1. (*noun*) In statistical analysis, a broad term describing how a change in one variable is associated with a similar pattern of variation in another variable across cases in a dataset.

2. (*noun*) A normalized indicator of covariance.

**Example:** In infants, shoe size is positively correlated with the number of teeth

**Audio Pronunciation: **(cor·re·la·tion)

**Download Audio Pronunciation:** correlation.mp3

**Usage Notes:**

- Plural:
**correlations** - A correlation can be positive or negative. When variables move in the same direction they are positively correlated and when an increase in one variable, decreases another variable they are negatively correlated.
- Perfect correlation is unlikely in the social sciences.
- Correlation does not always equal causation. Inappropriate inference of causality is referred to as a spurious relationship (not to be confused with spurious correlation). Correlation only reveals a relationship between variables but not the context; the presence of a third factor that accounts for the association between variables is a confounding variable. Researchers must employ a rigorous methodology in order to adequately infer the nature and direction of correlated variables. For example, the positive correlation between hat sales and ice cream sales is likely not because the wearing of hats promotes a craving for ice cream but because of a third, confounding variable: hot weather.
- Correlation between two variables is measured statistically by a correlation coefficient (coefficient of association). There are different tests of correlation depending on the type of data and the characteristic of relationship being examined. A common test is the Pearson product-movement correlation (commonly referred to as Pearson correlation,
*r*) which tests the linear correlation between two variables where data approximates interval level characteristics. When data does not meet the assumptions for the Pearson correlation test, nonparametric tests for rank correlation may be applied. - Types:
- Also called
**correlativity**. - When two variables (
*verb*)**correlate**they are (*adverb*)**correlationally**or (*adverb*)connected and (**correlatively***adjective*)**correlational**or (*adjective*)**correlative**and (*adjective*)**correlational****analysis**determines the strength of two variables using (*n*).**correlational statistics**.

**Additional Information:**

- Abelson, Robert P. 1995.
*Statistics as Principled Argument*. Hillsdale, NJ: Lawrence Erlbaum Associates. - Agresti, Alan, and Barbara Finlay. 2009.
*Statistical Methods for the Social Sciences*. Upper Saddle River, NJ: Pearson Prentice Hall. - Babbie, Earl. 2013.
*The Practice of Social Research*. 13th ed. Belmont, CA: Wadsworth Cengage Learning. - Cole, Stephen. 1972.
*The Sociological Method*. Chicago: Markham Publishing Company. - Huff, Darrell, and Irving Geis. 1954.
*How to Lie with Statistics*. New York: Norton. - Kenny, David A. 1979.
*Correlation and Causality*. New York: Wiley - Krieg, Eric J. 2012.
*Statistics and Data Analysis for Social Science*. Boston: Allyn & Bacon. - Marsh, Catherine, and Jane Elliott. 2008.
. 2nd ed. Cambridge: Polity Press.*Exploring Data: An Introduction to Data Analysis for Social Scientists* - Paulos, John Allen. 1995.
*A Mathematician Reads the Newspaper*. New York: Anchor Books. - Rosenberg, Morris. 1968.
*The Logic of Survey Analysis*. New York: Basic Books. - Singleton, Royce, and Bruce C. Straits. 2010.
*Approaches to Social Research*. New York: Oxford University Press. - Vogt, W. Paul. 2011.
*Dictionary of Statistics and Methodology: A Nontechnical Guide for the Social Sciences*. London: SAGE.

**Related Terms: **