WRITING THE RESEARCH REPORT AND THE POLITICS OF SOCIAL RESEARCHfrom sponsors, and disseminate results in an open,
clear manner. The politics of social research over-
laps with many issues in sponsored research. In ad-
dition, many social researchers face economically
or politically powerful groups who attempt to limit
what they study, how they conduct research, or how
they disseminate the findings.Limits on What Researchers Study
Direct Limits on Research.Governments or pow-
erful groups in society may try to restrict free sci-
entific inquiry. Some limits on research have always
existed but in particular times or places they become
very restrictive. In nondemocratic societies, control
over or censorship of social research is the rule, not
the exception. This is particularly the case with
politically sensitive topics including public opinion
surveys. Thus, during the late twentieth century in
China, eastern Europe, South Africa, and Taiwan,
for example, social researchers were suspect, lim-
ited to “safe” topics, or forced to support official
government policy.^18 In a number of countries, the
study of sociology itself was banned as subversive
after a military coup. In an extreme case, 40 percent
of German scientists were dismissed from their jobs
for political reasons when the Nazis “purified” uni-
versities and research centers in 1937.^19 Hundreds
of professors and researchers in the United States
who did not publicly swear to anticommunism and
cooperate with the McCarthy investigations of the
1950s were purged. At that time, people who ob-
jected to mandatory loyalty oaths, supported racial
integration, or advocated the teaching of sex edu-
cation were suspected of subversion and threatened
with dismissal. For instance, at the University of
California alone, twenty-five professors were fired
for refusing to sign a loyalty oath.^20
Two limitations on social research are (1) gate-
keepers who control access to data or subjects and
(2) controls over how official statistics are collected.
Gatekeepers can limit what we study and may try to
protect themselves or their organizations from crit-
icism or embarrassment. They often limit access to
subjects or areas with which they have concerns.
For example, in 1997, the U.S. Army dropped
several questions from a 153-item questionnaire on
sexual harassment to be sent to 9,000 soldiers. The
reason for eliminating the six questions was that
“senior Army officials feared that the responses
could be highly embarrassing to the Army”
(Schmitt, 1997). A social anthropologist and a law
professor who were consultants on the project were
upset and noted that preliminary results from an
early version of the questionnaire suggested that
sexual harassment at military bases was correlated
with questions that asked about certain soldier be-
haviors (e.g., going to strip clubs, watching X-rated
movies). Gatekeeper army officials did not want
such questions because the answers could prove em-
barrassing to widespread practices on military
bases.
Another limitation involves official or existing
statistics that government or other large organiza-
tions collect. Whether agencies decide to collect in-
formation and how they collect it can affect research
findings. Political factors often determine how phe-
nomena (e.g., unemployment, income, educational
success, poverty level) are defined in official statis-
tics and whether such data are collected.^21
Hundreds of social scientists regularly rely on
the data collected by the U.S. Census Bureau for
conducting demographic, economic, and other stud-
ies. The original purpose of a census was to allocate
elected representatives among states and districts.
Later the Census Bureau began to gather informa-
tion for making policy decisions, providing social
programs, and distributing government funds based
on the population in an area. The Census Bureau
has become a major source of social science infor-
mation and a clearinghouse for official statistics on
many topics. Serious distortions (e.g., systematic
overcounts or undercounts of some people or areas)
in Census Bureau statistics weaken research find-
ings based on them, prevent full democratic repre-
sentation, and undermine a fair distribution of social
programs or funds.
Researchers who rely on existing statistics de-
pend on the government to supply information or
documents. In the United States, the Paperwork Re-
duction Act of 1980 created the Office of Informa-
tion and Regulatory Affairs to determine whether