Hyrkäs Kristiina, Appelqvist-Schmidlechner Kaija, Paunonen-Ilmonen Marita
Faculty of Nursing, University of Alberta, Alberta, Canada.
Scand J Caring Sci. 2003 Dec;17(4):358-64. doi: 10.1046/j.0283-9318.2003.00236.x.
Evaluation research provides new perspectives for clinical supervision (CS), and international collaboration offers advantages to develop valid instruments for this purpose. Besides translation, an instrument developed and tested in another culture requires systematic validation. The study focuses on the translation process of the Manchester Clinical Supervision Scale for testing in Finland carried out collaboratively between the Universities of Tampere and Manchester. The instrument is a 45-item questionnaire with a Likert-type (1-5) scale comprising seven sub-scales: trust and rapport, supervisor advice and support, improved care and skills, importance and value of CS, finding time, personal issues and reflection and total score. At first, a licensed translator translated the instrument into Finnish. A native British language teacher at the University language centre performed the blind back-translation into English. The translations were compared by both collaborative parties and by three experienced Finnish supervisors. A pilot sample (n = 182) was collected to test the translated instrument. In this sample Cronbach's alpha value for the total score was 0.9227 and in the sub-scales 0.6393-0.8838. The mean values in the sub-scales were 14.2-29.3, SDs 3.02-3.88 and modes 14.0-30.0. The British test sample had almost similar values. Translating an instrument into another language not only requires expertise in language, but also in practice. The cultural validation is the most important phase in the process that can be accomplished with pilot testing and statistical methods. However, further expert evaluation is required for the validity of the instrument.
评估研究为临床督导（CS）提供了新的视角，国际合作有助于开发适用于此目的的有效工具。除了翻译之外，在另一种文化中开发和测试的工具还需要进行系统的验证。本研究聚焦于坦佩雷大学和曼彻斯特大学合作开展的将曼彻斯特临床督导量表翻译成芬兰语以便在芬兰进行测试的过程。该工具是一份包含45个条目的问卷，采用李克特式（1 - 5）量表，包括七个子量表：信任与融洽关系、督导建议与支持、护理与技能提升、临床督导的重要性与价值、抽出时间、个人问题与反思以及总分。首先，一名有资质的翻译人员将该工具翻译成芬兰语。大学语言中心的一位以英语为母语的教师进行了盲态回译。双方合作人员以及三位经验丰富的芬兰督导对译文进行了比较。收集了一个试点样本（n = 182）来测试翻译后的工具。在这个样本中，总分的克朗巴哈系数值为0.9227，子量表的系数值在0.6393 - 0.8838之间。子量表的均值为14.2 - 29.3，标准差为3.02 - 3.88，众数为14.0 - 30.0。英国测试样本的数值与之几乎相似。将一种工具翻译成另一种语言不仅需要语言方面的专业知识，还需要实践经验。文化验证是该过程中最重要的阶段，可通过试点测试和统计方法来完成。然而，该工具的有效性还需要进一步的专家评估。
