Designs
A single-subject design, also referred to as a small-N design, uses the data from one individual participant without averaging it as part of a group of participants. It is assumed that the topic under study is accurately reflected in the single individual and can be controlled appropriately. With this approach, statistical tests are typically replaced with graphical changes. For example, if you wanted to know if a child with autism would respond to a particular type of praise, you could determine this by having a condition in which you give praise and a baseline condition in which you do not. A stronger design, referred to as ABAB, would have a baseline condition (A) followed by the treatment condition (B). This is referred to as a reversal design. Such a design would appear as in Figure 3.9.
single-subject design: also referred to as small-N design, an experiment that uses the data from one individual participant without averaging it as part of a group of participants
Hersen and Bellack (1976) used a multiple-baseline design to demonstrate the effects of a treatment program for a schizophrenic patient. The patient made little contact with others, rarely engaged in conversation, and was compliant even to unreasonable requests. The treatment consisted of training in the development of assertiveness skills and skills for making contact with others. The measures taken over the baseline and treatment sessions were the amount of eye contact while talking, the amount of speaking without prolonged pauses, the number of requests made of another person, and the number of unreasonable requests not complied with. This design requires that baselines be taken for the four measures and that treatments be introduced at different sessions for each of the behaviors to be changed, while measurements of all behaviors are continued. This type of design helps us to determine whether the treatment was specific to a particular behavior (see Figure 3.10). Notice that the treatment was introduced at a different time (dotted vertical line) for each of the four targeted behaviors.
Figure 3.9 ABAB Single-Subject Reversal Design
Source: Dyer, Dunlap, & Winterling (1990).
Figure 3.10 Probe Sessions During Baseline, Treatment, and Follow-Ups for Subject 1
Source: Hersen & Bellack (1976, p. 243).
In a longitudinal study on children with ADHD, a researcher may observe the same group of individuals for a period of years in order to detect changes over time.
© iStockphoto.com/KatarzynaBialasiewicz
Longitudinal Research
A longitudinal design allows the researcher to follow a specific group of individuals across a period of time to document any changes that take place during that period. For example, one study followed children diagnosed with attention deficit/hyperactivity disorder (ADHD) for 9 years and noted changes in specific symptoms (Lahey & Willcutt, 2010). By comparing the children with ADHD to children without the disorder, it was possible to determine normal developmental changes as opposed to changes related to ADHD itself. These researchers were also interested in knowing if ADHD symptoms in year 1 were predictive of symptoms and teacher ratings in later years, which they were. Another study asked if living in difficult neighborhoods and being maltreated as a child would predict drug use in middle adulthood, which it did (Chauhan & Widom, 2012).
longitudinal design: a research technique that allows the researcher to follow a specific group of individuals across a period of time to document any changes that take place
There are a number of advantages to using a longitudinal design. First, it allows us to study the natural history of the development of a mental disorder in comparison with a similar group without the disorder. Second, we can note when and in what manner the changes in the disorder take place. Some disorders, such as schizophrenia and bipolar disorder, tend to have an abrupt onset. Third, longitudinal designs are particularly useful for studying prevention or treatment programs regarding longer-term changes. Of course, the disadvantage of these designs is that they require a significant period of time to complete the study.
Epidemiological Research
Epidemiology is the study of the distribution and determinants of the frequency of a disorder in humans (see Tsuang, Cohen, & Jones, 2011, for an overview). Within psychopathology, epidemiological approaches have been used to determine how frequently a particular disorder is present in men or women and if a particular disorder is related to other factors such as income or level of industrialization of a country.
epidemiology: the study of the distribution and determinants of the frequency of a disorder in humans
Epidemiological research is particularly helpful in determining the nature, etiology, and prognosis of a given disorder. From this type of research, we know that autism begins early in life and continues throughout one’s lifetime. Anxiety disorders, on the other hand, can begin in adolescence, but by early adulthood, a number of individuals who experienced one or more of these in adolescence will no longer experience the disorder. We also know that schizophrenia is seen in similar percentages throughout the world.
prevalence: the proportion of individuals who have a particular disorder at a particular time period
There are a variety of measures used in epidemiological research to describe the statistical profile of psychological disorders. One common measure is prevalence, which is the proportion of individuals who have a particular disorder at a particular time period. If you go to the National Institute of Mental Health (NIMH) website—http://www.nimh.nih.gov/statistics/index.shtml—you can see the prevalence rates for the major disorders from the Diagnostic and Statistical Manual of Mental Disorders (DSM). For example, you will see that the 12-month prevalence for general anxiety disorder (GAD) in adults is 3.1%. What this means is that 3.1% of the U.S. population during a 12-month period had the disorder. Another type of prevalence is referred to as lifetime prevalence. This is the percentage of a specific population that had the disorder at some point in their life, even if they no longer show symptoms of the disorder currently. As would be expected, lifetime prevalence is always larger than the number of individuals who have the disorder during a given 12-month period. In the case of GAD, lifetime prevalence is 5.7%. Further types of epidemiology data can include prevalence rates of the disorder in terms of specific factors such as gender, age, or average age at onset of the disorder. These data are shown for GAD in Figure 3.11.
lifetime prevalence: the percentage of a specific population that had the disorder at some point in their life, even if they no longer show symptoms of the disorder currently
Epidemiological studies also allow us to look at particular populations. For example, we could ask what percentage of individuals in prison has a mental disorder. These data were collected in 2002 and 2004 and are shown in Figure 3.12. From this graph, it can be seen that there are higher rates of inmates with mental disorders in local as compared with federal prisons. Although not shown in the graph, in the United States, there are now more than 3 times more seriously mentally ill individuals in jails and prisons than in hospitals (Torrey, Kennard, Eslinger, Lamb, & Pavle, 2010).
