Lin Li-Chan, Kao Chieh-Chun, Tzeng Yu-Ling, Lin Yeh-Jen
Institute of Clinical Nursing, National Yang-Ming University, Taipei, Taiwan.
J Adv Nurs. 2007 Jul;59(2):178-85. doi: 10.1111/j.1365-2648.2007.04303.x. Epub 2007 May 25.
This paper is a report of a study undertaken to translate the Cohen-Mansfield Agitation Inventory from the original English into a Chinese language version and to establish the content, semantic and technical equivalence of the Chinese version.
Ensuring the equivalence of a translated Chinese version of the Cohen-Mansfield Agitation Inventory for patients with dementia is an essential requirement for identifying culturally specific expressions of agitated behaviours.
In Phase I, five experts scrutinized the Inventory items to determine whether the behaviour described was relevant to the agitated behaviours present in the Taiwanese culture. In Phase II, experts did the initial translation into Chinese and then back-translated it into English. Four bilingual observers rated 31 residents with dementia to determine semantic equivalence across versions. In Phase III, the agitated behaviours of 32 residents were assessed by four observers using staff ratings and direct observation to determine the technical equivalence of the Chinese version. The study was carried out in Taiwan from August 2004 to April 2005.
In Phase I, the rating of behaviours described on the Cohen-Mansfield Agitation Inventory was relevant to the agitated behaviours based on Taiwanese culture with a content validity index (CVI) value of 0.993. In Phase II, intra-class coefficients for all items on the Chinese version and split-halves of Chinese and English versions ranged from 0.69 to 0.74. In Phase III, six possible results for estimation of reliability across methods and raters for the Chinese version ranged from 0.63 to 0.86.
The Chinese Cohen-Mansfield Agitation Inventory had modest psychometric properties. Further studies are needed to assess criterion and conceptual equivalence.
本文报告一项研究，该研究旨在将科恩-曼斯菲尔德激越量表从原始英文翻译成中文版，并确立中文版在内容、语义和技术方面的等效性。
确保科恩-曼斯菲尔德激越量表的中文版对于痴呆患者具有等效性，是识别激越行为中文化特定表达方式的一项基本要求。
在第一阶段，五名专家仔细审查了量表项目，以确定所描述的行为是否与台湾文化中存在的激越行为相关。在第二阶段，专家们先将量表翻译成中文，然后再回译成英文。四名双语观察员对31名痴呆症患者进行评分，以确定不同版本之间的语义等效性。在第三阶段，四名观察员使用工作人员评分和直接观察的方式，对32名患者的激越行为进行评估，以确定中文版的技术等效性。该研究于2004年8月至2005年4月在台湾进行。
在第一阶段，基于台湾文化，科恩-曼斯菲尔德激越量表上所描述行为的评分与激越行为相关，内容效度指数（CVI）值为0.993。在第二阶段，中文版所有项目以及中英文版本的分半的组内相关系数在0.69至0.74之间。在第三阶段，中文版在不同方法和评分者之间可靠性估计的六种可能结果在0.63至0.86之间。
中文版科恩-曼斯菲尔德激越量表具有适度的心理测量学特性。需要进一步研究来评估标准等效性和概念等效性。
