Chapter 5 Research Methods in the Study of Abnormal Behavior Ch 5.

Chapter 5Research Methods in the Study of

Abnormal Behavior

Ch 5

The Scientific Method• Science is the pursuit of systematized

knowledge through observation– Science is both a method and a goal

• Scientific propositions must be – Stated clearly and precisely– Testable (allows for disconfirmation)

• Tests of hypotheses must be reproducible by other scientists

Ch 5.1

The Role of Theory in Science• A theory is a set of propositions meant to

explain a class of phenomena– Propositions are causes, the phenomena are the

effects induced by the causes

• Theories often lead to the generation of hypotheses that confirm or disconfirm the theory

• Theories are constructed by scientists

Ch 5.2

Nature of Scientific Explanation

• Critical what kind of question is asked– What makes the person behave like this (i.e., cause

distress, impair functioning)?– Why does the person behave in this unusual way?– Under what conditions does this behavior occur?– How do we help the person behave in a more adaptive

way?– What conditions are necessary, sufficient, and

contributory in the etiology, maintenance, and treatment of abnormal behavior? (Barlow & Durand)

Research Methods of Abnormal Psychology

• A case study includes historical and biographical information on a single person

• Case studies are used to– provide a detailed description of a rare

phenomenon– disconfirm allegedly universal aspects of a

particular theoretical proposition– generate new research hypotheses

Ch 5.3

Epidemiological Research• Epidemiology is the study of the frequency

and distribution of a disorder in a population– Prevalence is the proportion of a population that

has a disorder at any given time– Incidence refers to the number of new cases

that occur during some time period– Risk factors are conditions that increase the

likelihood of developing the disorder

Ch 5.4

Lifetime Prevalence Rates of Selected Diagnoses (%)

Major depressive episode 12.7 21.3 17.1

Manic episode 1.6 1.7 1.6

Simple phobia 6.7 15.7 11.3

Antisocial personality disorder 5.8 1.2 3.5

Diagnosis Male Female Total

Ch 5.5

The Correlational Method• The correlational method examines the relationship

between or among two or more variables• The variables are assessed as they exist in nature

(no experimental manipulation)• Correlational studies seek to determine the

magnitude and direction of a relationship among variables– E.g. smoking & mortality; sexual performance & number

of drinks;

Ch 5.6

The Correlation Coefficient

• The relationship between values of two variables can be quantified by the correlation coefficient “r”– r can take on values between +1.0 and -1.0

– sign of r indicates direction of relationship– size of r indicates strength of relationship

-1.0 0 +1.0

Ch 5.7

Scatter Diagrams

Ch 5.8

Statistical Significance • A correlation of zero implies no relationship between 2

variables– Chance factors can result in a correlation different from zero

• Statistical significance tests examine the likelihood that a correlation is different from zero simply because of chance variation– The probability of a chance finding is set at p < 0.05 (5 in 100

or less)

• Statistical vs. clinical significance

Ch 5.9

Problems of Causality in Correlational Research

• The directionality problem: one cannot tell the direction of causation from a correlation coefficient– Correlation does not imply causation

(involvement of 3rd variable)– Causation does imply correlation (correlational

research can be used to disconfirm a predicted causal relation)

Ch 5.10

The Experiment• The experiment allows for determination of a

causal relation between two variables• An experiment involves

– random assignment of subject to experimental conditions

– manipulation of an independent variable (IV: believed to be a causal variable)

– measurement of a dependent variable (DV is assumed to the controlled by the IV)

Ch 5.11

Overview of an Experiment• Determine the hypothesis:

– Cause ----> Effect

• Define the variables of the experiment– IV, DV

• Subjects are randomly assigned to treatments• Evaluate the dependent measures

– Statistical significance testing: are the differences between group due to chance?

Ch 5.12

Clinical Contexts

• Process vs. outcome Research

• Compare therapies

• Compare therapies +/- medications– (Barlow & Durand)

Validity Issues• Internal validity assesses whether the differences

between groups is due to the influence of the IV or some other factor– Experimenter uses a control group that is treated

identically to the experimental group, with the exception of receiving the treatment

• Confounds are variables that complicate experimental interpretation– E.g. passage of time or life events

Ch 5.13

Placebo Effects• A placebo effect reflects a therapy effect due to the

expectations of the person rather than a treatment effect– Placebo effects can exert significant change and can endure

for some time

• Experimenters employ double-blind procedures to control for placebo effects– Neither the subject or the experimenter know which group the

subject is in; the subject is led to believe they are receiving a treatment

Ch 5.14

External Validity

• External validity refers to the extent to which the results of an experiment can be generalized beyond the conditions of the experiment– Results from laboratory studies may generalize to

more natural conditions – Results from animal studies may generalize to that

of humans

• Balancing internal vs. external validity

Ch 5.15

Analogue Experiments

• Many key issues in abnormal psychology cannot be examined in experiments due to ethical concerns– Effects of stress on likelihood of developing

schizophrenia

• Analogue studies examine a related phenomenon in the laboratory– Analogue is related to the disorder of interest– The participants are similar to those who have the

disorder

Ch 5.16

Single-Subject Research Designs

• Single-subject research involves the sequential application of control and experimental variables for a single person– Reversal design involves measurement of a

behavior during baseline (A), during treatment (B), during baseline (A) and again during treatment (B)

• ABAB design

Ch 5.17

Studying Behavior Over Time

• Cross sectional designs• Study different groups (Cohorts)• Problem: Cohort effect

– (Barlow & Durand)

Longitudinal Designs

• Study same persons• Over time• Problem: Cross generational effect

– (Barlow & Durand)

Combined Research Strategies

• Theory and hypotheses• Correlation and Epidemiology• Lab and Clinic Experiment• Refine hypotheses, methods, etc.• SHOW ME, and SHOW ME AGAIN

» (Barlow & Durand)