that the teen is not currently engaging in sexual activity requiring a condom.
Researchers control for this type of bias by clearly specifying criteria required
for subjects to be included in a study. After subjects are enrolled in a study, it is
important for researchers to control how the study is actually conducted. This
includes controlling the environment in which the study is conducted, providing
consistency in how interventions are delivered, and collecting data in a careful
and consistent manner. For example, asking students to report their condom
use during an interview where others might hear their responses could bias
the results because students may not feel comfortable responding truthfully.
A better way to obtain this information may be to have students complete a
self-report of their condom use on an anonymous questionnaire.
Randomization
Randomization is an effective way to control extraneous variables. Randomiza-
tion can occur either with random sampling of the subjects to be studied or by
random assignment of subjects to the intervention or control group. Random
sampling means that all people in the population of interest have the same prob-
ability of being selected to be included in the study. This occurs infrequently
in health-related research because researchers usually do not have the time or
money to randomly sample all people who could potentially participate. For
example, it would be very difficult to randomly sample sexually active teens
from around the country. Instead, in this example the researcher would most
likely use a convenience sample of students from high schools within the area.
The researcher would then randomly assign the students into the intervention
and the control groups. Random assignment means that all subjects in the
sample (not the population) have an equal chance of being assigned to either
the treatment or the control group. Random assignment increases the likelihood
that extraneous variables that may affect the DV will be equally distributed
between the two groups. Using random assignment allows researchers to be
more confident that the IV rather than the extraneous variables caused the
effect on the DV. Although researchers may be very careful about selecting
representative samples and using random assignment to groups to control
for extraneous variables, these techniques do not guarantee that extraneous
variables will be equally distributed between the two groups. When groups are
different, researchers need to apply certain statistical tests to control for the
effect of extraneous variables on the DV.
Between-Groups and Within-Groups
Quantitative studies often are designed for the purpose of making comparisons
either by comparing different groups of subjects or by comparing the same
subject at different points in time. Studies developed to compare two different
KEY TERMS
randomization:
The selection,
assignment, or
arrangement of
elements by chance
random sampling:
Technique for
selecting elements
whereby each has
the same chance of
being selected
random
assignment:
Assignment
technique in which
subjects have an
equal chance of
being in either the
treatment or the
control group6.1 Chart the Course: Selecting the Best Design 153