QUALITATIVE AND QUANTITATIVE SAMPLINGstrategy and vice versa. Nonetheless, most quantita-
tive studies use probability or probability-like samples
while most qualitative studies use a nonprobability
method and nonrepresentative strategy.
SAMPLING STRATEGIES
We want to avoid two types of possible sampling
mistakes. The first is to conduct sampling in a sloppy
or improper manner; the second is to choose a type
of sample inappropriate for a study’s purpose. The
first mistake reminds us to be very meticulous and
systematic when we sample. To avoid the second
mistake, we need a sampling strategy that matches
our specific study’s purpose and data. Sampling
strategies fall into two broad types: a sample that will
accurately represent the population of cases, and all
others. We primarily use the first strategy in quanti-
tative studies and the latter in qualitative studies.
Strategies When the Goal Is to Create
a Representative Sample
In a representative sample, our goal is to create
sample data that mirror or represent many other
cases that we cannot directly examine. We can do
this in two ways. The first is the preferred method
and considered the “gold standard” for representa-
tive samples, the probability sample. It builds on
more than a century of careful reasoning and applied
mathematics plus thousands of studies in natural sci-
ence and quantitative social science. With a proba-
bility sampling strategy, we try to create an accurate
representative sample that has mathematically pre-
dictable errors (i.e., precisely known chances of
being “off target”). This sampling approach is com-
plex with several subtypes. Before we examine it,
let us look at the second, simpler way to produce
a representative sample: to use a nonprobability
technique. It is a less accurate substitute when we
want a representative sample; however, it is accept-
able when probability sampling is impossible, too
costly, time consuming, or impractical.Nonprobability Sampling Techniques.Ideally, we
would prefer probability samples when we want to
create a representative sample, as a less demand-
ing alternative there are two nonprobability alter-
natives: convenience and quota samples. In
convenience sampling(also called accidental,
availability,orhaphazard sampling), our primary
criteria for selecting cases are that they are easy to
reach, convenient, or readily available. This sample
type may be legitimate for a few exploratory pre-
liminary studies and some qualitative research
studies when our purpose is something other than
creating a representative sample. Unfortunately, it
often produces very nonrepresentative samples, so
it is not recommendedfor creating an accurate
sample to represent the population.
When we select cases based on convenience,
our sample can seriously misrepresent features in the
entire population.^1 You may ask why, if this method
is so bad and samples can be seriously nonrepre-
sentative, anyone would use it. The reason is simple:
convenience samples are easy, cheap, and quick to
obtain. Another reason might be that people are
ignorant about how to create a good representative
sample. An example of such sampling is the
person-on-the-street interview conducted by televi-
sion programs. Television interviewers go out on
the street with camera and microphone to talk to a
few people who are convenient to interview. The
people walking past a television studio in the middle
of the day do not represent everyone. Likewise, tel-
evision interviewers tend to pick people who look
“normal” to them and avoid people who are unat-
tractive, disabled, impoverished, elderly, or inartic-
ulate. Another example is a newspaper that asks
readers to clip a questionnaire and mail it in, a Web
site that asks users to click on a choice, or a televi-
sion program that asks viewers to call in their
choices. Such samples may have entertainment
value, but they easily yield highly misleading dataConvenience sampling A nonrandom sample in
which the researcher selects anyone he or she happens
to come across.