Skip to main content icon/video/no-internet

Odds ratios (OR) compare the probability of an outcome occurring to absence when exposed to treatment/dependent variables. The OR provides effect size information on the relative odds of two groups. Common uses of OR include medical research and cross-sectional studies. The ratio is easy to calculate and easy to interpret when written appropriately. Probability and odds build the foundation of the OR calculation. Probability describes the chance of one specified outcome happening as a percentage of the total outcomes. Odds represent the ratio of the probability of an event occurring divided by the probability of the event not occurring. In addition to presenting an outcome, OR foundationally contributes to logistical regression, enabling researchers to model categorical variables. This entry presents an overview of OR by explaining and discussing the calculation in three steps: (a) probability, (b) odds, and (c) OR. Next, a hypothetical example is used to demonstrate a practical application. Then, the entry describes how to interpret results. Finally, the entry explores the relationship of OR and logistical regression analysis.

Odds Ratio Calculation

The OR expresses the association of the odds of two groups as a ratio. Calculating the OR involves three steps. First, for each group, compute the probability of outcome occurrence and nonoccurrence as a proportion of the total outcome. Second, for each group, divide the two probabilities to calculate the odds. Third, divide the two groups’ odds to calculate the OR. This section summarizes each step, presents a hypothetical example, and discusses how to interpret results.

Step 1: Calculate the Probabilities

For each group, calculate two probabilities: probability of outcome occurrence and outcome nonoccurrence. Outcome occurrence represents the number of times a desired outcome, for example, low anxiety in public speaking, happens within a sample. Outcome nonoccurrence represents the number of times an outcome does not happen within a sample. These numbers are complementary; occurrence outcomes plus nonoccurrence outcomes must equal total outcomes. In addition, occurrence and nonoccurrence outcomes must be mutually exclusive, meaning that each recorded outcome fits in one, and only one, category, preventing double counting and/or data omission.

Probability and odds represent two different, but related, statistics. Generally speaking, probability (P) indicates the number (n) of occurrence (1) or nonoccurrence (2), as a percentage of total occurrences, denoted as a number between, or equal to, 0 and 1. Greater numbers reflect greater chances of a given outcome. Mathematically, the following formulas reflect probability calculations and a check figure.

Probability ( P ) of occurrence ( 1 ) : P 1 = n 1 n . Probability ( P ) of nonoccurrence ( 2 ) : P 2 = n 2 n . Check: n 1 + n 2 = n .

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading