Correlation as a Measure of Association

Christina Johnson

BSHS/435

3/18/2015
Dr. Michele Howser

Correlation as a Measure of Association

When performing academic research, a researcher must determine not only what it is they are studying but how they are going to describe what they are studying. What a person is studying can be defined as a characteristic or any qualities that possess two or more potential values (Leedy & Ormrod, 2010). Variables usually come in two forms, independent and dependent where the former is manipulated by the researcher and the latter is influenced by the first (Leedy & Ormrod, 2010). Using a statistical process, determining if two or more variables are associated to each other defines correlation (Leedy & Ormrod, 2010). Describing the connection between the variables is how association is measured. Knowledge of correlation as a measure of association, the methods used, and its application in research will assist a researcher in understanding when and how to establish correlation.
Correlational Research Studies using correlation are done to observe to what degree different types of variances exists with one variable in relation to another variable (Leedy & Ormrod, 2010). Basically, it is to determine if one characteristic affects another. “A correlation exists if, when one variable increases, another variable either increases or decreases in a somewhat predictable fashion” (Leedy & Ormrod, 2010). When data is collected for this type of research, a researcher might use a chart or diagram to help better determine if there is in fact a correlation between the variables (Leedy & Ormrod, 2010). In example, a researcher charting a study using the variables of children’s age and belief in the tooth fairy, may find that as children get older the belief in the mythical creature fades. If this was the case the chart would show that while the independent variable of the children’s ages increase, the dependent variable of believing in the tooth fairy decreases. Even though data can prove to show a strong relationship between two variables, correlation is not meant to assume that one variable causes the other; this type of study merely shows the relationship that may or may not exists (McLeod, 2008).
Measures of Association The strength of the relationships between the variables studied is referred to as the measure of association (Monette et al., 2011). Since seeing a predictable change in both variables constitutes a correlation, seeing how they change is now the key. There are different types of relationships data can show for variables. These relationships are; positive, negative, and curvilinear. Positive relationships shows the value of both variables going in the same direction, negative relationships show variable values going in opposite directions, and curvilinear data do not show consistent changes between variables.
Measures of Association: Correlation When measuring how closely variables are associated, statistics are used to reference the correlation (Leedy & Ormrod, 2010). “A correlation coefficient is a number between negative one and positive one; most correlation coefficients are decimals (either positive or negative) somewhere between these two extremes” (Leedy & Ormrod, 2010). By this definition the connection of the correlation is a very small one, between three numbers. The positive or negative coefficient dictates the direction of the correlation used to show the relationship of variables in a study (Leedy & Ormrod, 2010). The measurement of association is only determined after the instrument used to measure the relationship is decided. The instrument is the tool which by proper implementation, the study can be replicated by a researcher using the same measurements and variables (Monette et al., 2011). With the proper measurement tools a researcher can conduct a correlational study and have accurate statistical data to prove the relationship.

References
Leedy, P. D., & Ormrod, J. E. (2010). Practical Research Planning and Design (9th ed.). Upper Saddle River, New Jersey: Pearson Education Inc.
McLeod, S. A. (2008). Correlation. Retrieved from http://www.simplypsychology.org/correlation.html
Monette, D. R., Sullivan, T. J., & DeJong, C. R. (2011). Applied Social Research: A Tool for the Human Services (8th ed.). Mason, Ohio: Cengage Learning.