Harran University.
J Gen Psychol. 2021 Jan-Mar;148(1):84-104. doi: 10.1080/00221309.2020.1834346. Epub 2020 Oct 24.
The purpose of the study was to psychometrically characterize the Ethical Leadership Scale (ELS) through confirmatory factor analysis (CFA) and Rasch analysis. Data were collected from 469 teachers across several countries. The one-factor structure of the ELS was confirmed in CFA. Rating scale diagnosis, item fit assessment, reliability analysis, unidimensionality, local independence, and differential item functioning (DIF) analyses were conducted using the Rasch Rating Scale Model. Item and rating scale analyses indicated that the ten items and five-category scale performed as intended. High person separation and reliability statistics supported the consistency of the ELS scores. Only one item (Item 4) demonstrated misfit in the Rasch model and two items (i.e., Items 2 and 10) showed DIF across Turkish and European respondents. None of the items showed DIF across gender. Overall CFA and Rasch models provided evidence for the psychometric quality of the 10-item ELS. However, reassessment of the scale after removing Item 4 showed improvement in the psychometric properties of the scale.
本研究旨在通过验证性因子分析（CFA）和 Rasch 分析对道德领导力量表（ELS）进行心理计量学特征描述。数据来自于来自多个国家的 469 名教师。在 CFA 中，ELS 的单因素结构得到了确认。使用 Rasch 评分量表模型进行了评分量表诊断、项目拟合评估、可靠性分析、单维性、局部独立性和项目区分功能（DIF）分析。项目和评分量表分析表明，十个项目和五个类别的评分量表按预期运行。高的个体分离和可靠性统计支持 ELS 分数的一致性。只有一个项目（项目 4）在 Rasch 模型中表现出不拟合，两个项目（即项目 2 和 10）在土耳其和欧洲受访者中表现出 DIF。没有一个项目在性别上表现出 DIF。总的来说，CFA 和 Rasch 模型为 10 项 ELS 的心理计量质量提供了证据。然而，在删除项目 4 后对量表进行重新评估显示，量表的心理计量特性得到了改善。
