adjusted than non–gang members, and saw gang membership as a reasonable eco-
nomic alternative to unemployment and poverty. Gangs provided good steady jobs,
high wages (with high risks), and the rich social relationships that come from
community. Similarly, Elijah Anderson’s research on young black men in the inner
city (1992, 2000) gave a far deeper understanding of the complex of meanings
and motives for behavior that had often been reduced to rather one-dimensional
stereotypes.
Ethnography taxes our powers of observation and stretches our sociological mus-
cles to try to see the world from the point of view of other people. Philippe Bourgois
(2002) lived for three years in New York City’s Spanish Harlem, studying the culture
of crack dealers. Loic Wacquant (2003) trained for over three years right alongside
local boxers in a training gym in Chicago’s South Side. Nancy Sheper-Hughes (1992)
studied the poor in Brazil, revealing the physical and psychological violence that
permeates their everyday lives and structures social interaction. Javier Auyero (2000)
studied clients’ own views of the patronage systems that sustain survival in shanty-
towns on the outskirts of Buenos Aires, Argentina. Chen Hsiang-Shul (1992) stud-
ied the transnational worlds of Taiwan immigrants in New York. Ethnographic
methods enable us to see people’s worlds up close, in intimate detail, bringing out
both subtle patterns and structural forces that shape social realities.
Interview Studies.The most typical type of qualitative study uses interviews with a
small sample. These studies use a purposive sample, which means that respondents
TYPES OF SOCIOLOGICAL RESEARCH METHODS 117The Likert scale
is the most
widely used
scale in survey research. Developed by
Rensis Likert (1932), it is a technique
that presents a set of statements on a
questionnaire, then asks respondents to
express levels of agreement or disagree-
ment with these statements. Their
responses are given numerical value,
usually along a five-point or a seven-
point scale. By tallying these numeric
values, sociologists can gauge people’s
attitudes.
Likert scales can be used to gauge
many types of attitudes, from agreement
or disagreement to relative importance,
likelihood, quality, or frequency. Some
Likert scales provide a middle value that
is neutral or undecided; others use a
“forced-choice” scale, with no
neutral value, that requires
respondents to decide whether they
lean more toward agreement or
disagreement.
For example, let’s say you are doing a
survey examining employee self-esteem.
You want to gauge levels of self-satisfac-
tion in the workplace. You might present
people with a series of statements such
as, “I feel good about my work in school
on the job,” and “I can tell my co-work-
ers respect me,” among others. Then you
would ask respondents to record the
extent of their agreement or disagree-
ment with these statements along a Lik-
ert scale. The scale could look something
like this:Measuring Attitudes with
a Likert ScaleHow do we know
what we know
Or, they could record their answers
on a “forced-choice” scale that looks
more like this:disagree
stronglydisagree
somewhat neutralagree
somewhatagree
strongly12345disagree
stronglydisagree
somewhatagree
somewhatagree
strongly123 4You would take the different scaling
structure into account when analyzing
and reporting your results. But in either
case, the Likert scale would help you to
see the extent or intensity of atti-
tudes—more or less, stronger or weaker,
bigger or smaller—registered by your
survey subjects.