Page 53 - Essencials of Sociology

26 CHAPTER 1 The Sociological Perspective

Down-to-Earth Sociology

Loading the Dice: How Not to Do Research

he methods of science lend themselves to distortion, would be the most popular in the coming year, their list of
misrepresentation, and downright fraud. Consider these choices included no other jeans but Levi’s 501!
Tfindings from surveys: 4. Discard undesirable results. Researchers can keep
silent about results they don’t like, or they can continue
Americans overwhelmingly prefer Toyotas to Chryslers. to survey samples until they find one that matches what
Americans overwhelmingly prefer Chryslers to Toyotas.
they are looking for.
Obviously, these opposite conclusions cannot both 5. Misunderstand the subjects’ world. This route can
be true. In fact, both sets of findings are misrep- lead to errors every bit as great as those just cited. Even
resentations, even though the responses came researchers who use an adequate sample and word their
from surveys conducted by so-called indepen- questions properly can end up with skewed results. They
dent researchers. It turns out that some con- might, for example, fail to anticipate that people may be
sumer researchers load the dice. Hired by firms embarrassed to express an opinion that isn’t “politi-
that have a vested interest in the outcome of the cally correct.” For example, surveys show that 80
research, they deliver the results their clients are look- percent of Americans are environmentalists. Is this
ing for (Armstrong 2007). Here are six ways to load the dice. an accurate figure? Most Americans are probably
embarrassed to tell a stranger otherwise. Today,
1. Choose a biased sample. If you want to “prove” that would be like going against the flag, mother-
that Americans prefer Chryslers over Toyotas, interview hood, and apple pie.
unemployed union workers who trace their job loss to 6. Analyze the data incorrectly. Even when researchers
Japanese imports. The answer is predictable. You’ll get strive for objectivity, the sample is good, the wording
what you’re looking for. is neutral, and the respondents answer the questions
2. Ask biased questions. Even if you choose an unbiased honestly, the results can still be skewed. The researchers
sample, you can phrase questions in such a way that you may make a mistake in their calculations, such as enter-
direct people to the answer you’re looking for. Suppose ing incorrect data into computer programs. This, too, of
that you ask this question:
course, is inexcusable in science.
We are losing millions of jobs to workers overseas who Of these six sources of bias, the first four demonstrate
work for just a few dollars a day. After losing their jobs, fraud. The final two reflect sloppiness, which is also not ac-
some Americans are even homeless and hungry. Do you ceptable in science.
prefer a car that gives jobs to Americans, or one that As has been stressed in this chapter, research must be ob-
forces our workers to lose their homes?
jective if it is to be scientific. The underlying problem with the
This question is obviously designed to channel peo- research cited here—and with so many surveys bandied about
ple’s thinking toward a predetermined answer—quite in the media as fact—is that survey research has become big
contrary to the standards of scientific research. Look business. Simply put, the money offered by corporations has
again at the Doonesbury cartoon on page 25. corrupted some researchers.
3. List biased choices. Another way to load the dice is to The beginning of the corruption is subtle. Paul Light, dean
use closed-ended questions that push people into the at the University of Minnesota, put it this way: “A funder will
answers you want. Consider this finding: never come to an academic and say, ‘I want you to produce
finding X, and here’s a million dollars to do it.’ Rather, the
U.S. college students overwhelmingly prefer Levi’s 501 to subtext is that if the researchers produce the right finding,
the jeans of any competitor. more work—and funding—will come their way.”

Sound good? Before you rush out to buy Levis, note Sources: Based on Crossen 1991; Goleman 1993; Barnes 1995; Resnik
what these researchers did: In asking students which jeans 2000; Augoustinos et al. 2009.

Watch on MySocLab
Video: Everything’s Cool, Clip 1

Improperly worded questions can
steer respondents toward answers
that are not their own, which
produces invalid results. Doonesbury © G. B. Trudeau. Reprinted with permission of Universal Press Syndicate. All rights reserved.