Understanding the different types of scientific studies that psychologist’s use is not difficult at all. In fact, once you learn the basic methodologies and a few terms it is easy to put into practice. While I probably practice Naturalistic Observation the most, my two favorite types of studies are Surveys and the Ex Post Facto. I love the survey method because of the challenge it presents and the practicality of it, while it is not the most reliable method, it is fun to try and come up with survey questions that will work. I also like the Ex Post Facto method. It seems more reliable than the Survey method and it could be just as practical.
Naturalistic Observation is a type of study that observes people in their environment in the setting and situation in which behaviors naturally occur (Myers & Hansen, 2006). For example, a researcher may observe a family at home or in a location that they normally frequent. While this type of study can help a researcher observe real behaviors, it does not tell the researcher why the behaviors are occurring and it is not always a practical type of research (Myers & Hansen, 2006).
Survey research is usually accomplished in the form of a questionnaire or scale. This type of research is simple to conduct and is relatively easy to find participants for, however if the researcher is not careful he or she can word the questions in a leading fashion (Myers & Hansen, 2006). This type of study is also very difficult to measure. Data that is collected can be flawed and participants do not always answer questions unbiased (Myers & Hansen, 2006).
Correlational Studies are designed to study the relationship between two events (Myers & Hansen, 2006). For example, a researcher may study the relationship between children’s behavior and the time of day or the amount of sugar the child has eaten. These studies are often very difficult and time consuming to conduct. There are many variables that have to be eliminated. It is important not to confuse a correlational study with an experiment. A correlational study cannot prove cause and effect, it simply shows a relationship.
Ex Post Facto
The Ex Post Facto study is a type of study that falls somewhere between a correlational study and an experiment. In an Ex Post Facto study the researcher examines preexisting subject variables by forming groups based on the differences (Myers & Hansen, 2006). For example, a researcher may put the introverts in one group and the extraverts in the one group and then determine how many in each group choose a certain food. Unlike experimental research, the researcher in the Ex Post Facto study does not manipulate anything.
Important Terms to Know
External validity determines how well an experiment applies to real life, while internal validity refers to the reliability of the study itself. To say that an experiment has internal validity means that the researcher can be reasonably certain that the independent variable caused the changes in behavior that was observed (Myers & Hansen, 2006).
Predictive validity refers to the study itself; a study has predictive validity will measure and produce enough information where the researcher can actually predict a behavior (Myers & Hansen, 2006). Content validity is similar to predictive validity, but more specific; content validity measures the methodology of the experiment (Myers & Hansen, 2006). Construct validity refers to the validity of the operational definitions used in the experiment (Myers & Hansen, 2006).
Single Subject Design
According to All Psych Online, the single subject design is likely the result of the research done by B.F. Skinner. Single Subject Design is a type of treatment that involves a single subject or group of subjects. A Single Subject Design is very practical as it acts as its own control group. It is most often used in the field of Behavioral Psychology when measuring behavioral change and/or modification (All Psych, 2008).
In order for a single subject design to be valid it must possess three components; continuous assessment, baseline assessment, sometimes called baseline measure, and variability in data. The baseline assessment being the most important as it contains the pretest information (Wikipedia, 2008).
Single Subject Design in Common Use
The single subject design is used in all forms of Applied Psychology. It can be used on both human and nonhuman subjects. The most common use of the single subject design is in the field of Behavioral Psychology. When trying to help a client change his or her behavior, a psychologist will use the single subject design to measure results.
A very common example of the single subject design in daily life is the parent child relationship. Seeking to modify his child’s behavior, a father may employ a disciplinary technique such as the time out chair. After using this technique on several occasions with the hopes of modifying the child’s behavior, the father may determine that the technique is not working. The father may then seek to modify the child’s behavior by revoking a privilege.
An example of the single subject design applied to a group might occur in an elementary school classroom. A teacher may have a group of students that is misbehaving at recess everyday. She may attempt to separate the group of children every day at recess for a few days in hopes that the behavior will change.
Within-subjects designs are those designs in which each subject participates in more than one part of the experiment (Myers & Hansen, 2006). There are two main problems with these types of designs. One is that the subjects can become tired of the experiment and not do well or, problem number two, the subjects may in fact do very well after repeated trials. This is referred to as the practice effect (Matheson, 2008). I would use this design to test a product’s repeated appeal, for example a cosmetic. Many times women really like a night cream when they first purchase it, however as the newness wears off they grow tired of it.
Between-subjects designs are those designs in which different subjects participate in each part of the experiment (Myers & Hansen, 2006). This type of design provides a good way to avoid the problems that one would find with the within-subjects design, however if there is a shortage of available subjects this may cause a new problem.
A Small N design is an experiment in which just a few subjects participate (Myers & Hansen, 2006). This is a very convenient design to use as it does not require a large number of time, resources and subjects, however it will not render a strong supportable conclusion. This would be a good design to use for a pilot study.
Large N designs are experiments that evaluate the behavior of groups of subjects (Myers & Hansen, 2006). This would be a type of design that a Social Psychologist would use. It could be used to study group behavior as a whole or the effects of an independent variable on different groups of people.
Myers, Anne, & Hansen, Christine. (2006). Experimental psychology. Thomson Wadsworth: Belmont,CA.
All Psych Online. (2008). Single Subject Design.
Wikipedia. (2008). Single Subject Design. http://en.wikipedia.org/wiki/single.subject_design