Background: In the light of recent concerns about the reliability of scientific research, the open science movement has attracted considerable attention and interest from a variety of sources, including researchers, research institutions, the business industry, intergovernmental organizations, the media and the public. However, the current extent of openness in demographic research remains unknown. Methods: All relevant publications in four leading journals of anglophone demography -- Demography, Population and Development Review, Population Studies and Demographic Research -- over the last decade are analysed. Using a text search algorithm, two quantitative metrics of open scientific knowledge are estimated: the share of publications that can be openly accessed, and the share of publications providing open software codes for reproducibility or replicability purposes. Results: Two contrasting patterns emerge from these indicators. Access to demographic research papers is increasingly available to everyone, with more than 90\% of Open Access publications in the first ten months of 2023. Conversely, the provision of open software codes is considerably low, ranging from an average of about 2-4\% in three journals to about 29\% in Demographic Research, with only small signs of improvement over the last three years. Contribution: This reflection provides the first assessment of the adoption of some open science practices in demographic research and their evolution over the last decade. An urgent change is needed in the sharing of software codes (along with the data used, where possible) to contribute to the advancement of demographic research. Some recommendations for promoting this change are discussed.

中文翻译：

---

人口研究中的开放科学实践：评估（Ugofilippo Basellini）


背景：鉴于最近人们对科学研究可靠性的担忧，开放科学运动引起了各方的广泛关注和兴趣，包括研究人员、研究机构、商业行业、政府间组织、媒体和公众。然而，目前人口统计研究的开放程度仍然未知。方法：对过去十年中四种英语人口学主要期刊《人口学》、《人口与发展评论》、《人口研究》和《人口统计研究》中的所有相关出版物进行了分析。使用文本搜索算法，估计开放科学知识的两个定量指标：可公开访问的出版物的份额，以及出于可重复性或可复制性目的提供开放软件代码的出版物的份额。结果：这些指标出现了两种截然不同的模式。每个人都可以越来越多地获取人口统计研究论文，到 2023 年前 10 个月，超过 90% 的出版物是开放获取的。相反，开放软件代码的提供量相当低，平均约为 2-4 个三种期刊的比例下降到人口统计研究的 29% 左右，过去三年只有很小的改善迹象。贡献：本反思首次评估了人口研究中一些开放科学实践的采用及其在过去十年中的演变。迫切需要改变软件代码的共享（以及可能的情况下使用的数据），以促进人口研究的进步。讨论了促进这一变革的一些建议。