SURVEY RESEARCHAlmost all people complete surveys at some
time, and the reporting of survey or poll results in
major newspapers grew rapidly after the 1960s. By
the 1970s, it seemed that every day a newspaper story
cited survey or poll results. As surveys became
increasingly used, response rates have declined. Non-
response rates in surveys vary greatly; for academic
organizations, they range from 25 to 33 percent. In
the United States, nonresponse rates for major aca-
demic surveys rose from less than 10 percent in the
1950s to 25 percent in the 1980s. Public cooperation
in survey research has declined across most countries
with the Netherlands having the highest refusal rate;
it is as high as 30 percent in the United States.^44 The
nonresponse rates to commercial polls (Roper, Gal-
lup, CBS, etc.) and campaign polls tend to be higher,
however, reaching as high as 50 percent.
Researchers discovered a growing group of
“hard core” refusing people who decline all sur-
veys. In addition, general survey participation has
declined because people believe there are too many
surveys. Other reasons for declining survey partic-
ipation include a fear of strangers, a more hectic
lifestyle, a loss of privacy, and a rising distrust of
authority. The misuse of a survey to sell products
or persuade people, poorly designed question-
naires, and inadequate explanations of surveys also
increase refusals for legitimate, serious ones.
The most interested, informed, and active
members of society tend to participate in surveys.
This means that nonresponse both harms survey
validity and omits a particular segment of the pop-
ulation. In the United States, nonrespondents tend
to be young non-White males and the less educated.
Nonresponse rates have five components (see
Expansion Box 9, Confusion about Response
Rates).^45
- Location—Could a sampled respondent be
located? - Contact—Was a located respondent at home or
reached after many attempts? - Eligibility—Was the contacted respondent the
proper age, race, gender, citizenship, and so on
for the survey purpose? - Cooperation—Was an eligible respondent will-
ing to be interviewed or fill in a questionnaire?
5. Completion—Did a cooperating respondent
stop answering before the end or start answer-
ing most questions with “do not know” or “no
opinion”?
Improving the overall survey response rate
requires us to reduce each type of nonresponse.EXPANSION BOX 9
Confusion about Response RatesThere is some confusion about response rates
because the total response rate depends on the suc-
cess rate of five component responses, each of which
has its own rate:
Location rate: Percentage of respondents in the
sampling frame who are located.
Contact rate: Percentage of located respondents
who are contacted.
Eligibility rate: Percentage of contacted respondents
who are eligible.
Cooperation rate: Percentage of contacted, eligible
respondents who agree to participate.
Completion rate: Percentage of cooperating respon-
dents who complete the survey.
Total response rate: Percentage of all respondents
in the initial sampling frame who were located, con-
tacted, eligible, agreed to participate, and completed
the entire questionnaire.For example, researchers begin with 1,000 respon-
dents in a sampling frame, locate 950 by telephone
or an address, are able to contact 800 (by an inter-
viewer or successful mailing), and determine that 780
are eligible (i.e., meet basic criteria, speak the lan-
guage, are mentally competent). They find that 700
people cooperate with the questionnaire or interview,
and 690 complete the entire questionnaire or inter-
view. This yields the following rates: location rate:
95 percent; contact rate:84.2 percent; eligibility
rate:97.5 percent; cooperation rate:89.8 percent;
completion rate:98.6 percent; total response rate:
69 percent. The total response rateis the product
of all of the individual rates: .95 .842 .975
.898.986 .690.