Volume 127, December 2025, 102954Author links open overlay panel, , , Highlights•

Interval-valued data allow to capture richer information in psychological research.

•

Some indices of Cohen’s d family of standardized mean difference effect sizes are stated for these data.

•

A distance-based approach is followed to extend such measures.

•

The applicability of these indices is illustrated through a real-life example.

Abstract

There is a large literature in psychological and behavioral sciences describing mean difference effect size indices for real-valued data. These indices are essential for integrating results from different studies, diverse types of data, or various rating scales. The emergence of new types and sources of data motivates the need to adapt the existing effect size measures or to develop new ones in order to facilitate the comparison of the observed experimental outcomes. To this purpose, some indices of the Cohen d family are to be extended throughout this article in order to deal with interval-valued data by following a distance-based approach and its utility will be illustrated by means of a real-life example where interval-valued responses were collected in a questionnaire.

Keywords

Effect size

Standardized mean difference

Cohen’s d index

Glass’ Δ index

Interval-valued data

Interval-valued rating scales

Data availabilityThe data and the source code that support the findings of this study are openly available in RUO (Repositorio Institucional de la Universidad de Oviedo) at https://hdl.handle.net/10651/80598.

© 2025 The Authors. Published by Elsevier Inc.