Читать книгу Practitioner's Guide to Using Research for Evidence-Informed Practice - Allen Rubin - Страница 63
3.3.5 Matrix of Research Designs by Research Questions
ОглавлениеAs we have discussed, different research designs are better suited for certain types of EIP questions. As an alternative to presenting multiple research hierarchies, some researchers have represented the fit between various EIP questions and the research designs most equipped to answer those questions with a matrix to emphasize that multiple designs can be used for different types of questions, but that some designs are stronger than others depending on the question type. Table 3.2 is an example of such a matrix that we've created using the four EIP questions and the study designs that we have briefly explored in this chapter.
The larger checks indicate the study designs that are best suited to answer each EIP question, and the smaller checks indicate designs that are less well suited, or less commonly used, to answer each EIP question. Those designs without a large or small check do not provide good evidence to answer the EIP question. For example, when answering questions about factors that predict desirable and undesirable outcomes, correlational studies are the study design of choice and therefore are marked with a large check. Also, systematic reviews that combine the results of these types of studies to answer prognosis and risk questions are also marked with a large check. Sometimes the results of experimental or quasi-experimental studies are used to determine risk and prognosis as well, especially if they are large studies that collect a lot of data about the participants in order to look at factors related to risks and benefits, not just the treatments received. Therefore, each of these designs is marked with a small check. Qualitative studies and single-case designs are neither well suited to answering risk and prognosis questions nor commonly used for this purpose, so these are not marked with a check.
TABLE 3.2 Matrix of Research Designs by Research Questions
Qualitative | Experimental | Quasi-Experimental | Single Case | Correlational | Systematic Reviews or Meta-analyses | |
---|---|---|---|---|---|---|
What factors predict desirable and undesirable outcomes? | √ | √ | √ | √ | ||
What can I learn about clients, service delivery, and targets of intervention from the experiences of others? | √ | √ | √ | |||
What assessment tools should be used? | √ | √ | √ | √ | ||
What intervention, program or policy has the best effects? | √ | √ | √ | √ | √ |
You should keep in mind that this is not an exhaustive list of types of research studies you might encounter and that this table assumes that these study designs are executed with a high level of quality. As you read on in this book, you'll learn a lot more about these and other study designs and how to judge the quality of the research evidence related to specific EIP questions.