当前位置: X-MOL 学术Psychology and Aging › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Data overuse in aging research: Emerging issues and potential solutions.
Psychology and Aging ( IF 4.201 ) Pub Date : 2021-04-29 , DOI: 10.1037/pag0000605
Daniel K Mroczek 1 , Sara J Weston 2 , Eileen K Graham 1 , Emily C Willroth 1
Affiliation  

Aging and lifespan development researchers have been fortunate to have public access to many longitudinal datasets. These data are valuable and see high utilization, yet this has a considerable downside. Many of these are heavily overused. Overuse of publicly available datasets creates dependency among published research papers giving the false impression of independent contributions to knowledge by reporting the same associations over multiple papers. This is a potentially serious problem in the aging literature given the high use of a relatively small number of well-known studies. Any irregularities or sampling biases in this relatively small number of samples have outsize influence on perceived answers to key aging questions. We detail this problem, focusing on issues of dependency among studies, sampling bias and overfitting, and contradictory estimates of the same effect from the same data in independent publications. We provide solutions, including greater use of data sharing, pre-registrations, holdout samples, split-sample cross-validation, and coordinated analysis. We argue these valuable datasets are public resources that are being diminished by overuse, with parallels in environmental science. Taking a conservation perspective, we hold that these practices (pre-registration, holdout samples) can preserve data resources for future generations of researchers. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

中文翻译:

老化研究中的数据过度使用:新出现的问题和潜在的解决方案。

老龄化和寿命发展研究人员很幸运能够公开访问许多纵向数据集。这些数据很有价值,并且利用率很高,但缺点是很大。其中许多被严重滥用。过度使用公开可用的数据集会在已发表的研究论文中产生依赖性,从而通过在多篇论文中报告相同的关联来给知识的独立贡献带来错误的印象。考虑到大量使用相对较少的知名研究,这在老化的文献中是一个潜在的严重问题。在相对较少的样本中,任何不规则或抽样偏差都会对关键的老化问题的感知答案产生巨大影响。我们将详细介绍这个问题,重点是研究之间的依赖关系,抽样偏差和过度拟合,在独立出版物中,根据相同数据得出的相同效果的相互矛盾的估计。我们提供解决方案,包括更多地使用数据共享,预注册,保留样本,拆分样本交叉验证和协调分析。我们认为这些有价值的数据集是公共资源,但由于过度使用而减少了,与环境科学类似。从保护的角度来看,我们认为这些做法(预注册,保留样本)可以为后代研究人员保留数据资源。(PsycInfo数据库记录(c)2021 APA,保留所有权利)。我们认为这些有价值的数据集是公共资源,但由于过度使用而减少了,与环境科学类似。从保护的角度来看,我们认为这些做法(预注册,保留样本)可以为后代研究人员保留数据资源。(PsycInfo数据库记录(c)2021 APA,保留所有权利)。我们认为这些有价值的数据集是公共资源,但由于过度使用而减少了,与环境科学类似。从保护的角度来看,我们认为这些做法(预注册,保留样本)可以为后代研究人员保留数据资源。(PsycInfo数据库记录(c)2021 APA,保留所有权利)。
更新日期:2021-04-29
down
wechat
bug