed way. The direction of
these statements would be reversed in the other questionnaire. An example of dual statements was
prov ided in the summated Likert scale in Chapter 9 designed to measure attitudes toward Sears;
some statements about Sears were positive whereas others were negative
In the department store project, the questionnaire was to be answered by the male or female head of the
household who did most of the shopping in department stores. The first question asked was, “Who in your
household does most of the shopping in department stores?” Thus the opening question helped in identifying
the eligible respond-nts. It also gained cooperation because of its simple and nonthreatening nature
Type of Information
The type of information obtained in a questionnaire may be classified as (I) basic information, –
(2) classification information, and (3) identification-information. Basic information relates directly
to the research problem. Classification information, consisting of socioeconomic and demographic
characteristics, is used to classify the respondents and understand the results.
Identification information includes name, postal address, e-rnail address, and telephone number.
Identification information may be obtained for a variety of purposes, including verifying that the
respondents listed were actually interviewed, remitting promised incentives, and so on. As a
general guideline, basic information should be obtained first, followed by classification and, finally,
identification information. The basic information is of greatest importance to the research project
and should be obtained first, before we risk alienating the respondents by asking a series of personal
questions. The questionnaire given in problem 7 (see exercises for this chapter) incorrectly obtains
identification (name) and some classification (demographic) information in the beginning
effect on Subsequent Questions
Questions asked early in a sequence can influence the responses to subsequent questions. As a
rule of thumb. general questions should precede specific questions. This prevents specific questions
from biasing responses to general questions. Consider the following sequence of questions
Questions should be asked in a logical order. All of the questions that deal with a particular topic
should be asked before beginning a new topic. When switching topics. brief transitional phrases
should be used to help respondents switch their train of thought.Branching questions should be designed carefully.35 Branching questions direct respondents
to different places in the questionnaire based on how they respond to the question at hand. These
questions ensure that all possible contingencies are covered. They also help reduce interviewer and
respondent error and encourage complete responses. Skip patterns based on the branching
questions can become quite complex. A simple way to account for all contingencies is to prepare a
flowchart of the logical possibilities and then develop branching questions and instructions based
on it. A flowchart used to assess the use of credit in store purchases is shown in Figure 10.2.
Form and Layout
The fonnat, spacing, and positioning of questions can have a significant effect on the results. as illustrated by the Census 2000 questionnaire in the opening example. This is particularly important for self-administered questionnaires. Experiments on mail questionnaires for census of population revealed that questions at the top of the page received more attention than those placed at the
bottom. Instructions printed in red made little difference except that they made the questionnaire appear more complicated to the respondents
Form and Layout
It is a good practice to divide a questionnaire into several parts. Several parts may be needed for questions pertaining to the basic information. The questions in each part should be numbered, particularly when branching questions are used. Numbering of questions also makes the coding of responses easier. The questionnaires should preferably be preceded. In precoding. the codes to enter in the computer are printed on the questionnaire. T) pically, the code identifies the line number and the column numbers in which a particular response will be entered. Note that when CAT! or CAP! is used, the precoding is built into the software. Coding of questionnaires is explained in more detail in Chapter 14 on data preparation. Here we give an example of a precoded questionnaire. To conserve space, only part of the questionnaire is reproduced
Reproduction of the Questionnaire
How a questionnaire is reproduced for administration can influence the results. For example, if the questionnaire is reproduced on poor-quality paper or is otherwise shabby in appearance. the respondents will think the project is unimportant and the quality of response will be adversely affected. Therefore, the questionnaire should be reproduced on good-quality paper and have a professional appearance.
Vertical response columns should be used for individual questions. It is easier for interviewers and respondents to read down a single column rather than sideways across several columns. Sideways formatting and splitting, done frequently to conserve space, should be avoided. This problem can be observed in The American Lawyer questionnaire.
Directions or instructions for individual questions should be placed as close to the questions as possible. Instructions relating to how the question should be administered or answered by the respondent should be placed just before the question. Instructions concerning how the answer should be recorded or how the probing should be done should be placed after the question (for more information on probing and other interviewing procedures, see Chapter 13). It is a common practice’to distinguish instructions from questions by using distinctive type, such as capital letters. (See the department store project in the section titled “Effect of Interviewing Method on Questionnaire Design.”)
The questionnaire should be reproduced in such away that it is easy to read and answer. The type should be large a~ elear, Reading the questionnaire should not impose a strain. Several technologies allow researchers to obtain better print quality and simultaneously reduce costs. One effort along these lines resulted in a lowering of printing costs from $1,150 to $214.