Selecting a Research Method: The Selection Questions
Two slides carry the whole of the source's method-selection material, and a closing sentence then names five dependencies that do not match them. This page prints the questions as given, puts them in an order you can apply, and marks the three that rest on terms the material never defines.
Nine questions on two slides, framed as prompts and not as a checklist
The selection material sits on two consecutive slides titled "Choosing a method (1)" and "Choosing a method (2)". Nothing supports them in prose: the week's study notes document is headed with the week's own title and contains no research design content at all, so the slides are the whole of it. See What Research Design Is.
THE NINE SELECTION QUESTIONS, AS THE SOURCE STATES THEM
| # | Question, quoted from the slide | Slide |
|---|---|---|
| 1 | "Does the hypothesis point you in a certain direction for obtaining sources of data?" | Choosing a method (1) |
| 2 | "Is there any previous research/literature that would indicate an appropriate research method?" | Choosing a method (1) |
| 3 | "Can you access specific data archives, access specific data or people? If not, does this preclude some methods?" | Choosing a method (1) |
| 4 | "Resources - do you have money to survey people, access to people?" | Choosing a method (1) |
| 5 | "Time - Will this research be done within the schedule of your study?" | Choosing a method (1) |
| 6 | "Validity, reliability, trustworthiness - will the research be considered valid, reliable or trustworthy when reviewed?" | Choosing a method (2) |
| 7 | "Generalisability - Consider the ways in which you hope to generalize any specific results from one setting to a larger number of settings?" | Choosing a method (2) |
| 8 | "Ethics - Are there any ethical issues attached to the choice of research method?" | Choosing a method (2) |
| 9 | "Uses/users of the findings - Will this research be useful and add to knowledge?" | Choosing a method (2) |
Five questions on the first slide, four on the second. The numbering above is this page's, added for reference; the source numbers nothing and states no total.
What each question is actually deciding
A question you can ask yourself is not yet a decision. The source supplies the question and stops: no answer scale, no weighting, no order of application, and no rule for what a bad answer obliges you to do.
The "if" side restates the source's question. The "then" side is this page's synthesis, not source content.
From prompt to decision - the "then" column is this page's synthesis, not the source's
Three lists of criteria, and no statement of which one governs
The second slide does not end with the ninth question. It closes with a sentence that names dependencies rather than questions: "The appropriate techniques depend on the type of topic, your skills and experience, availability of access, time factors, ethical matters." That is five items, on the same slide as the four questions above it.
The week's overview slide had already named a third set. Its second sentence imposes three conditions on a design: that the design is appropriate to addressing your questions, that you have the skills and interests to implement the plan, and that all elements of the plan are achievable. The three lists overlap and disagree, and the source never says which governs.
THREE OVERLAPPING ACCOUNTS OF WHAT DRIVES THE CHOICE OF METHOD
| Criterion | In the nine questions | In the five dependencies | In the overview's three conditions |
|---|---|---|---|
| Access to data or people | Yes (q3) | Yes | No |
| Time and schedule | Yes (q5) | Yes | No |
| Ethical matters | Yes (q8) | Yes | No |
| Type of topic | No | Yes | No |
| Your skills and experience | No | Yes | Yes, as "skills and interests" |
| Your interests | No | No | Yes |
| Money and resources | Yes (q4) | No | No |
| Direction given by the hypothesis | Yes (q1) | No | No |
| Prior research and literature | Yes (q2) | No | No |
| Validity, reliability, trustworthiness | Yes (q6) | No | No |
| Generalisability | Yes (q7) | No | No |
| Uses and users of the findings | Yes (q9) | No | No |
| Appropriateness to the research questions | No | No | Yes |
| Achievability of every element of the plan | No | No | Yes |
Compiled from the two "Choosing a method" slides and the week's overview slide. The source presents no such comparison and states no relationship between the three lists.
The first question presumes a hypothesis you may not have
Question 1 asks whether "the hypothesis" points you toward sources of data. The definite article does the damage: the question presupposes that every project has one, and it is the opening move of the whole selection sequence.
The qualitative week of the same material sets out a different starting point. Its notes state that qualitative research questions are "How" or "What" questions - "How did this happen?", "What is going on here?" - and its comparison table opposes quantitative work that "Tests hypotheses that the researcher begins with" to qualitative work that "Captures and discovers meanings once the researcher becomes immersed in the data".
The three questions the material raises and cannot support
Questions 6, 7 and 8 each ask you to judge something against a standard. In each case the standard is named and not supplied. The table records exactly what the material puts behind each term.
WHAT THE MATERIAL SUPPLIES BEHIND THE TERMS IN QUESTIONS 6 TO 8
| Term as the question uses it | What the material supplies | Status |
|---|---|---|
| Validity (q6) | An example in place of a definition. The relevant slide opens "Validity means, for example..." and gives a study of student turnover that "must do that". The general form has to be inferred by the reader. | Example only |
| Reliability (q6) | A clean general definition: "the extent to which a research instrument, such as a survey, will produce the same results if the research was repeated". | Defined |
| Trustworthiness (q6) | Named once, in this question, and nowhere else in the week. | Not defined in source |
| Generalisability (q7) | Named, with the question assuming the reader already knows what it is and asking only about "the ways in which you hope to generalize". | Not defined in source |
| Ethics (q8) | Raised as a selection criterion with no principle, no framework, no approval process and no worked example anywhere in the week. | Not defined in source |
The qualitative week names four alternative criteria for judging qualitative work - credibility, transferability, dependability and confirmability - and never uses the word "trustworthiness" to group them. The source does not connect the two.
Validity and reliability are themselves handled unevenly on the slide that treats them, and that is set out in Validity and Reliability.
Criteria without options: where this page sits against the week's objectives
The research design week states three objectives, quoted exactly: "Create methodology appropriate to proposal"; "Compare types of methods"; "Determine benefits and disadvantages of types". The nine questions are part of what the week does deliver against the first of those.
The deferral is at least recoverable: the seven methodologies are defined in the earlier material and are covered in The Methodology Catalogue. The ten method labels are not recoverable at all; see The Method Catalogue.
Working the nine into a selection you can defend
The source presents the questions in the printed order and states no sequence, priority or weighting. The order below is this page's, chosen so that the questions that can eliminate options are asked before the ones that only shade a judgement.
An order of application (this page's, not the source's)
Eliminate on access and time
Questions 3 and 5. The only two that remove candidates outright, and the cheapest to answer. Anything that fails here is not a method you are choosing between.
Let the question type do its work
Questions 1 and 2. If you have a hypothesis, follow it to the data sources. If you do not, record question 1 as inapplicable and use question 2 - what comparable studies did - as the substantive input instead.
Cost the survivors
Question 4. Money and access to people, in the source's own terms. Put a figure against each surviving candidate, though the material gives you no benchmark for it.
Test how the work will be judged
Question 6. Say in one sentence per candidate how a reviewer would assess it, and cite the text your definitions came from.
Decide what the results are meant to reach
Question 7. Name the settings you hope the findings apply to and the basis for that hope. A method that cannot support the reach you want is the wrong method.
Run the ethics screen before you commit
Question 8. Treat approval as a scheduled activity with a lead time, not a form, and use a published set of principles since the week gives none.
Name the users
Question 9. Two bars - useful, and adds to knowledge. Write one sentence for each. If you cannot, the problem is upstream of method selection.
Record the criteria you added
Cost, your skills, your interest and the type of topic appear in one or another of the source's three lists but not in the nine. If you used them, say which list they came from.
What to carry forward
- Nine questions, presented as prompts for initial consideration. The source states no count, no order and no weighting, and does not call the set a checklist.
- The same slide closes with five dependencies that do not match the nine, and the overview slide adds three conditions that match neither. Choose which list you are working to, and say so.
- Question 1 presumes a hypothesis. If you have none, mark it inapplicable rather than forcing an answer.
- Trustworthiness, generalisability and the ethics of a method are asked about and never defined. Bring those definitions from a cited text.
- These are criteria without options. Pair them with the methodology catalogue from the earlier material, because the week's own method grid supplies nothing to apply them to.
Frequently asked questions
Does the source require me to answer all nine questions?
No. It introduces them as questions to ask yourself in your "initial consideration", states no count, numbers nothing and sets no completion rule. Answering all nine is good practice and is this page's recommendation, but presenting the set as a required framework would misrepresent the material.
Which list should I work to - the nine questions or the five dependencies?
The source does not say, and both appear on the same slide. In practice, use the nine as the working set because they are the more specific, then add the two dependencies the nine omit - type of topic, and your skills and experience - and record that you did. Being explicit about which criteria you applied is more defensible than silently blending the lists.
What do I do about question 1 if my project has no hypothesis?
Record it as inapplicable and say why. The qualitative material in the same source describes research that begins from "How" or "What" questions and discovers meaning rather than testing a hypothesis stated in advance. The source never reconciles that with question 1, so the honest move is to note the mismatch rather than invent a hypothesis to satisfy the prompt.
Where can I get a definition of trustworthiness?
Not from this material - it uses the word once and defines it nowhere. The qualitative week offers four alternative criteria for judging qualitative work, credibility, transferability, dependability and confirmability, but never groups them under trustworthiness. Take a definition from a methods text you have read and cite it.
Do the nine questions tell me which method to choose?
No. They are criteria, and the week never describes the options the criteria would be applied to: the ten methods it lists are bare labels and the seven methodologies are deferred to an earlier topic. You will need the methodology definitions from that earlier material, plus method definitions from a cited text, before the questions can discriminate between anything.
Is the ethics question answered anywhere in the week?
No. Ethics is raised as a method-selection criterion and no principles, framework, approval process or example follows it. The qualitative week raises ethics twice more, once as a checklist item and once through case studies, and again supplies no framework against which either could be judged.
References and source attribution
- O'Leary, Z. 2010, The essential guide to doing your research project, Sage Publications - the only work referenced by the research design week's slide deck.
- O'Leary, Z. 2017, The Essential Guide to Doing Your Research Project, 3rd ed., Sage Publications, London.
- Bryman, A. 2016, Social Research Methods, 5th ed., Oxford University Press, Oxford.
- Quinlan, C. 2011, Business Research Methods, 1st ed., Cengage Publishing.
- The supplied teaching source: the research design week's slide deck, slides 3 and 9 to 13, and the accompanying study notes document, used here as the basis for every statement attributed to "the source".
Suggested questions for Ask KEVOS
- Turn the nine selection questions into a one-page method justification for my project.
- Which of the nine questions can rule a method out on their own, and which only shade the judgement?
- My project has no hypothesis - how should I handle selection question 1 in my proposal?
- Give me a cited definition of generalisability, since the teaching material does not define it.
- Draft the ethics screening answer for question 8 for an interview study inside my own organisation.
- Compare the nine selection questions with the five dependencies and tell me what each list misses.