Longitudinal dataset of journal data sharing policies across 22 disciplines for 2014 2019 and 2023

This dataset accompanies the Data Descriptor published in Scientific Data. The associated article provides a detailed description of the data collection methods, classification framework, and technical validation.Ikeuchi, U. Longitudinal dataset of journal data sharing policies across 22 disciplines for 2014 2019 and 2023. Sci Data 13, 119 (2026). https://doi.org/10.1038/s41597-025-06434-2This dataset documents journal data sharing policies across 22 disciplines for 2014, 2019, and 2023. A total of 220 high-impact journals were surveyed, representing the top ten journals (by Impact Factor) from each discipline listed in the Essential Science Indicators. For each journal, policies concerning two types of data sharing were reviewed: repository-based data sharing and supplementary materials. Policy requirements were classified into four categories based on their strength: require, recommend, accept, or no policy. Data were collected at three time points through systematic reviews of journal websites and submission guidelines. The dataset includes journal metadata —such as publishers, ISSN, and Impact Factor—along with detailed policy descriptions and classifications. This longitudinal dataset provides evidence of changes in data sharing requirements over time and enables comparative studies of journal policies. These data may be useful for research on open science practices, the development of science policy, and the evolution of scholarly publishing standards.