SURVEY RESEARCHCollege respondents are more responsive to Web
surveys, but other respondents (e.g., medical doc-
tors, teachers, consumers) prefer mail surveys. Fol-
low-up reminders appear to be less effective for Web
than for mail surveys (Shih and Fan, 2008). Many
of the techniques suggested follow the total design
method and help to make the task easy and inter-
esting for respondents.TYPES OF SURVEYS: ADVANTAGES
AND DISADVANTAGES
Mail and Self-Administered
Questionnaires
Advantages.We can give or mail questionnaires
directly to respondents, who read the instructions
and questions and then record their answers. A sin-
gle researcher can conduct this type of survey at
very low cost and cover a wide geographical area.
The respondent can complete the questionnaire
when it is convenient and can check personal
records for information if necessary. Mail ques-
tionnaires offer anonymity and avoid interviewer
bias. They are very effective and can achieve
acceptable response rates from an educated sam-
ple that has a strong interest in the topic or the sur-
vey organization.
Disadvantages.Because many people do not com-
plete and return mail questionnaires, their biggest
problem is a low response rate. Most questionnaires
are returned within 2 weeks, but others trickle in for
up to 2 months. We can improve response rates by
sending nonrespondents reminder letters, but this
adds to the time and cost of data collection.
We lack control over the conditions under
which a mail questionnaire is completed. A ques-
tionnaire completed during a drinking party by a
dozen laughing people may be returned along with
one filled out by an earnest respondent. Also, no
one is present to clarify questions or to probe for
more information when respondents give incom-
plete answers. Someone other than the sampled
respondent (e.g., spouse, new resident) may open
the mail and complete the questionnaire withoutthe researcher’s knowledge. We cannot visually
observe the respondent’s reactions to questions,
physical characteristics, or the setting. For example,
an impoverished 70-year-old White woman living
alone on a farm could falsely state that she is a pros-
perous 40-year-old Asian male doctor with three
children living in a town. Such extreme lies are rare,
but serious errors can go undetected. In addition,
different respondents can complete the question-
naire weeks apart or answer questions in a differ-
ent order than intended. Incomplete questionnaires
can also be a serious problem.
The mail questionnaire format limits the ques-
tions that we can use. Those that require visual aids
(e.g., look at this picture and tell me what you see),
open-ended questions, many contingency ques-
tions, and complex questions cannot be used in most
mail questionnaires. Likewise, mail questionnaires
are ill suited for people who are illiterate or nearly
illiterate (see Table 4).Telephone Interviews
Advantages.The telephone interview is a popular
survey method because about 95 percent of the
population can be reached by telephone. An inter-
viewer calls a respondent (usually at home), asks
questions, and records answers. Researchers sam-
ple respondents from lists and telephone directo-
ries or use RDD and can quickly reach many people
across all geographic areas. A staff of interviewers
can interview 1,500 respondents across a nation
within a few days and, with a dozen callbacks,
achieve response rates as high as 80 percent. The
telephone survey is more expensive than a mail
questionnaire because it requires interviewer time.
In general, the telephone interview is a flexible
method with most of the strengths of face-to-face
interviews but at a much lower cost. Interviewers
control the sequence of questions and can use some
probes. A specific respondent is chosen and is
likely to answer all questions alone. We know when
the questions were answered and can use contin-
gency questions effectively.
Most researchers use computer-assisted tech-
nologies in telephone interviews, two of which