How is correlational research different than experimental research?

From PsychWiki - A Collaborative Psychology Wiki

Jump to: navigation, search

There are two general categories under which most empirical research falls: correlational and experimental. In correlational research variables are not manipulated, but are only measured. The purpose is to identify the relationship between the variables. For example, correlation between the set of variables -- such as blood pressure and cholesterol level -- is examined. In experimental research, variables are manipulated, and the effects of this manipulation is measured upon the dependent variable.

In experimental research a treatment is deliberately imposed on a group of objects or participants. In other words, blood pressure can be artificially increased by a researcher; and then cholesterol level is recorded, showing increase or decrease. If changes that are made in variable A lead to changes in variable B, then the conclusion that “A influences B” can be made. Causal relations between variables can only be conclusively demonstrated by experimental data. Data from correlational research cannot prove causality: correlation does not equal causation.

Example / Application

Example: French Replication of Milgram Obedience Study

Application: In the "Milgram" Study, participants were randomly assigned to different conditions, such as the presence or absence of an authority figure. The purpose of the study was to identify if the manipulation influenced the level of obedience in shocking a stranger. Is this correlational or experimental? Given the indendent variable was manipulated, this is an experiment.

Is it possible to study the same variables correlationally? Yes. Instead of manipulating variables, if the researchers only measured variables, it would be correlational. For example, instead of manipulating the presence or absence of an authority figure, the researchers could have asked participants to self-report instances in their life when they felt pressure to conform, and then examine the correlation between the self-report instances of pressure from authority and obedience to that authority.

References

Field, A. (2006). Discovering Statistics Using SPSS: Second Edition. London. Thousand Oaks. New Delhi. Sage Publication

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox