定性研究实用指南：背景、研究问题和设计

Full text

Is it normal that my research question seems to change during the study?

During the research process, the research question might change to a certain degree because data collection and analysis sharpens the researcher’s lenses. Data collection and analysis are iterative processes that happen simultaneously as the research progresses. This might lead to a somewhat different focus of your research question and to additional questions. However, you cannot radically change your research question because that would mean you were performing a different study. In the methods section, you need to describe how and explain why the original research question was changed.

For example, let us return to the problem that GPs are hesitant to ask about intimate partner violence despite their awareness of family violence. To design a qualitative study, you might use SPIDER to support you in formulating your research question. You purposefully sample GPs, varying in age, gender, years of experience and type of practice (S-1). You might also decide to sample patients, in a variety of life situations, who have been faced with the problem (S-2). You clarify the phenomenon of family violence, which might be broadly defined when you design your study—e.g. family abuse and violence (PI-1). However, as your study evolves you might feel the need for fine-tuning—e.g. asking about intimate partner violence (PI-2). You describe the design, for instance, a phenomenological study using interviews (D), as well as the ‘think, feel or do’ elements you want to evaluate in your qualitative research. Depending on what is already known and the aim of your research, you might choose to describe actual behaviour and experiences (E-1) or explore attitudes and perspectives (E-2). Then, as your study progresses, you also might want to explain communication and follow-up processes (E-3) in your qualitative research (R).

Each of your choices will be a trade-off between the intended variety, depth and richness of your findings and the required samples, methods, techniques and efforts for data collection and analyses. These choices lead to different research questions, for example:

‘What are GPs’ and patients’ attitudes and perspectives towards discussing family abuse and violence?’ Or:

‘How do GPs behave during the communication and follow-up process when a patient’s signals suggest intimate partner violence?’

全文翻译（仅供参考）

在研究过程中，研究问题可能会发生一定程度的变化，因为数据收集和分析使研究者的视角更加敏锐。数据收集和分析是反复的过程，随着研究的进展同时发生。这可能会导致你的研究问题的重点有些不同，并导致更多的问题。但是，你不能从根本上改变你的研究问题，因为那将意味着你在进行一项不同的研究。在方法部分，你需要描述如何改变原来的研究问题，并解释其原因。

例如，让我们回到这样一个问题：尽管全科医生意识到家庭暴力，但他们对询问亲密伴侣的暴力问题犹豫不决。为了设计一个定性研究，你可以用SPIDER来支持你制定你的研究问题。你有目的地对全科医生进行抽样调查，在年龄、性别、工作年限和诊所类型方面各有不同（S-1）。你也可以决定对处于不同生活环境中的病人进行抽样调查，他们都曾面临过这个问题（S-2）。你澄清了家庭暴力现象，在你设计研究时，可能有广泛的定义--例如，家庭虐待和暴力（PI-1）。然而，随着研究的发展，你可能觉得有必要进行微调--例如，询问亲密伴侣的暴力（PI-2）。你描述设计，例如，使用访谈的现象学研究（D），以及你想在定性研究中评估的 "想、感觉或做 "的要素。根据已经知道的情况和你的研究目的，你可能会选择描述实际行为和经验（E-1）或探索态度和观点（E-2）。然后，随着研究的深入，你也可能想在你的定性研究中解释沟通和后续过程（E-3）。

你的每一个选择都是在你的研究结果的预期多样性、深度和丰富性与数据收集和分析所需的样本、方法、技术和努力之间进行权衡。这些选择导致了不同的研究问题，例如。

全科医生和病人对讨论家庭虐待和暴力的态度和观点是什么？或者

当病人的信号表明有亲密伴侣暴力时，全科医生在沟通和跟进过程中是如何表现的？

THE

END

打赏