Correlational Study

Date: 2015-10-07; view: 669.

Descriptive Studies

Research Designs

Although there are many different kinds of research designs in psychology, studies may be categorized into descriptive or qualitative, correlational, and experimental. The method of data collection also varies, with self-report on one end of the spectrum, and naturalistic observation on the other.

Studies that do not test specific relationships between variables are called descriptive studies. In this research method, general or specific behaviors or attributes are observed and measured, without respect to each other. These studies are generally the design of choice for breaking into new areas, as the vast but often inconclusive amount of information collected can be drawn upon for future hypotheses.

An example of such a study would be a researcher inquiring into the quality of mental health institutions. This would be done by observation or measurements of various criteria, as opposed to relationships between variables. Alternatively, the study could be conducted without any specific criteria in mind.

This method of statistical analysis shows the relationship between two variables. For example, research has shown that alcohol dependence correlates with depression. That is to say, the more alcohol people consume, the more depressed they become. On the other hand, it could be the other way around as well: the more depressed people become, the more likely they are to consume alcohol.

The attributes of correlations include strength and direction. The direction may be positive (both variables both increase or decrease together), negative (one variable increases while the other decreases) or unrelated (a random relationship between variables). The strength of a correlation ranges from -1 to +1 with a 0 reflecting no relationship between variables. A correlational study serves only to describe/predict behavior and not to explain it. This is so because a third variable could be shown to cause the occurrence of one of the variables. Furthermore, only experiments can prove causation.