The Science of Psychology 25could randomly select a certain number of college freshmen from several different col-
leges across the United States, for example. Why randomly? Because the sample has to
be representative of the population, which is the entire group in which the researcher is
interested. If researchers selected only freshmen from Ivy League schools, for example,
they would certainly get different opinions on politics than they might get from small
community colleges. But if they take a lot of colleges and select their participants (people
who are part of the study) randomly, they will be more certain of getting answers that a
broad selection of college students would typically give.
Getting a representative sample is not always easy (Banerjee & Chaudhury, 2010).
Many researchers (even more so in the past than now) use people for their samples who
are readily available. Since many researchers work in educational settings, that means
that they often use college students. College students aren’t really good representatives
of the general population even if you sampled many different kinds of schools as in the
previous example—they are mostly white, well educated (and, in the early days of psy-
chology, nearly all men). The general population is not all of those things, obviously.
Sometimes the best way to get a truly representative sample is to pick names from a large
phone book, starting by opening the book to a random page, closing one’s eyes, and jab-
bing a finger at a name. Starting from that name and taking every tenth name from that
point would be fairly random.
Another major disadvantage of the survey technique occurs because people aren’t
always going to give researchers accurate answers. The fact is, people tend to misremem-
ber things, distort the truth, and may lie outright—even if the survey is an anonymous
questionnaire. Remembering is not a very accurate process sometimes, especially when
people think that they might not come off sounding very desirable or socially appropri-
ate. Some people deliberately give the answer they think is more socially correct rather
than their true opinion so that no one gets offended in a process called courtesy bias.
Researchers must take their survey results with a big grain of salt—they may not be as
accurate as they would like them to be.
Simulate the Experiment, Participating in a Research SurveySurvey PARTICIPATING IN A RESEARCH SURVEY
Previous NextINTRODUCTION SURVEY RESULTSInteractiveThis survey asks you about your attitudes
towards and experiences with a broad range
of psychological principles and theories.
Click Next to begin the surveyrandomly: in this sense, selected so that each member of the group has an equal chance of
being chosen.
anonymous: not named or identified.
grain of salt: a phrase meaning to be skeptical; to doubt the truth or accuracy of something.
population
the entire group of people or animals
in which the researcher is interested.© The New Yorker Collection 1989 George Price
from cartoonbank.com. All Rights Reserved.