disconfirm theories or explanations believed to apply
to everyone, and facilitate the generation of hypoth-
eses. However, case study methods typically do not
result in universal laws or principles applicable to all
and cannot lead to cause–effect conclusions.
Epidemiological methods are used to estimate
the prevalence and incidence of a disorder or con-
dition in the population. These methods also enable
us to identify risk factors.
Correlational methods assess and quantify the
relations between variables; this may stimulate
ideas or theories about causal relations among vari-
ables. However, correlation is a necessary but not
sufficient indicator of causation, which can only be
established through experimental methods.
Cross-sectional designs involve an examination
of participants at one point in time, whereas longitu-
dinal designs evaluate the same participants over a
period of time. Longitudinal studies allow scientists
to evaluate time-order relationships among factors
that vary together and help address third-variable
explanations of observed relations. Unfortunately,
longitudinal studies are time-intensive and expensive
and are conducted less frequently than cross-
sectional studies.
The experimental method is powerful because it
enables scientists to evaluate cause–effect questions
and to exert control over a number of potentially
important factors that affect the behavior in question.
Internal validity refers to the degree to which, based
on the design of the experiment, we can be confi-
dent that the manipulation of the independent
variable affected the dependent variable. External
validity refers to the extent to which the results of
the experiment are generalizable to other, preferably
“real-world,” conditions. In some instances, for
practical or ethical reasons, it is not possible to con-
duct experiments on real-life problems. In these
cases, analog studies may be used.
An important variant of the experimental and
case study methods is the single-case design (e.g.,
ABAB design, multiple baseline design). The
mixed design combines features of the correlational
and experimental methods and helps us evaluate
which interventions are best for whom. Finally, a
number of ethical considerations are involved in
conducting research, including obtaining informed
consent, assuring confidentiality, using deception
appropriately, providing debriefing information,
and protecting against fraud.KEY TERMS
ABAB designsSingle-case designs that observe
systematic changes in the participant’sbehavioras
the treatment and no-treatment conditions alter-
nate. The initial baseline period is followed by a
treatment period, a treatment reversal period, and
a second treatment period.
analog studyA study conducted in the laboratory
under conditions that are purportedly analogous to
real life.
between-group designsDesigns in which two
or more separate groups of participants each
receive a different kind of treatment.
case study methodA research method consisting
of the intensive description or study of one person
(usually a client or patient who is in treatment).
confidentialityIn research on human subjects,
the principle of protecting individual participants’
data from public scrutiny.
confoundA situation in which extraneous
variables are not controlled or cannot be shown to
exist equally in one’s experimental and control
groups. When there is a confound, one cannot
attribute changes in the dependent variable to the
manipulation of the independent variable.
control groupThe group in an experimental
design that does not receive the treatment of
interest. In the perfect experimental design, the
experimental and control groups are similar on all
variables except the treatment variable.124 CHAPTER 4