CHAPTER 6: Case Studies and Observational Research1

Chapter 6: Case Studies and Observational Research

A. LEARNING OUTCOMES. After studying this chapter students should be able to:

Discuss the general characteristics of case studies.
Describe advantages and disadvantages of case study research.
Discuss the general characteristics and purposes of observational research.
Identify and illustrate different types of observational research, and explain the pros and cons of each.
Describe how observational researchers sample and record behavior.
Discuss potential problems that can arise when observing behavior.
Describe several types of unobtrusive measures and archival records.
Discuss the limitations of using unobtrusive measures and archival records.

B. KEYWORDS

Archival records / Observational research
Behavioral coding system / Observer bias
Blind observation / Observer rating and ranking scales
Case study / Participant observation
Diary / Physical trace measure
Disguised versus undisguised observation / Reactivity
Ethnography / Scan sampling
Focal sampling / Situational sampling
Habituation / Structured observation
Interobserver reliability / Time sampling
Naturalistic observation / Unobtrusive measures

C. BRIEF CHAPTER OUTLINE

I. Case Studies

Basic Characteristics
Why Conduct Case Studies?
Concerns About Case Studies

II. Observational Research

Basic Characteristics
Why Conduct Observational Research?
Types of Observational Research
Recording Observation
Sampling Behavior
Concerns About Observational Research

III. Unobtrusive Measures and Archival Records

D. EXTENDED CHAPTER OUTLINE

*Much of this summary is taken verbatim from the text.

Introduction

This chapter focuses on two important types of descriptive research, case studies and observational research, the latter placing the scientist in the role of systematic observer. The chapter also examines how behavior can be observed indirectly.

Part I: Case Studies

Basic characteristics. A case study is an in-depth analysis of an individual, social unit, or event. Case studies employ a variety of research methods to understand, fully, the case of interest. Researchers typically begin by observing and interviewing the patient and his or her family and friends, as well as performing physiological and psychological tests. A case study may be reported as an entire book, or many case studies might be reported in a single journal article.
Why conduct case studies? First, unlike some other research methods, case studies provide the unique opportunity to peer deeply into the psychological world of the case. Second, case studies can provide insight into the possible causes of behavior and thus lead to hypotheses that are tested using other research methods. Case studies can also test hypotheses and provide evidence that supports or contradicts a theory. Lastly, case studies may provide support for the external validity of findings obtained in experiments and other types of research.
Concerns about case studies typically center around three issues.
First, the methodology makes it difficult to draw causal conclusions. This is because there are many factors that affect the way a person behaves, and case studies do not allow a researcher to identify specific factors while controlling all other possible factors.
Second, the generalizability of the findings is limited. Studies that involve a single case run an especially high risk of having low external validity.
Finally, there is potential for observer bias since a researcher’s expectations may affect how he or she interprets the case.

Part II: Observational Research

Basic characteristics. Observational research encompasses different types of nonexperimental studies in which behavior is systematically watched and recorded. In most observational studies, researchers measure the behavior of multiple people or nonhuman animals in real time or from after-the-fact video recordings. The findings may be reported qualitatively, quantitatively, or both ways (mixed models).
Why conduct observational research? An advantage of observational research over a case study is that it can measure the behavior of many participants. This not only increases the external validity of the research, but it also enables the researcher to examine relations among variables through exploration or hypothesis testing. Observational research also has the advantage of examining relationships between variables that cannot be done experimentally due to practical and ethical constraints.
There are many types of observational research, their differences being based on the naturalness of their setting, whether participants are aware they are being observed, and the degree to which the observer intervenes with the situation.

In naturalistic observation researchers passively observe the behavior of participants in their natural environment. Undisguised naturalistic observation occurs when participants are aware they are being observed. This type of observation introduces the problem of reactivity, when the process of being observed causes participants’ behavior to change. One way to reduce reactivity is to conduct disguised naturalistic observation so participants do not realize they are being measured.
The advantage of naturalistic observations is that they are more externally valid than laboratory experiments. Still, naturalistic observations occur in a single setting, but in order for results to be generalizable they must apply to a variety of settings.
The disadvantages of naturalistic observation have to do with the complexity of behavior, the lack of control over the research setting, and the practical difficulties of observing every important behavior that takes place. There are also ethical considerations when conducting observational research. While informed consent is mandatory in most research, it is not required if (1) the study will not cause participants harm or distress, (2) confidential information is protected, and (3) the participants would not be exposed to social, economic, or legal risks if their responses were to become known.
In participant observation the observer becomes a part of the group or social setting being observed. In disguised participant observation the researchers do not reveal themselves to the participants as such, while in undisguised participant observation the participants are fully aware of the observer’s role. By making it explicit that the “new person among us” is a researcher, the potential ethical issues surrounding deception are alleviated. Ethnography is a qualitative research approach in which participant observation is combined with interviews to gain an integrative description of social groups.
The key advantage of participant observation is the opportunity to study people’s behavior from the viewpoint of an insider. In addition, there are some instances in which the only way to understand a group is to gain access covertly.
Disadvantages of participant observation include an increased risk that the researcher will influence the way the participants behave due to his or her more active involvement with the group.
In structured observation a researcher fully or partially configures the setting in which behavior will be observed. The setting (situation) is designed to be analogous to what occurs in real life and can be simulated in either a laboratory or in a natural setting.
The advantages of structured observation is increased efficiency and control.
The key disadvantage of this type of observation is that the environment is only a simulation of an environment and not real life setting.

Recording observations can be done in many ways, including:
Narrative records provide a detailed, ongoing description of behavior that is used for later analysis. The record can be created electronically or by writing it down. Field notes are less comprehensive than narrative records in that only the most important behaviors are recorded after the observation has taken place.
Behavioral coding systems involve classifying participants’ responses into mutually exclusive categories. Curtis, Smith, and Smoll (1979), for example, created a list of twelve behaviors exhibited by youth sports coaches, then made note of the frequency of each behavior when exhibited in a game.
Observer rating scales and ranking scales are used to evaluate participants’ behavior or other characteristics. Observers are typically the researchers themselves, or a highly trained assistant. Sometimes a researcher will ask participants to act as self-observers by having them keep a diary. While diaries have their use in research, studies that rely solely on participant diaries are not considered true observational studies. It is the presence of trained observers and a focus on research that makes research “observational.”
Sampling behavior. In observational research it is impossible to observe every member of the population or to observe every behavior exhibited by any given participant. Before any observational study, therefore, a researcher must decide who they are going to observe, what behaviors they will observe, and how often and for how long they will be observed.
In focal sampling a particular member is observed at any given time. Over the course of the study all members of a group will have been measured multiple times (i.e., 10 minutes during recess).
In scan sampling the observer rapidly scans each member of a group at a preselected time so that the entire group is observed within a relatively short period of time.
Situational sampling is used to establish diverse settings in which behavior is observed to increase the external validity of the study.
If researchers cannot record behavior continuously, they may use time sampling to select a representative set of time periods during which observations will occur.

Concerns about observational research. Like case studies, observational research has several concerns.

First, observational research is unable to draw causal conclusions about the relatedness among variables, primarily due to the lack of control.
Second, observational research increases the likelihood of observer bias. Some ways to minimize bias in measurement include creating a well-developed coding system with clear operational definitions of the various categories, employing rigorous observer training, and conducting periodic checks to make sure that observers are using the system reliably. Using blind observation whenever possible to obtain data is another safeguard against observer bias.
Observer training and reliability is another concern in observational research. One way to determine whether the measures obtained through observation is by using more than one observer to record behavior. Doing that allows the researcher to use statistical procedures to establish interobserver reliability, the degree to which independent observers show agreement in their observations. Obtaining high interrater reliability requires a well-developed coding system and extensive observer training.
Reactivity is also a potential issue in observational research. One way to minimize the likelihood that the measured behavior is due to reactivity is based on the principle of habituation—the idea that one’s response to a stimulus decreases with repeated exposure to the stimulus. In other words, if the researcher wants to obtain a nonreactive measure of behavior she should wait until the participants get used to her presence.

Part III: Unobtrusive Measures and Archival Research

An unobtrusive measure assess behavior without making people aware that their behavior is being measured. This is what researchers use to reduce reactivity.

E. LECTURE AND CLASSROOM ENHANCEMENTS

PART I: Case Studies

A. Lecture/Discussion Topics

Psychology students know that the hippocampus is a brain region critically involved in certain types of learning and memory. Our understanding of the importance of the hippocampus is based on the case of Henry Molaison, known as H. M. until he died. H.M. suffered so badly from seizures that neurosurgeon William Scoville decided to take out the part of H. M.’s brain from which the seizures originated. This part of the brain included the hippocampus. The surgery was a success, but it came with a high price—H.M. could no longer make new long-term memories. His profound anterograde amnesia led to an extensive case study. The first report on his case (of many) was published in 1957 by Scoville and neuropsychologist Brenda Milner. H. M. died in 2008 but he and his brain are still being studied today. His brain was donated to science and is being examined by experts in the field to better understand the contribution of the hippocampus to learning and memory. Although it’s possible that without H. M. the function of the hippocampus would have still been discovered eventually, his unique case created an impetus for studying the brain region that until then had been believed to have no role in learning and memory.
The case study is presented to students as a specific research method, yet it’s really more of a collection of research methodologies than a method in and of itself. When a researcher does a case study he or she uses a variety of other research methodologies to examine a patient or an event in a comprehensive way. The combination of surveys, physiological tests, observation, and interviews allows the researcher to collect a broad array of data in order to obtain a more in-depth understanding of the patient or event.

B. Classroom Exercise

Have students select one of the case studies from the list below. Ask them to summarize the case, making sure to include what significant event happened to the individual and what the psychological effects were observed as results of the event. Finally, have them identify one empirical paper that tests a hypothesis based on the case. This exercise can be administered as an individual assignment or groups can work together to create a presentation. The purpose of the exercise is to help students appreciate how unique cases are sometimes an important first step in learning about behavior.
Potential cases: H. M., Genie, Phineas Gage, Little Hans, The Wild Boy, and Joan/John

C. Web Resources

This webpage describes the postmortem analysis of H.M.’s brain. There are links to news article and videos about “Project H.M.” http://thebrainobservatory.ucsd.edu/hm
The National Center for Case Study Teaching in Science provides a database of case studies. Students may search the database by keyword or discipline. http://sciencecases.lib.buffalo.edu/cs/

D. Film Suggestions

Sybil is a movie based on a true story of a woman diagnosed with multiple personality disorder. It can be used to discuss case studies in psychological research.

Babbin, J., Capice, P. & Dunne, P. (Producers). Petrie, D. (Director). (1976). Sybil. Lorimar Productions: Los Angeles

E. Additional References

Sacks, O. (1998). The Man Who Mistook His Wife for a Hat: And Other Clinical Tales. Simon & Schuster.

Stake, R. E. (1995). The Art of Case Study Research. Sage.

PART II: Observational Research

A. Lecture/Discussion Topics

Students as participant observers. Participant observers are those who become a part of a group in order to study its behavior. Many counseling psychology and clinical psychology graduate programs use this methodology to help future therapists understand group dynamics. For example, in my university’s Group Therapy course students spend a portion of each class period in a group setting, with the instructor serving as the group counselor. Throughout this process each student becomes not only a member of the group who not only participates in its activities but also observes the behavior of the group itself.
The ethics of electronically recording behavior. The fact that the APA code of ethics has a specific guideline (8.03) for recording voices and images in research suggests that the use of audio and/or video recordings add another ethical element to research. Why might recording behavior in this manner be of such concern?
Observational research. Discuss with the class the types of research that might benefit from observational research. What particular disciplines in psychology are most likely to employ observational research methods (e.g., social psychology, anthropology)? What particular disciplines are least likely to use observational methodology (e.g., biological psychology, cognitive neuroscience)? This discussion can be used to emphasize the importance of how the research question drives the research methodology used.

B. Classroom Exercise

Developing behavioral checklists. Help students gain hands-on experience in developing behavioral checklists. First, divide students into groups of 3 to 5. Second, assign to each group a particular behavior that they will observe. Example behaviors include (1) engaging another person in conversation, (2) studying, (3) napping. Next, have the students establish a working checklist for the assigned behavior. Then have them observe students in common areas on campus (e.g., cafeterias, library) and use the checklist to record the behavior. This exercise can be used to demonstrate not only how tricky behavioral checklists can be but that they typically need several revisions in order to be good instruments to measure behavior.

C. Web Resources

Observational research. This great clip describes what observational research is. It also provides a clear and concise description of its value to observational methods.

D. Film Suggestions

The Departed is a movie about a police officer who goes undercover as a mobster. He is accepted into the crime family and uses his position to help bring down their operation.

Brown, G. M., Davison, D. Hahn, Kristin, K., Lee, R., & Nunnari, G. (Producers). Scorsese, M. (Director). (2006). The Departed. Plan B: New Orleans.

E. Additional References

Studies with participant observers.

Humphreys, L. (1975). Tearoom Trade: Impersonal Sex in Public Places. Aldine.

Kawulich, B. B. (2005, May). Participant observation as a data collection method. In Forum Qualitative Sozialforschung/Forum: Qualitative Social Research (Vol. 6, No. 2).

PART III: Unobtrusive Measures and Archival Records

A. Lecture/Discussion Topics

Archival research. Historians rely heavily on archives in conducting research. A friend of mine, a medieval historian, travels to England each year to go through archives that date back hundreds of years. Historians, though, are not the only people who conduct archival research. A concerned citizen may examine police records to determine if the new stoplight was effective at reducing speeding by counting the number of traffic violations in the year before and the year after the light was installed. The concept of archives use is really the same whether it is a medieval collection of records or a present-day collection of records. A lot of things have changed from the 1500s to today, however, most notably the use of computers to create and store archives. That said, what issues, good or bad, are associated with using electronic archives and/or handwritten archives? How might each medium affect one’s analysis of the data?

B. Classroom Exercise