Psychological discovery 4
Between-subjects design. Different subjects are assigned to each group.
Post-test-only control group. DV is measured after the manipulation of the IV.
Pre-test/post-test control DV measured before and after the manipulation.
Confound. Uncontrolled extraneous variable/flaw in an
Internal validity. Extent to which results can be attributed to the
manipulation of the IV.
Non-equivalent control Problem in subject selection/assignment may lead to
group. important differences between subjects.
History effect. Outside event that is not part of the manipulation.
Maturation effect. Naturally occurring changes within the subjects.
Testing effect. Repeated testing leads to better/worse scores.
Regression to the mean. Extreme scores, upon retesting, tend to be less
extreme, moving towards the mean.
Instrumentation effect. Changes in the DV.
Mortality/attrition. Differential dropout rates, leading to inequality
Diffusion of treatment. Observed changes due to info being received from
Experimenter effect. Consciously/unconsciously affecting the results.
Subject effect. Subject consciously/unconsciously affects results.
Floor effect. Limitation of the measuring instrument that decreases
its capability to differentiate between scores at the
bottom of the scale.
Ceiling effect. Limitation that decreases its capability to differentiate
between scores at the top of the scale.
External validity. Extent to which results can be generalised.
College sophomore problem. Using mainly college sophomores as subjects.
Exact replication. Replicating a study using the same method of
Conceptual replication. Using different methods, manipulation, or method.
Systematic replication. Study that varies from an original study in one way.
Counterbalancing. Including all orders of treatment presentation, or by
randomly determining the order for each subject.
Latin square. Counterbalancing technique to control for order
effects without using all possible orders.
Non-manipulated IV. Subjects not randomly assigned to conditions but
come to the study as members of each condition.
Cross-sectional design. Subjects of different ages observed at the same time.
Cohort. Groups of individuals born at about the same time.
Cohort effect. When the era effects how subjects respond.
Single case design. Only one participant at a time.
Small-n design. Only a few subjects studied. The experimental research strategy
Involves systematically manipulating an IV to examine its effects on a DV,
whilst controlling/limited extraneous variables so that a cause-and-effect
relationship can be established.
Goal of an experimental study is to demonstrate a cause-and-effect
relationship between two variables.
To accomplish they need two characteristics:
o Manipulate one IV to create two or more treatment conditions.
o Measure a DV so that you have a set of score for each treatment
o Compare the DV scores of each treatment condition.
o Control all extraneous variables so that you can determine that it is the
IV that is causing the changes in the DV.
Categories of variables that researchers must consider:
o Participant variables – e.g. age, gender, intelligence.
Vary from one individual to another.
o Environmental variables – e.g. lighting, time of day, weather.
Individuals in treatment A need to be tested in the same
environment as the individuals in treatment B.
E.g. examine whether a new antidepressant medication causes a decrease in
depression levels by comparing scores of placebo group with treatment group.
Quasi-experimental research methods
Attempts to achieve the same goal as the experimental strategy, but is unable
to definitively establish cause-and-effect explanations.
The quasi-IV is indirectly manipulated to examine its effects on a DV.
E.g. attempt to explain whether high doses of antidepressant causes a decrease
in depression (compare people already taking 10mg with those taking 20mg).
o Comparing groups of individuals but the groups occur naturally.
o Subjects not randomly assigned to the groups.
o E.g. not truly manipulating smoking by assigning subjects to either
smoke or not smoke – they come to the study as either smokers or non-
Single-group designs. Non-equivalent control
Posstest-only. Open to many confounds. Control group is non-
No comparison group. equivalent.
No equivalent control No pre-test measures to
group. establish equivalence of
Can compare groups on
pre-test and post-test
measures, but differences
may be due to treatment or
Pre-test/post-test. Compare scores on pre-test Can compare between to those on post-test. groups on pre-test and
No equivalent control post-test.
group for comparison. Can compare within
If change is observed, it groups from pre-test to
may be due to treatment or post-test.
confounds. Because subjects not
randomly assigned, we
cannot say that they are
If change is observed, it
may be due to treatment or
Time series. Because many measures Because many measures
are taken, can see effect of are taken, we can see the
treatment over time. effect of treatment over
No control group for time.
comparison. Non-equivalent control
If change is observed, it group available for
may be due to treatment or comparison.
confounds. Because subjects are