Lecture 3: Chapter 4: Research Methods in Psychopathology

Question 1

What is a theory and what is a hypothesis?

Answer 1

Theory: set of propositions meant to explain a class of observations

Hypothesis: expectation about what should occur if a theory is true

Question 2

What are 3 common research designs in psychopathology?

Answer 2

1. Case study
2. Correlation
3. Experiment

Question 3

What are case studies?

Answer 3

Collection of detailed biographical information, often from 1 patient

Question 4

Give 3 advantages and 2 disadvantages of case studies

Answer 4

+:
1. Source for hypotheses that can be tested in quantitative research
2. Provides info about novel cases or procedures
3. Can disconform a universal relationship

-:
1. May be biased by observer
2. Can’t provide causal evidence, because alternative hypothesis can’t be eliminated

Question 5

What is a correlational study?

Answer 5

Study of the relationship between two or more variables, measured as they exist in nature

No assessment of causality possible

Question 6

When are correlational studies often used?

Answer 6

1. When we can’t manipulate risk variables (trauma, genes) or diagnoses
2. Used by epidemiologists to study incidence, prevalence and risk factors of disorders
3. Used in behavioral genetics research to study heritability

Question 7

What is an experimental study?

Answer 7

Manipulated independent variable and measured dependent variable, often with at least one control group and random assignment

Can assess causality

Question 8

What is the downside of single-case experimental designs?

Answer 8

Limited external validity

Question 9

When is the experimental method used most in psychopathology?

Answer 9

For determining causal relationships in studies of treatment

Question 10

What is the difference between statistical and clinical significance?

Answer 10

Statistical: with numbers saying if a relationship is significant in relation to the population

Clinical: whether a relationship between variables is large enough to matter

Question 11

What is the directionality problem? How can you deal with it (2)?

Answer 11

Often in correlational studies: don’t know which way the relation goes

Fix by using a longitudinal design/ high-risk method: test whether causes are present before a disorder developed

Question 12

What is the difference between a cross-sectional and a longitudinal design?

Answer 12

Cross: researcher measures causes and effects at same point in time

Long: research over a long period of time

Question 13

What is the high-risk method?

Answer 13

It overcomes the problem of doing a longitudinal study to claim causation with the risk that there is no good quantity of data yielded by this study

High-risk method: researchers study only people with above-average risk of developing a specific disorder

Question 14

What is the third variable problem?

Answer 14

When a third factor/confound may have produced the result of research

Question 15

What is the family method in correlational behavior genetic research?

Answer 15

Method to study genetic predisposition among members of a family

Question 16

What are index cases or probands?

Answer 16

The sample of recruited people with the diagnosis in question

Question 17

What is the cross-fostering method? How does this differ from the adoptees method?

Answer 17

Assessment of children who are adopted and reared completely apart from their biological parents. One of their adoptive parents has a disorder and not their biological parents

Adoptees method: biological parent has disorder, adoptive parents don’t

Question 18

What is the difference between internal and external validity?

Answer 18

Internal: the extent to which the experimental effect can be attributed to the independent variable

External: the extent to which results can be generalized beyond the study

Question 19

What is a reversal design or ABAB design?

Answer 19

Participant’s behavior is carefully measured in a specific sequence

1. An initial time period, the baseline (A)
2. A period when treatment is introduced (B)
3. Reinstatement of conditions of baseline period (A)
4. Reintroduction of treatment (B)

Question 20

What is the main focus of treatment outcome research?

Answer 20

To see if treatment actually works

Question 21

Which 7 criteria should a treatment study include?

Answer 21

1. Clear definition of sample
2. Clear description of treatment offered
3. Inclusion of control condition
4. Experimental design with random assignment
5. Reliable/valid outcome measures
6. Evaluation of outcomes by rater unaware of the patient’s treatment
7. Large enough sample size

Question 22

What are randomized controlled trials (RCTs)?

Answer 22

Studies where clients are randomly assigned to receive active treatment or a comparison (no treatment, placebo, other treatment etc.)

Question 23

What is a double-blind procedure and why would we use it? Why is it hard to implement though?

Answer 23

When the psychiatrist and patient aren’t told whether the patient is receiving active medication or a placebo

Used to reduce bias

Hard to implement, because treatment providers and patients may guess who is getting the treatment, due to side effects of the treatment

Question 24

What is the difference between efficacy and effectiveness?

Answer 24

Efficacy: whether treatment works under the purest of conditions

Effectiveness: how well treatment works in the real world

Question 25

Do RCTs (randomly controlled trials) determine efficacy or effectiveness of treatment? What is the consequence of that

Answer 25

Efficacy –> so RCTs might not inform us about how these treatments work with broader samples with nonacademic therapists –> poor external validity

Question 26

What is dissemination and how is it applied?

Answer 26

The process of facilitating adoption of efficacious treatments in the community

Done by providing guidelines about the best available treatments + trainings for clinicians

Question 27

What is an analogue experiment?

Answer 27

Create/observe a related but less severe phenomenon (analogue of the risk variable) in the lab to allow more intensive study

This would make it possible to study ethically bad experiments (e.g. give mothers both a different condition (good or bad) and wait and see if the kids develop GAD in adulthood)

Question 28

What are 3 types of analogue experiments?

Answer 28

1. Create temporary symptoms through experimental manipulation
2. Use analogue samples of people who are similar to people with certain diagnoses
3. Use animals to understand human behavior

Question 29

What does it mean to have successful replication of findings?

Answer 29

When findings hold up when the study is repeated a second time. Findings are more believable when they’re replicated independently

Question 30

What is the publication bias?

Answer 30

Tendency to publish only positive results

Question 31

What are 4 issues in research methods that can contribute to replication failures?

Answer 31

1. Small samples
2. Use of unreliable measures
3. If methods aren’t described well
4. Some phenomena aren’t generalizable

Question 32

What is p-hacking?

Answer 32

Tweaking analyses or comb through data until they arrive at a significant result

Question 33

What are 5 policies that can address issues in replication?

Answer 33

1. Preregistering hypotheses, measures and analysis plan
2. Make data publicly available
3. Weigh methods more heavily than results
4. Use large sample size
5. Let consumers consider whether findings have been replicated

Question 34

What is a meta analysis and what problem does it try to tackle?

Answer 34

Using different statistical tests to compare different studies. They compare effect sizes across studies and try to estimate publication bias.

It tackles the problem that two scientists read the same set of studies and reach very different conclusions

Question 35

Why are meta analyses criticized?

Answer 35

Researchers sometimes include studies that are of poor quality. Some researchers then give equal weight to all studies, so a poorly controlled study was as important as a well-controlled one