Psychology 350
Psychopathology

Dr. Salvatore (Sam) Catanzaro
Executive Associate Dean, College of Arts and Sciences
Professor of Psychology

Illinois State University

Dr. Catanzaro's Home Page

PSY 350 Links

What's new

Syllabus

• General
• Evaluation
• Policies
• Study Hints
• Schedule

WebBoard
Assignments
Handouts
Lectures
Grades

Psych Resources
on the Web

Association for Psychological Science

American Psychological Association

The National Institute of Mental Health

APA Style Home Page

Additional Resources

Computer & Web Training at Illinois State

The IRB at Illinois State University

This handout is designed to provide an overview of material we did not get to in class. It complements the material in Chapter 2 of the textbook on research methods.

Case Studies are used to provide an intense analysis of a particular person's problem(s). Freud was famous for his insightful case studies. Unfortunately, the conclusions we can draw from case studies are limited, because they are easily influenced by biases and assumptions of the person doing them. They can never be used to confirm or prove a hypothesis or theory. However, case studies can be very useful to suggest new hypotheses that can be evaluated more carefully with other research methods. They can also disconfirm a theory by showing an example that demonstrates that the theory is wrong.

Correlational Studies examine the naturally occuring relationships between measured variables. We use correlational studies to answer questions like:

• Do people with higher incomes have greater levels of happiness?
• Do people who experience more life stress have poorer health?

The correlation coefficient is a number that can be anywhere from -1.0 to +1.0. The sign, minus or plus, tells us the direction of the relationship: Negative correlations tell us that as one variable increases, the other variable decreases, whereas positive correlations tell us that as one variable increases, the other variable also increases. The size of the number in the correlation coefficient tells us the strength of the relationship. A correlation of -.50 is a stronger relationship than a correlation of +.35.

Correlational studies can tell us how variables are related but they provide limited insight into causal processes--correlation is not evidence of causality. We have to be especially careful about what is called the "Third Variable." This refers to the possibility that the two variables we correlate are only related because there is another (third) variable that influences them both but because we haven't measured it, we know nothing about it. Here's a couple of examples that illustrate correlations best explained by third variables:

• In northern European cities, the birth rate is positively correlated with the number of stork's nests. We cannot conclude that babies are brought by storks. Rather, if we think about it, we realize that storks like to build nests on rooftops, and in cities with more people, there are more roofs, therefore more storks nests--and more babies.
• If we measured the vocabulary and the show size of all the children in a grade school, we would see a strong positive correlation. But we would know that learning more words doesn't cause children's feet to grow (nor would we think that having bigger feet causes kids tohave bigger vocabularies). We would realize that both show size and vocabulary are influenced by age, the third variable that explains the correlation.

Longitudinal studies (sometimes called follow-up studies) are a variation on correlational studies in which people are tracked over time--and the variables are correlated with each other over time. This method allows us to examine developmental processes. we might be able to learn whther changes in one variable precede changes; this might help us learn more about possible causal processes.

Another important version of the correlation study is the Comparative Study. This is a study in which we identify groups based on different characteristics and compare them on some variable. Examples include comparisons of males and females, or of individuals with different diagnoses. Even if such a study does not use a correlation coefficient as a test statistic, the logic of the study is correlational--we are interested in the relationship between the grouping haracteristic (e.g., gender) and the other variable (e.g., level of depression).

Epidemiological Research is dedicated to identifying the rates of disorders for particular groups of people, or populations. It can also identify demographic correlates or risk factors for disorders. When we hear about the rate of bird flu spreading around the world, or the risk of HIV infection in different countries, we are hearing the results of epidemiological research. Epidemiological research has shown us that all other things being equal, women have about twice the risk of being diagnosed with depression as men do--and that this gender difference appears in early adolescence.

Two key terms from epidemiological research are lifetime prevalence and relative risk.

• Lifetime Prevalence:  proportion of people in a sample who have ever experienced disorder (Table 2-2)

• Relative risk:  odds of having a disorder given some information (Table 2-3)

True experiments involve special features that make them powerful methods for testing causal hypotheses--See pages 67-70. They have some, but limited utility in psychopathology, because of ethical and practical limitations .

Experiments of Nature is a term used to refer to research that investigates the impact of some significant and naturally occurring event. These are usually major stressful events, like a natural disaster or war. Sometimes, we are able to make these studies prospective if we happened to be doing some other kind of research and then the event occurs. A famous example studied a group of college students who had been part of a study (for extra credit!) in California and then experienced a major earthquake. We learned, among other things, that earthquake survivors experience more nightmares.

I urge you to use this handout, as well as the other study aids available to assist in your preparation for this exam. Good luck!

Salvatore J. Catanzaro, Illinois State University, Department of Psychology