Trying to create a schema of sorts to understand research methods, what we are taking to mean the approach used to gather research data. But research methods can be confusing and overwhelming because of the differences in terminology (from our own backgrounds) and even in published explanations of methods.

So we’re gonna try to find a common ground to define some of the most common approaches to research methods across the disciplines/fields related to RCID. Let’s think in terms of  the scientific idea of family, genus, and species, which is used to classify animals. By trying to classify research methods, we can hopefully gain some clarity in. the strengths and weaknesses of them.

For our purpose, we’re thinking of the families of research methods as quantitative (how many?; how often? Causation?), qualitative (how? Why? What purpose?), mixed methods (both a mixture of quantitative and qualitative) but it can also be a mixed qualitative.

A sub-family is empirical (based on observation typically with people involved) and textual (artifacts that examined) and field (which is actually just empirical but with attention to the communities or the in situ practice). All of these families are such because they contain both genus and species or they contain categories and specific types of method.

Our genus methods are case studies, ethnography, user research, and experiments. These are genus methods because they typically use more than one method. Big data as a method is likely here too since it will use different methods based on the methodology chosen.

So our singular methods (or species) that belong to the family of empirical:

• Interviews
• Focus groups
• Surveys
• Questionnaires
• Usability (can potentially be a genus; typically mixed quant/qual)
  • Think aloud
  • Keystroke (or movement)
  • A/B
• Delphi
• experiments ( & quasi-experimental)
• observation
• case study

linguistics analysis (sits between textual and quantitative

Textual

• rhetorical
• literary
• historical
• archival
• thematic
• visual
• multi modal
• meta
• critical

Your method needs to produce the types and kinds of data that you think you need to answer your research question(s). For example, if you really want people’s perceptions of things in a qualitative fashion, then an open-ended questionnaire is not typically the method will produce the type of results that you want. Another example, a survey is a quantitative instrument meant to garner large number of responses so that when the responses are viewed through the sampling plan the results can be generalizable. If this is not your goal or the sampling is not rigorous and produces internal validity, then the survey is likely not the method you need or want for your research question.

When considering the type of method that will be best for your question you also need to consider some practical factors such as

• time and energy
• money
• other available resources
• access to participants and/or sites