基于鲁棒标准误差的小簇二值结果分析

IF 1.7 4区教育学 Q2 EDUCATION & EDUCATIONAL RESEARCH

Journal of Research on Educational Effectiveness Pub Date : 2022-08-08 DOI:10.1080/19345747.2022.2100301

Francis L. Huang, Bixiu Zhang, Xintong Li

{"title":"基于鲁棒标准误差的小簇二值结果分析","authors":"Francis L. Huang, Bixiu Zhang, Xintong Li","doi":"10.1080/19345747.2022.2100301","DOIUrl":null,"url":null,"abstract":"Abstract Binary outcomes are often analyzed in cluster randomized trials (CRTs) using logistic regression and cluster robust standard errors (CRSEs) are routinely used to account for the dependent nature of nested data in such models. However, CRSEs can be problematic when the number of clusters is low (e.g., < 50) and, with CRTs, a low number of clusters is quite common. We investigate the use of the CR2 CRSE and an empirical degrees of freedom adjustment (dofBM) proposed by Bell and McCaffrey with a simulation using binary outcomes and illustrate its use with an applied example. Findings show that the CR2 (w/dofBM) standard errors are relatively unbiased with coverage and power rates for group-level predictors that are comparable to that of a multilevel logistic regression model and can be used even with as few as 10 clusters. To promote its use, a free graphical SPSS extension is provided that can fit logistic (and linear) regression models with a variety of CRSEs and dof adjustments.","PeriodicalId":47260,"journal":{"name":"Journal of Research on Educational Effectiveness","volume":null,"pages":null},"PeriodicalIF":1.7000,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Using Robust Standard Errors for the Analysis of Binary Outcomes with a Small Number of Clusters\",\"authors\":\"Francis L. Huang, Bixiu Zhang, Xintong Li\",\"doi\":\"10.1080/19345747.2022.2100301\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Binary outcomes are often analyzed in cluster randomized trials (CRTs) using logistic regression and cluster robust standard errors (CRSEs) are routinely used to account for the dependent nature of nested data in such models. However, CRSEs can be problematic when the number of clusters is low (e.g., < 50) and, with CRTs, a low number of clusters is quite common. We investigate the use of the CR2 CRSE and an empirical degrees of freedom adjustment (dofBM) proposed by Bell and McCaffrey with a simulation using binary outcomes and illustrate its use with an applied example. Findings show that the CR2 (w/dofBM) standard errors are relatively unbiased with coverage and power rates for group-level predictors that are comparable to that of a multilevel logistic regression model and can be used even with as few as 10 clusters. To promote its use, a free graphical SPSS extension is provided that can fit logistic (and linear) regression models with a variety of CRSEs and dof adjustments.\",\"PeriodicalId\":47260,\"journal\":{\"name\":\"Journal of Research on Educational Effectiveness\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2022-08-08\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Research on Educational Effectiveness\",\"FirstCategoryId\":\"95\",\"ListUrlMain\":\"https://doi.org/10.1080/19345747.2022.2100301\",\"RegionNum\":4,\"RegionCategory\":\"教育学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"EDUCATION & EDUCATIONAL RESEARCH\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Research on Educational Effectiveness","FirstCategoryId":"95","ListUrlMain":"https://doi.org/10.1080/19345747.2022.2100301","RegionNum":4,"RegionCategory":"教育学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"EDUCATION & EDUCATIONAL RESEARCH","Score":null,"Total":0}

引用次数: 2

摘要

在聚类随机试验(CRTs)中，二元结果通常使用逻辑回归分析，聚类稳健标准误差(CRSEs)通常用于解释此类模型中嵌套数据的依赖性。然而，当集群数量较少时(例如，< 50)，crse可能会出现问题，而对于crt，集群数量较少是很常见的。我们研究了由Bell和McCaffrey提出的CR2 CRSE和经验自由度平差(dofBM)的使用，并使用二进制结果进行了模拟，并通过一个应用实例说明了它的使用。研究结果表明，与多水平逻辑回归模型相比，CR2 (w/dofBM)标准误差与组级预测因子的覆盖率和功率率相对无偏，即使只有10个聚类也可以使用。为了促进其使用，提供了一个免费的图形SPSS扩展，可以适合逻辑(和线性)回归模型与各种CRSEs和dof调整。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Using Robust Standard Errors for the Analysis of Binary Outcomes with a Small Number of Clusters

Abstract Binary outcomes are often analyzed in cluster randomized trials (CRTs) using logistic regression and cluster robust standard errors (CRSEs) are routinely used to account for the dependent nature of nested data in such models. However, CRSEs can be problematic when the number of clusters is low (e.g., < 50) and, with CRTs, a low number of clusters is quite common. We investigate the use of the CR2 CRSE and an empirical degrees of freedom adjustment (dofBM) proposed by Bell and McCaffrey with a simulation using binary outcomes and illustrate its use with an applied example. Findings show that the CR2 (w/dofBM) standard errors are relatively unbiased with coverage and power rates for group-level predictors that are comparable to that of a multilevel logistic regression model and can be used even with as few as 10 clusters. To promote its use, a free graphical SPSS extension is provided that can fit logistic (and linear) regression models with a variety of CRSEs and dof adjustments.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Research on Educational Effectiveness EDUCATION & EDUCATIONAL RESEARCH-

CiteScore

4.00

自引率

11.10%

发文量

期刊介绍： As the flagship publication for the Society for Research on Educational Effectiveness, the Journal of Research on Educational Effectiveness (JREE) publishes original articles from the multidisciplinary community of researchers who are committed to applying principles of scientific inquiry to the study of educational problems. Articles published in JREE should advance our knowledge of factors important for educational success and/or improve our ability to conduct further disciplined studies of pressing educational problems. JREE welcomes manuscripts that fit into one of the following categories: (1) intervention, evaluation, and policy studies; (2) theory, contexts, and mechanisms; and (3) methodological studies. The first category includes studies that focus on process and implementation and seek to demonstrate causal claims in educational research. The second category includes meta-analyses and syntheses, descriptive studies that illuminate educational conditions and contexts, and studies that rigorously investigate education processes and mechanism. The third category includes studies that advance our understanding of theoretical and technical features of measurement and research design and describe advances in data analysis and data modeling. To establish a stronger connection between scientific evidence and educational practice, studies submitted to JREE should focus on pressing problems found in classrooms and schools. Studies that help advance our understanding and demonstrate effectiveness related to challenges in reading, mathematics education, and science education are especially welcome as are studies related to cognitive functions, social processes, organizational factors, and cultural features that mediate and/or moderate critical educational outcomes. On occasion, invited responses to JREE articles and rejoinders to those responses will be included in an issue.