Predom

Predom is a term used in statistics and data analysis to denote a simple measure of dominance within a categorical dataset. It is commonly described as the predominance proportion of the most frequent category. The exact definition of Predom is not standardized, and its interpretation can vary by field, which means it may be used informally rather than as a formal index.

Calculation: Given N observations across k categories, let f_i be the frequency of category i and p_i = f_i/N the relative frequency. Predom is defined as max_i p_i. It ranges from 1/k (when categories are equally distributed) to 1 (when all observations are in a single category). When expressed as a percentage, Predom = 100 × max_i p_i.

Interpretation and uses: A high Predom indicates strong concentration in a single category; a low Predom indicates a more even distribution. It is used as a quick descriptor in market research, ecology, linguistics, and survey analysis to summarize dominance without detailing the full distribution. It is typically complemented by other measures that capture diversity and evenness.

Limitations: Predom discards information about all but the top category and can be misleading if there are many categories or if sample size is small. It is sensitive to how categories are defined and may be biased by rare categories or aggregation choices. For a fuller picture, researchers often pair Predom with diversity indices such as the Shannon index or Simpson’s index, or with concentration measures like the Herfindahl index.

See also: Dominance, Predominance, Diversity index, Herfindahl index, Shannon index, Simpson index.