Assessment and Research
Psychologists use many tools and methods to help answer questions pertaining to abnormal psychology. Regardless of the setting or their primary role (clinician, researcher, or some combination of both), psychologists strive to be systematic and scientific in their approach to studying and treating abnormal behavior so that their findings might help further the knowledge base of the field and improve the lives of other people who may be struggling with a similar set of distressing symptoms.
In order to collect data, psychologists must perform some kind of assessment. Assessment comes in many forms to suit many purposes, but all assessments have in common the properties of reliability and validity. Reliability is similar to the concept of consistency or the ability to get the same result each time you measure something. Measuring a person’s height repeatedly with a steel tape measure will give you a highly reliable value (assuming there is no human error in using the tape measure). The scale used (inches and feet) is extremely precise, and the material used to construct the tape measure is highly durable. Imagine instead that you were trying to measure someone’s height with a device employing scale that simply divided everyone into two categories (short or tall) and was constructed of elastic. You would not be able to reliably measure height (especially for those whose height put them around the cutoff between “tall” and “short”) because the scale is extremely broad and the elasticity of the instrument might cause that cutoff to shift around each time you stretched the elastic from head to toe.
If reliability is similar to consistency, then validity is comparable to the concept of fidelity or the idea that you are truly measuring what you think you are measuring. If you wanted to measure a person’s weight, then you would have them step on a good (i.e., reliable) scale of some kind that would give you information about total pounds, kilograms, and so on. However, borrowing from the previous example, imagine again that you are using a steel tape measure (using a scale of inches and feet) but you are interested in measuring a person’s weight. You extend the tape measure from head to toe, record that value, and then draw a conclusion about how much the person weighs based on the numbers generated by your measurement using the steel tape measure. Although height and weight are correlated (the taller a person is, the more he or she weighs, generally speaking), height and weight are separate concepts with different units of measurement (i.e., distance vs. mass). In this example, even though the tape measure is still very precise (i.e., reliable), you have not measured what you originally intended to measure. Therefore, using a measure of distance to assess mass is not a valid assessment. This example also illustrates a very important relation between reliability and validity, because although it is possible to have reliability without validity, you can never have validity without reliability.
Psychologists use statistical procedures (usually a correlation or some variation of it) to help establish and communicate the reliability and validity of an assessment, and both reliability and validity come in various forms. Once multiple types of reliability and validity are established, other psychologists can begin to use an assessment to collect data.
The type of assessments psychologists use depends on the nature of the data and the questions they are trying to answer. Assessments come in several formats, such as self-report instruments, behavioral observations, psychological tests, and physiological measures. Examples of self-report assessments include the Beck Depression Inventory (BDI-II; Beck, Steer, & Brown, 1996), the State-Trait Anxiety Inventory (STAI Form Y; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983), and the Minnesota Multiphasic Personality Inventory (MMPI-2; REF ). The BDI and STAI have a series of statements that are rated on a Likert-type scale based on how much the reader believes the statement applies to his/her symptoms and behavior. The MMPI is a well-known and widely used assessment that consists of over 500 true/false statements. All of these instruments have been standardized and normed on various groups so that psychologists can take the scores obtained in clinical and research settings and meaningfully interpret and compare them relative to the norms.
Behavioral observation is a popular assessment strategy that psychologists have used for many years. It offers objectivity and translates easily into quantifiable values, and can be a simple but elegant way to determine if changes in behavior are occurring. Psychologists design and use various devices (called apparatus) to collect different types of data, depending on what the researcher needs to quantify. Counting the number of times a behavior occurs would require some measure of frequency. The length of time a behavior lasts would be a measure of duration. The amount of time that passes before a behavior occurs is latency. The amount of effort required to engage in an activity or accomplish a task would be a measure of intensity. Variations and combinations of these measures allow psychologists to quantify a number of important dimensions of behaviors. A good example of this combined approach is behavioral coding, which is a form of behavioral observation where expertly trained and supervised researchers watch (either live in an unobtrusive manner or via video recording) for target behaviors and then rate or code them along various dimensions using predetermined definitions and coding sheets.
A good example of a psychological test is the Wechsler Adult Intelligence Scale (WAIS-III; Wechsler, 1997), which has various tasks and several subscales, including domains assessing motor performance and verbal abilities. A trained professional gives standardized instructions for each task, records the responses, and then scores the responses and generates subscale scores as well as an overall intelligence score (sometimes called the intelligence quotient, or IQ). A psychological test like the WAIS is essentially a combination of self-report (since the subject is required to report his or her own level of knowledge about certain things) and behavioral observation or behavioral coding (since the clinician must observe, record, and score responses based on standardized instructions and scoring criteria).
As technology continues to evolve, what psychologists can measure (and the precision they can measure it with) continues to expand as well. Recent advances in medical technology have allowed psychologists to go inside the active mind and body and measure responses that previously they could only speculate about or measure indirectly. A good example of a relatively recent advance is functional magnetic resonance imaging (fMRI), a technique that maps the structure and function of the brain (and other parts of the body) in real time. Previously, psychologists were more limited to techniques that singularly measured electrical activity, muscle activity, and rates of metabolism, for example. Assessments of these characteristics are still useful and being performed in many labs, and have also improved with technological advances such as faster and more powerful computers. All of these technological assessment strategies are highly precise, and the equipment used to measure and record data, while often very expensive, is highly valuable to a psychologist who is helping to study, diagnose, and treat conditions that would otherwise remain a mystery.
No matter the question and the measures used to help provide answers, psychologists recognize that using a single measure of a single type and in a single way is very limiting. Therefore, a good rule of thumb is to take multiple measurements in multiple formats and under various conditions, often referred to as a multimodal, multimethod assessment strategy. This approach gives psychologists a much broader, clearer picture of the individual and his or her behavior and helps put everything into context and perspective.
The research questions and assessment strategies will also directly influence the research methods used. Many studies in abnormal psychology are not true experiments, per se, because it is not possible or ethical to experimentally control or manipulate a certain condition. For example, it is important to know if there are significant sex differences in the prevalence of certain disorders, because this may provide clues as to the etiology of symptoms. However, it is not possible to randomly assign research participants to their sex; they bring this characteristic and all things related to it with them into the study. Psychologists can statistically analyze these variables, also known as subject variables, and treat them as independent variables. However, this does not change the fact that they have not experimentally manipulated anything, and therefore cannot draw cause-and-effect conclusions based on their results. Much of the epidemiological data gathered on the prevalence of mental disorders and their associated features is correlational because of the design limitations mentioned above. However, this research strategy is still useful and important, because it not only provides good descriptive data but also often points researchers toward important questions needing additional experimental exploration.
Research designs that are experimental in the more traditional sense include longitudinal studies and twin studies. Longitudinal studies involve tracking the same people over time, or gathering repeated measures of a behavior, trait, physiological response, and so on. This design is very helpful in terms of documenting changes within people over time. However, its primary limitation lies in the inability to attribute any changes that occur to a specific cause. In other words, if a young man’s symptoms of social phobia improve over time, is it because he received good treatment, is it a matter of development and maturation, or is it simply a matter of severe symptoms moving closer to an average level because statistically there is nowhere else to go? In order to overcome this limitation, scientists must use a control group of some kind.
Studies of twins, particularly identical or monozygotic twins, can provide the control group that longitudinal studies may lack. Because identical twins are exactly the same genetically, studying them provides important clues about the respective influences of nature and nurture in the development of abnormal behaviors and mental disorders. Identical twins reared separately provide a unique opportunity to study two genetically identical people raised in different environments. Tracking identical twins reared together and apart over time can be a very powerful, albeit very resource-intensive, research strategy because it combines the advantages of both longitudinal and cross-sectional methods.
Because of the relative rarity of many disorders, and because of practical and ethical limitations on what researchers are allowed to do, psychologists have developed research strategies and observational methods based on very small sample sizes or even single cases. Descriptive case studies were informal precursors to single case designs used today, because they provided a wealth of information about the background and baseline of a client’s functioning and his or her progression through treatment. The simplest single case design is the treatment withdrawal design, or the ABAB design. This design strategy involves measuring target behaviors during a baseline phase, “A,” followed by the intervention phase, “B,” and then a withdrawal of the treatment, the second “A,” and finally a second implementation of the intervention, the second “B.” This design helps psychologists be confident that any changes that occur within the person are not simply due to the passage of time or spontaneous remission of symptoms. As you can imagine, withdrawing a successful treatment is not always ethical or practical, and in the absence of large, grant-funded treatment outcome studies, many practicing clinicians must be satisfied with AB designs and the hope that it is, in fact, the treatment that contributes to positive changes and symptom reductions.
Read more about Abnormal Psychology:
- What is Abnormal Psychology
- History of Abnormal Psychology
- Diagnosis and Treatment
- Summary and References