MEASURES OF ASSOCIATION

Measures of Association

Measures of Association

Introduction

The odds ratio (OR) is a measure of association that is used to describe the relationship between two or more categorical (usually dichotomous) variables (e.g., in a contingency table) or between continuous variables and a categorical outcome variable (e.g., in logistic regression). The OR describes how much more likely an outcome is to occur in one group (i.e. eating alfalfa sprouts) as compared to another group (i.e. disease). ORs is particularly important in research settings that have dichotomous outcome variables (e.g., in medical research).

To determine whether or not this OR is significantly different from 1.0 (implying the observed relationship or effect is most likely not due to chance), one can perform a null hypothesis significance test. Usually, the OR is first transformed into the log odds ratio [log (OR)] by taking its natural logarithm. Then, this value is divided by its standard error and the result compared to a test value.

Is this odds ratio statistically significant?

An odds ratio equal to 1 indicates no association. An odds ratio significantly greater than 1 means that the exposure factor is a risk factor for the disease, while an odds ratio significantly less than 1 means that the exposure factor is a protective factor against disease (Bland, 2000). These results are still an estimated risk of error nearly represented by the level of significance p, traditionally allows to conclude if it is below 5% (p <0.05). In practice, the association is statistically significant; the confidence interval 95% of the odds ratio does not include the value 1. For example, an odds ratio equal to 3.0 with a confidence interval (2.1 to 4.4) indicates a statistically significant association with a three times higher risk of occurrence of disease in exposed subjects compared to unexposed subjects (Strachan & ...