`CHAPTER 4: RESEARCH METHODS

Examining Psychopathology: Psychopathology is complex, influenced

by biological, psychological, and social factors, and hard to study directly.
Researchers use scientific methods to identify problem behaviors, their
causes, and effective treatments. Understanding research helps
distinguish fact from fiction and make informed mental health decisions.
Important Concepts: Psychopathology research examines three key
questions: First, “What problems cause distress and impair functioning?”
Second, “Why do people behave in unusual
ways?” Third, “How do we help them behave in more adaptive ways?”
Understanding research methods is essential for evaluating these
questions.
Basic Components of a Research Study: A research study starts with
a hypothesis, then uses a design to test it by measuring the dependent
variable (what is affected) and the independent variable (what influences
it). Internal validity checks if the independent variable truly causes the
change, while external validity assesses how well results apply beyond the
study.
Hypothesis: Humans naturally seek meaning and order, which drives the
study of psychopathology. Scientists form hypotheses—educated, testable
guesses—about behavior or mental disorders, such as whether violent TV
increases aggression or specific therapies reduce anxiety. For example,
Wood et al. (2020) tested whether traditional CBT or an ASD-adapted CBT
reduced anxiety in children with autism compared to treatment as usual
(TAU). The dependent variable was anxiety, measured with tools like the
Pediatric Anxiety Rating Scale, and the independent variable was the type
of treatment. The study found both CBT approaches effective, with the
adapted CBT showing extra benefits.
Internal and External Validity
Internal validity refers to whether the independent variable (e.g., a
treatment) truly causes changes in the dependent variable (e.g., anxiety).
Confounding variables—like differences in IQ across groups—can threaten
internal validity. Researchers use control groups, randomization, and
analogue models to reduce these confounds.
External validity is the extent to which results generalize beyond the
study participants to the broader population. High control can improve
internal validity but may reduce generalizability, creating a trade-off.
Balancing internal and external validity often requires multiple, related
studies.
Statistical versus Clinical Significance:
Statistical significance shows that a result is unlikely due to chance, but it
doesn’t indicate whether the effect is meaningful in real life. For example,
a study found anorexia patients drank more smoothie after a positive
video than a neutral one—a statistically significant difference—but still far
less than healthy individuals, so the change may not be clinically
meaningful.
Clinical significance considers whether results truly improve functioning.
Researchers use effect size to measure the magnitude of change and
social validity, gathering input from participants and significant others to
see if the change matters in daily life.
The “Average” Client
The patient uniformity myth occurs when researchers treat all participants
as if they respond the same, ignoring individual differences. Group
averages can hide variations—some may improve with a treatment, while
others worsen. Because people differ in age, abilities, gender, and history,
statistically significant results may not predict outcomes for every
individual. Clinicians must consider this heterogeneity when applying
research to practice.
Types of Research Methods:
Studying Individual Cases: case study method examines one or a few
individuals in detail to explore new or rare disorders. It relies on
observation rather than experimental control, so results may be
influenced by confounding factors and coincidences. Historically, case
studies contributed to major developments, such as Freud’s
psychoanalysis and Masters & Johnson’s sexual research. While
informative, case studies have limitations and can be misleading if
generalized too broadly or sensationalized.
Research by Correlation: examines whether two variables are related
without manipulating conditions. Positive correlations mean both increase
together, negative correlations mean one increases as the other
decreases, and zero means no relationship. Correlation does not imply
causation; a relationship could reflect reverse causality or a third factor.
Correlational research helps identify associations but cannot determine
which variable causes the other.
Epidemiological Research:
Epidemiology studies the incidence, distribution, and impact of disorders
in populations. By tracking prevalence (how many people have a disorder
at a given time) and incidence (new cases over a period), researchers
identify patterns and risk factors. For example, COVID-19 research
highlighted vulnerable groups, and studies of binge drinking show trends
across college students and states. Epidemiology has also revealed
causes of disorders, such as Joseph Goldberger linking niacin deficiency to
a mental disorder in low-income populations. While it cannot prove
causation, epidemiology provides valuable insights into the spread, risk
factors, and potential interventions for psychological disorders.
Research by Experiment: manipulates an independent variable to test
causality. For example, increasing social support and observing changes in
psychological disorders can suggest whether lack of support contributes
to these problems. While experiments provide stronger evidence than
correlations, results must be interpreted cautiously, as complete certainty
is rarely possible. Different experimental designs help researchers better
understand psychopathology.
Group Experimental Designs: In group experiments, researchers
manipulate an independent variable to see its effect on a group’s
behavior, such as testing a treatment for insomnia in older adults. Clinical
trials are formal experiments to evaluate treatment effectiveness and
safety. Variants include randomized clinical trials (participants
randomly assigned), controlled clinical trials (using control groups for
comparison), and randomized controlled trials (combining
randomization and control groups), which are considered the gold
standard in treatment research.
Control Groups: are similar to experimental groups but do not receive
the treatment, allowing researchers to see if changes are due to the
intervention. Placebo controls help account for participants’ expectations
of improvement, while double-blind controls prevent both participants and
researchers from biasing results. These strategies ensure that observed
effects are more likely due to the treatment itself rather than other factors
like expectations or researcher influence.
Comparative Treatment Research: instead of comparing a treatment
to no treatment, researchers can compare different treatments for the
same disorder. Process research examines why a treatment works by
identifying the active components that cause change, while outcome
research examines what changes occur, such as improvements or side
effects. Understanding both helps refine interventions, focus on effective
elements, and develop better treatments.
Single-Case Experimental Designs: developed by B. F. Skinner, this
method studies individuals systematically under different conditions. It
focuses on understanding the behavior of one person in depth rather than
averaging across groups. Unlike case studies, single-case designs use
strategies to improve internal validity and reduce confounding variables,
making them valuable for studying both the causes and treatment of
psychopathology.
Repeated Measurements: in single-case designs, behaviors are
measured multiple times rather than just before and after treatment. This
approach tracks variability and trends over time, helping determine
whether changes are due to the treatment or just natural fluctuations. For
example, Wendy’s daily anxiety ratings might vary from 5 to 10, so a
simple before-and-after comparison could be misleading. Repeated
measurements reveal the level, variability, and trend of behavior,
providing a clearer picture of treatment effects.
Withdrawal Designs: in single-case research, a withdrawal design tests
whether a treatment causes behavioral change by removing it after a
period of use. First, a baseline is established, then the treatment is
applied, and finally it is withdrawn to see if behavior returns to baseline. If
behavior worsens after withdrawal, the treatment likely caused the
improvement. Unlike case studies, withdrawal designs actively test
causality. However, they may be ethically or practically difficult when
treatments cannot be removed or unlearned. Brief withdrawals or drug
holidays can still provide useful insights.
Multiple Baseline: evaluates treatment effectiveness without
withdrawing it. Treatment is introduced at different times across settings,
behaviors, or individuals. For example, Wendy’s anxiety could be treated
first at home, then at work; improvement only after treatment begins in
each setting indicates effectiveness. This method improves internal
validity by ruling out alternative explanations. Multiple baseline designs
can be used with several participants to strengthen external validity and
often reflect natural treatment implementation, making them practical
and ethical when withdrawal designs are not feasible.
Genetics and Behavior across Time and Cultures: Understanding a
person’s behavior or disorder requires considering inherited influences,
how behavior changes over time, and cultural effects. Research replication
and ethics are also key elements in studying behavior.
Studying Genetics: genetics influences behavior through the interaction
of inherited traits and experiences. Behavioral geneticists study
phenotypes (observable traits) and genotypes (genetic makeup), as
well as endophenotypes, the genetic mechanisms underlying
psychological disorders. Advances like the Human Genome Project have
mapped genes linked to inherited conditions. Research strategies in
behavioral genetics include basic and advanced genetic epidemiology,
gene finding, and molecular genetics, using methods such as family, twin,
and adoption studies, as well as linkage and association analyses to
explore how genes contribute to psychological disorders.
Family Studies: examine whether a trait or disorder runs in families. The
individual studied is called the proband. If genetics play a role, first-
degree relatives should show the trait more often than distant relatives or
the general population. For example, blood–injury–injection phobia shows
up in about 60% of first-degree relatives, one of the highest rates of
familial aggregation. However, shared family environments can also
influence these patterns, making it difficult to separate genetic from
environmental effects.
Adoption Studies: help separate genetic from environmental influences.
Researchers study adoptees with a disorder and compare them to
biological relatives raised in different families. If the disorder appears in
relatives raised apart more often than by chance, genetics likely play a
role.
Twin Studies: compare identical (monozygotic) twins, who share nearly
all genes, with fraternal (dizygotic) twins, who share about 50% of their
genes. Greater similarity in identical twins suggests a genetic influence.
For example, identical twins resemble each other more than fraternal
twins in traits like height or antisocial behavior. Research shows family
environment strongly affects juvenile antisocial traits, but genetics play a
larger role in adulthood. Combining twin and adoption methods allows
researchers to better estimate the effects of genes versus environment.
Genetic Linkage Analysis and Association Studies
Family, twin, and adoption studies can suggest a genetic component for a
disorder but cannot locate the gene.
 Genetic linkage analysis looks for a connection between a
disorder and known genetic markers in families. If a marker and the
disorder are inherited together, the genes are likely near each other
on the chromosome. For example, studies on bipolar disorder
suggested possible links on chromosome 11, but such findings often
fail to replicate.
 Association studies compare genetic markers in people with a
disorder to those without it. Markers more common in affected
individuals are likely near relevant genes. This method is better for
identifying genes weakly associated with a disorder.
Both approaches help pinpoint genetic contributions and guide potential
treatments.
Studying Behavior over Time: Understanding how a behavior or
disorder changes over time helps determine whether treatment is needed
and how problems develop. For example, normal grief may fade without
intervention, while persistent aggression in children usually requires early
action. Prospective studies, which track changes as they happen, can
reveal patterns of disorder onset—such as autism spectrum disorder—that
differ from retrospective reports, giving more accurate insight into
development and treatment.
Prevention Research: aims to reduce psychological problems by
studying risk and protective factors over time. Strategies include health
promotion, universal prevention, selective prevention, and indicated
prevention. Health promotion builds skills in whole populations, universal
prevention targets specific risk factors, selective prevention helps at-risk
groups, and indicated prevention addresses early signs of problems.
Effectiveness is evaluated using correlational and experimental research
methods.
Cross-Sectional Designs: compare different age groups at a single point
in time to study development or behavior. Each age group, called a
cohort, may differ not only in age but also in cultural and historical
experiences, which can confound results (cohort effect). For example,
studies on teen alcohol use show that attitudes before drinking influence
later behavior. Cross-sectional designs are easier than longitudinal studies
but cannot show how problems develop within individuals over time.
Longitudinal Designs: track the same group over time to study
individual changes, avoiding cohort effects. They are time-consuming and
costly and may face cross-generational limits. Sequential designs combine
longitudinal and cross-sectional methods, such as studies following
children’s beliefs about smoking from middle school to adulthood, showing
changes in risk perception and the importance of early prevention.
Studying Behavior across Cultures: shows that psychological
disorders and treatments vary across cultures, making comparisons
complex. Symptoms, norms, and treatment models differ—for example,
depression looks different in Nigeria, the U.S., and China, and therapy
approaches vary in Japan and Saudi Arabia. Most research is Western, so
studying multiple cultures improves understanding and treatment of
disorders.
Power of a Program of Research: Research is most effective as a
program rather than a single study. Durand studied why children with
autism show behaviors like self-injury or aggression. Early single-subject
studies showed these behaviors often serve as basic communication, and
teaching functional communication reduced them.
A 3-year longitudinal study found that severe behavior problems
were linked to parents’ pessimism about helping their child. A
follow-up clinical trial using a cognitive-behavioral intervention for
parents increased optimism and improved child outcomes.
This shows that understanding and treating disorders requires combining
multiple research strategies in an integrated program of research.
Replication: is key in science. Researchers are skeptical of single
findings, so repeating studies builds confidence that results are reliable.
The more a behavior or outcome is observed consistently across studies,
the more certain scientists are about its cause.
Research Ethics: protect participants in psychopathology studies.
Informed consent is required, though it can be challenging with children,
cognitively impaired individuals, or across cultures. IRBs review studies to
ensure safety, confidentiality, and proper debriefing if deception is used.
Special guidelines exist for research with children. Participatory action
research now involves people with disorders in designing and interpreting
studies to improve ethics and relevance.