Psychology Research Methods, Essay Example

There are many different research study designs used to conduct research in psychology: the study design chosen likely depends on an investigator’s goals.For example, if there is a new and novel psychological case or syndrome, without recourse to a significant sample sixe, a single case study may be the best option.  A single case study essentially is a qualitative analysis of a patient’s symptoms, conditions, and treatment.  While there is no statistical or formal quantitative analysis offered, the research design is useful in identifying new phenomenon that were not previously studied.  The problem with the single case study design is generalizability; that is, one doesn’t know whether the results from treating one patient can be generalized to other patients who have similar symptoms.

Looking at other research study designs, there is a significant difference  between observational and experimental research designs.  Experimental designs are considered to be the “gold-standard” of research; this is because they control for a single variable in order to find out what effect it has on the dependent variable.  The results in an experimental study are achieved through testing how subjects react to a specified intervention- an intervention is a stimulus given to subjectsin  the treatment group but not given to subjects in the control group  with the results measured.  While the experimental design usually provides the most robust scientific evidence, it is also hard to achieve in many circumstances due to ethical or logistical constraints (Breakwell, 2009).  In that case, researchers may use a quasi-experimental design.  While a quasi-experimental design is also able to give valuable data and light on the independent variable, the main problem is confounding.  Confounding impugns the relationship between the independent and the dependentvariable (internal validity), and may make it difficult to understand the true underlying relationship.

The final types of study are observational in nature.  Observational studies are different from experimental studies in that no explicit treatment is given to subjects.  Rather, observational studies depend on observing patients or collecting data that already exists in the real world (Zechmesiter, 2005).  As a result, observational studies usually involve statistical analysis and are also open to the possibility of confounding because there may be numerous observations.  Finally, correlational studies analyze potential relationships between variables that may indicate a potential negative, positive, or no correlation between the variables examined.  While correlations may show associations between variables, they do not show causality- and thus must be followed up to make robust scientific inferences.

Regarding the first scenario,  a single case study approach should be used in order to record observations regarding the girl. There are several reasons that this is the case.  Primarily, the study design is limited by the sample size in this study; that is , there is only one person included in the study.  The main features of  a single case study include qualitative analysis of certain details in the case that may be of interest to researchers and cannot be captured in data analysis.  For example, the behavior in this case was “spoon banging during mealtime in school.”  In order to measure whether the patient has improved, there was research conducted pre and post intervention that arguably showed improvement in the individual’s overall condition. The case study would be able to give details on how this was possible.

An observational study is more appropriate for the second example, The second scenario looks at how gender (the independent variable) affects the need for affiliation (the dependent variable). In order to do this, the investigator observed students entering the college cafeteria and assessed whether the student was alone or with other students.  Based on observation, females were more likely to be in the presence of another person in the cafeteria.  Due to the nature of this study, it would be improbable to do a single case study- we have a sample size greater than one.  In addition, an experimental or quasi-experimental research design would not be feasible in this setting because it would likely draw attention to what is being examined and could inject bias into the study.

The experimental method is the most appropriate research design for the third scenario.  One can immediately tell that the experimental method is used because the individuals are randomly assigned into two groups: 1) 20 groups of two to use a cooperative approach; 2) 20 groups of two to use a competitive approach.  Randomization is the main methodology used to prevent confounding by ensuring that confounders are equally distributed in individuals in the group.  Once again, a single case study approach will not work due to the sample size.  An observational based approach would not attempt to randomize students into a group; it would simply observe students in groups.  A quasi-experimental method may work in this case, however, a quasi-experimental method does not use randomization.

References

Breakwell, G.M. (2009).  Research Methods in Psychology.  London: Sage Publications.

Zechmesiter, E.B.  (2005).  Research Methods in Psychology.  New York: McGraw-Hill.