that a sample of 200 students is necessary to achieve representativeness. After
selecting the start point, the researcher selects additional elements by proceed-
ing vertically through the columns until 200 subjects are selected. Computer
programs are also available that select a random sample, and these programs
are frequently used when the researcher is using a large sample.
Stratified Random Sampling
Stratified random sampling involves selecting elements from a population
that has been divided into groups or strata. Researchers identify characteristics
that they want to stratify. These determinations are frequently based on what
is already known about the phenomenon being studied. For example, findings
show that boys and girls respond differently to pain. Therefore, a researcher
may want to stratify subjects on the variable of gender to ensure that they have
enough boys and girls to make comparisons.
Strata must be mutually exclusive. This means that each element can be put
into one and only one stratum (Gray, Grove, & Sutherland, 2016). For example,
if the sample is stratified according to gender, subjects can be categorized only
as a boy or a girl. These strata are mutually exclusive because subjects cannot
fit into both categories. After strata are established, researchers assign each
element from the accessible population to a stratum. Participants are then
randomly selected from each stratum. For example, the researcher may want
to study nursing student attitudes toward treatment of pain. Because there are
significantly fewer male than female nursing students, the researcher needs
to account for gender disparity in the study. If less than 10% of the accessible
population is male, the researcher uses stratified random sampling so that
10% of the sample is male. After strata are determined, subjects are randomly
selected from each stratum.
One advantage to stratified random sampling is that sampling error can be
reduced because elements are selected by strata that are known to represent the
population. Representative samples increase the likelihood that findings can
be generalized to the population. Stratifying can also decrease data collection
time and costs of collecting data (Gray et al., 2016). However, care must be
taken when using this method. Each stratum must have a sufficient number
of elements. In the preceding example, if 50 males are needed to constitute
10% of the sample, there must be at least 50 males in the accessible population.
Cluster Sampling
Cluster sampling is also known as multistaging sampling. Cluster sampling
is an effective and efficient method to collect data from large populations.
A manageable sample is obtained by randomly selecting elements from larger
KEY TERMS
stratified random
sampling: Selecting
elements from
an accessible
population that has
been divided into
groups or strata
cluster sampling:
Random sampling
method of
selecting elements
from larger to
smaller subsets
of an accessible
population;
multistaging
sampling11.2 Sampling Methods 293