MEDD8001 Chap.2 Research as Scientific Inquiry
Research as Scientific Inquiry
Three sources of knowledge, and why the distinction is practical
The second lecture treats research as a knowledge-producing activity and begins by asking how anyone knows anything. Deductive reasoning moves from general to specific through logical argument. Inductive reasoning generalises from observed examples to a class.
Scientific method uses both, developing a hypothesis inductively and then putting it up for verification. The distinction is not decorative: a proposal that promises to test a hypothesis and then only describes cases has promised deduction and delivered induction, and a marker will notice in the analysis plan.
The three legs of a research problem
This is the most reusable idea in the course.
A research problem has three legs and each is answered by different evidence. What happens is the phenomenon, which can be established from the literature or described from data. So what is the impact or action that follows, and it needs analysis of empirical data. Why and how are the causes, and they need analysis of empirical data too.
A question made entirely of causes has nothing to attach a cause to; a question made entirely of phenomenon gives a reader no reason to care.
The lecture demonstrates the conversion on a running example about teachers' curricular beliefs, moving from what beliefs teachers hold, through how those beliefs show up in practice, to what shaped them, and only then producing a directional hypothesis.
Seven stages, and where a proposal sits in them
The research process is taught as a sequence: select the problem, review the literature, design the study, collect the data, analyse the data, interpret the data, report the findings.
A proposal is written before any of the later stages exist, which is why it is judged on the first three. The five components each have a stated job. The introduction, background and rationale identify the problem and justify selecting it, particularly against valid alternative designs.
The literature review synthesises published work associated with it. The research questions or hypotheses specify what is central.
The data collection plan describes what information is necessary for an adequate test and how it will be obtained.
The data analysis plan describes the methods that will answer the questions or decide whether the hypotheses hold.
The vocabulary the rest of the course assumes
A construct is an abstraction adopted for a research purpose and needs two definitions: a constitutive one saying what it means and an operational one saying how it will be measured.
A variable stands in for a construct and takes values, coming in continuous, two-category and many-category forms. A constant is a fixed value. The dependent variable is the phenomenon to be explained and the independent variable is what explains it, which maps them directly onto the first and third legs of the problem structure.
A hypothesis is a statement to be tested, and it is directional when it predicts which way the difference goes.
Design families, and what the data looks like
The lecture splits empirical work by the kind of data before it splits by design. Quantitative work uses numeric data such as performance scores, questionnaire responses and statistics collected in natural settings.
Qualitative work uses non-numeric data such as interviews, observation, documents, images and artefacts.
Mixed methods combines them.
The design list follows: experimental, non-experimental with a probability test and cross-sectional designs on the quantitative side; field research including interviews, ethnography and grounded theory, action research, and historical or content analysis on the qualitative side; with longitudinal designs and case studies available to either.
The lecture closes by putting six questions on screen and asking which design each one needs, and the reliable tell is the verb.
What this chapter covers
- 01
Deductive, inductive and scientific reasoning
- 02
Theoretical or philosophical research against empirical research
- 03
What happens, so what, why and how: the three legs
- 04
Turning the legs into questions, and one of them into a hypothesis
- 05
Seven stages of the research process
- 06
Five proposal components and the job each one does
- 07
Construct, constitutive and operational definition
- 08
Continuous, dichotomous and categorical variables
- 09
Directional against non-directional hypotheses
- 10
Design families, and reading the verb in the question
Six questions, six designs, one rule for choosing
- 4Assign a design family to each of the six questions.
- 3State the rule that produced the assignment.
- 2Name the one question whose design depends on what access you have.
Key terms
- Deductive Reasoning
- Moving from general knowledge to a specific conclusion through logical argument, which is the direction a hypothesis test runs in.
- Inductive Reasoning
- Generalising from observed examples to a whole class, which is the direction most qualitative analysis runs in.
- Operational Definition
- The concrete statement of how a construct will be measured, as distinct from the constitutive definition that says what the construct means.
- Dependent Variable
- The construct or phenomenon a study sets out to explain, which corresponds to the what-happens leg of a research problem.
- Independent Variable
- The construct or characteristic offered as an explanation of the dependent variable, corresponding to the causes leg.
- Directional Hypothesis
- A statement to be tested that predicts not only a difference but which way the difference will go, and which can therefore be wrong in a specific way.
- Cross-Sectional Design
- A design that observes many cases at one point in time, which can describe differences between groups but cannot by itself demonstrate change.
Research as Scientific Inquiry FAQ
What is the difference between a research question and a hypothesis?
A research question can be open and invites a complex answer. A hypothesis is a statement to be tested, so it has to be capable of turning out false, and a directional hypothesis has to be capable of turning out false in a specific direction. The test to apply to your own is short: write down the result that would make you withdraw it.
If nothing comes to mind, what you have written is a topic statement wearing the grammar of a hypothesis.
My question is about causes. Why does the course want me to describe first?
Because a cause needs something to be the cause of. The three-leg structure asks for the phenomenon before the explanation, and the practical benefit is that the description defines your dependent variable for you. Studies that start from the causes leg usually discover halfway through that nobody has established the thing being explained, at which point the design has to change.
Writing the phenomenon leg costs one sentence and prevents that.
Assessment move
Take the last three research questions you have seen, whether in a published article or in a classmate's draft, and label each leg: phenomenon, impact, cause. Most will have two legs and be missing one, and the missing one will tell you what the study could not claim. Then do it to your own question. This takes ten minutes and does more for a rationale section than a week of extra reading.
Working through Research as Scientific Inquiry in MEDD8001? Sia is AskSia’s AI Education tutor — ask any MEDD8001 Research as Scientific Inquiry question and get a clear, step-by-step explanation grounded in how MEDD8001 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
