【文章阅读 – 问卷调查的实践指导】Selecting, designing, and developing your questionnaire
标题 Hands-on guide to questionnaire research Selecting, designing, and developing your questionnaire
关键点 questionnaire research
论文地址
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC420179/pdf/bmj32801312.pdf
一篇2004年的文章,篇幅很短,只有4页,文章的目标是提出一个实用的指南,帮助研究人员进行设计出色,管理良好,公正客观的问卷调查。文章通过问答的形式组织。
What information are you trying to collect?
进行正式的范围界定,这会帮助明确潜在的实践问题,如问卷完成用时;如果对研究所针对的人群不够熟悉,可以先用定性的方法(如 focus-group)来寻找线索。
Is a questionnaire appropriate?
一些时候,一份问卷只有在混合方法中使用才合适,如例如,扩展和量化初始探索阶段的发现。参与问卷调查的人需要给予有效的回答,在physical, mental, social和linguistic等领域甚至需要在专业调查人员的指导下进行。
Examples of research questions for which a questionnaire may not be the most appropriate design
Could you use an existing instrument?
使用已经存在的成功问卷在提高效率的同时还能够和之前的数据进行对比,如果没有现成的问卷,在构建新问卷的时候,使用一种或多种标准工具以及较短的相关问卷,可以省去开发和验证一长串新问题的麻烦。
Is the questionnaire valid and reliable?
一个有效的文件评估它所声称评估的东西,真实情况是有很多的文件做不到这一点:
A self completion questionnaire that seeks to measurepeople’s food intake may be invalid because itmeasures what they say they have eaten, not what theyhave actually eaten.Similarly, responses on question-naires that ask general practitioners how they manageparticular clinical conditions differ significantly fromactual clinical practice.
在不同时间,国家或者文化背景下构建的问卷不一定同样适用自己当下的问题
The item “I often attend gay parties” may have been a valid measure of a person’s sociability level inthe 1950s, but the wording has a very different connotation(含义) today.
可靠的问卷在不同的样本之间得到的结果是一致的,不同的结果取决于不同的参与者个人而不是对问题不同的理解。在一个标准的问卷调查中,所有的参与者都以相同的格式问完全相同的问题,并以统一的方式记录回答。
How should you present your questions?
Pros and cons of open and closed-ended questions
问卷项目可以是开放式的,也可以是封闭式的,以不同的格式呈现。有两个词经常出现于不恰当的问卷中,分别是frequently和regularly。如果询问频率,应该给出具体的频次供选择。
封闭式的问卷可以高效集中收集信息,但是在信息的广度上面不够。封闭式问卷同时可能造成某些问题的遗漏而无法挽回。
尝试在某些问题,某些章节后面加上不限制内容的文本框收集信息是一个好的方法,同样需要基于填写什么内容的相关提示。这样做会给后续分析带来一定挑战(内容整理花时间)。
一些参与者更加倾向于回答"同意",在设计问题的时候,需要注意不要让"强烈认同"代表宽泛的意见。
On a patient satisfaction scale, if one questionis “my GP generally tries to help me out,” anotherquestion should be phrased in the negative, such as“the receptionists are usually impolite.”
Apart from questions, what else should you include?
Checklist for developing a questionnaire
What should the questionnaire look like?
研究人员很少关注自己的问卷的physical layout,但是有研究显示这些细节会影响问卷的回收率。
In general,questions should be short and to the point (around12 words or less), but for issues of a sensitive andpersonal nature, short questions can be perceivedas abrupt and threatening, and longer sentences arepreferred
How should you select your sample?
Types of sampling techniques for questionnaire research
What approvals do you need before you start?
下面给了一个例子
In the United Kingdom, aresearch study on NHS patients or staff must be:
- Formally approved by the relevant person in anorganisation that is registered with the Department of Health as a research sponsor (typically, a research trust,university or college)
- Consistent with data protection law and logged onthe organisation’s data protection files (see next articlein series)
- Accordant with research governance frameworks
- Approved by the appropriate research ethicscommittee