dependent variable is attributable to the manip-
ulation of the independent variable.
longitudinal designA research design that
compares the same group of individuals at two or
more points in time.
matchingA term used when research participants
in the experimental and control groups are
“matched”or are similar on variables (e.g., age,
sex) that may affect the outcome of the research.
mixed designsResearch designs that combine
both experimental and correlational methods. In
this design, participants from naturally occurring
groups of interest (e.g., people with panic disorder
and people with social phobia) are assigned to each
experimental treatment, allowing the experimen-
ter to determine whether the effectiveness of the
treatments varies by group classification.
multiple baseline designsDesigns used when it is
not possible or ethical to employ a treatment reversal
period. In this design, baselines are established for two
(or more) behaviors, treatment is introduced for one
behavior, and then treatment is introduced for the
second behavior as well. By observing changes in each
behavior from period to period, one may draw
conclusions about the effectiveness of the treatments.
naturalistic observationA research method in
which carefully planned observations are made in
real-life settings.
placebo effectThe case where the expectations
for the experimental manipulation cause the
outcome rather than (or in addition to) the
manipulation itself.
prevalenceThe overall rate of cases (new or old)
within a given period of time. Prevalence figures
allow us to estimate what percentage of the target
population is affected by the illness or disorder.
retrospective dataData based upon people’s
reports of past experiences and events.
risk factorA variable (e.g., demographic, envi-
ronmental) that increases a person’sriskof
experiencing a particular disease or disorder over
his or her lifetime.
scatterplotA visual representation of the relation-
ship between two variables. The scatterplot consists of
anx-axis (labeled to reflect one variable), ay-axis
(labeled to reflect the other variable), and a number of
data points, each corresponding to one person’s scores
on both variables.
single-case designsDesigns that focus on the
responses of only one participant. Usually, an
intervention is introduced after a reliable baseline
is established, and the effects of the intervention
are determined by comparing the baseline and
postintervention levels of behavior.
statistical significanceStatistical values that would
not be expected to occur solely on the basis of chance.
By convention, a value is considered statistically
significant if it would be expected to occur by chance
alone fewer than 5 times out of 100.
third-variable problemThe possibility that a
correlation between variables A and B is due to the
influence of an unknown third variable rather than
to a causal relationship between A and B.
within-group designsDesigns in which the
same group of participants is compared at different
points in time (say, before and after a treatment is
administered).WEB SITES OF INTEREST
To visit any of the following Web sites, go to
http://www.wadsworth.com and click Links.
4-1 APA Ethical Principles and Code of Conduct
(with 2010 amendments)
http://www.apa.org/ethics/code/index.aspx
4-2 Research with Non-human Animals in
Psychology
http://www.apa.org/science/leadership/care/
index.aspx126 CHAPTER 4