INT6067 Chap.1 Research and Research Design
Research and Research Design
Systematic is the word that does the work
The course defines research as systematic inquiry undertaken to describe, explain, predict or control an observed phenomenon. Systematic does not mean complicated and it does not mean numerical. It means the steps were decided in advance and written down, so that a second person could repeat them and disagree with you about the result rather than about what you did.
That is the difference between a proposal and an opinion, and it is why this course assesses a proposal before any study is run.
The course synopsis makes the target explicit: you are to understand what research design and method in artificial intelligence and educational technology rest on, well enough to develop your own proposal, with a literature review, a statement of the problem and a design that uses appropriate methods.
Four purposes, four different evidence burdens
Describe, explain, predict and control are not synonyms, and the first decision in any design is which of them you are claiming.
Description needs complete capture over a defined period and a scheme for reducing what was captured. Explanation needs accounts from the people involved, because a log file cannot supply a reason. Prediction needs an outcome recorded later than the things used to predict it, and some data held back so that performance is measured on cases the model has not seen.
Control, which is what words like effect and impact quietly claim, needs at least two conditions and a stated rule for how participants reached them.
The most common structural failure in a first proposal is a question that claims the fourth and a method that can only deliver the first.
Design and method are not the same thing
Method is the technique used to gather or analyse evidence: an interview, a log extraction, a test before and after, a coding scheme.
Design is the arrangement those techniques sit in: who is compared with whom, in what order, over what period. Two studies can share every method and answer different questions because their designs differ, which is why a list of techniques is never an answer to the question of what your design is.
A design is not settled until each of a short list of questions has an answer: who is studied and how many, what is measured and with what instrument, over what period and at how many points, against what comparison, and how the evidence is reduced to something a reader can inspect.
Where this course sits in the life of a study
The first lecture sets out the stages of conducting a study: identify a significant research problem, prepare a proposal, run a pilot, run the main study, and prepare a report.
Only the first two belong to this course. The rest happens in the independent project that follows, which is also why the quality of the proposal matters more than its length. It is the document the next course executes.
What this chapter covers
- 01
Systematic inquiry, and what the word systematic commits you to
- 02
Describe, explain, predict and control as four different claims
- 03
The evidence each purpose obliges you to collect
- 04
Design against method, and why a technique list is not a design
- 05
The five decisions a design has to settle before it can be judged
- 06
The five stages of a study, and the two this course assesses
- 07
Qualitative, quantitative and mixed approaches, and what each one costs
- 08
Why the purpose is chosen before the design, never after
Four questions about one classroom, four different designs
- 4Name the purpose each question claims.
- 4State the one design feature each purpose then forces.
Key terms
- Systematic Inquiry
- Investigation carried out to a plan decided in advance and written down, so that another person could repeat the steps and contest the result rather than the procedure.
- Research Purpose
- Which of description, explanation, prediction or control a study claims. It is chosen first because it determines what evidence the design is obliged to produce.
- Empirical Issue
- A question whose answer turns on observation rather than on argument alone. The course asks you to investigate empirical issues in artificial intelligence and educational technology.
- Research Proposal
- A document that argues for a study before it happens, so that a reader can judge whether it would work. In this programme it is also the plan the following project course carries out.
- Pilot Study
- A small run conducted before the main study to develop and try out the data collection methods and other procedures. It is where a coding scheme or an instrument earns its place.
- Mixed Methods
- An approach that combines qualitative and quantitative work in one study. It buys both a comparison and an account of it, and costs time, which is the binding constraint in a one semester project.
Research and Research Design FAQ
What is the difference between a research design and a research method?
Method is the technique used to gather or analyse evidence, such as an interview, a test before and after teaching, or a scheme for labelling recorded behaviour. Design is the arrangement those techniques sit in: who is compared with whom, in what order, over what period, and against what.
Two studies using identical techniques can answer completely different questions because their arrangements differ, so naming a tradition or listing instruments does not tell a reader what your design is.
Do I have to choose between qualitative and quantitative work in this course?
You have to choose what your final claim will say, and that settles it for you. A claim containing a comparison or a rate needs numbers from more than one condition or more than one point in time. A claim containing a reason, a perception or a process needs accounts from participants themselves.
Combining both is legitimate and is effectively two studies, so within one semester the honest arrangement is to name which one comes first and say what the other would require.
Why does the course assess a proposal rather than a finished study?
Because the proposal is the artefact the next course executes. The programme pairs this course, where you learn the concepts, principles and methods, with an independent project in which they are applied to a small scale study or a workable instructional solution. A weak proposal therefore costs twice, once in the mark here and again in the semester spent carrying out a plan whose question was never answerable.
Assessment move
Pick one deployment you could plausibly study in a school or university you already have access to, and carry it through every chapter of the course. Before each session, write your own answer to that session's question about your case; after the session, write one sentence on what you would now change.
By the time the topic submission arrives you will have a candidate that has survived nine tests rather than an idea you had on the weekend.
Working through Research and Research Design in INT6067? Sia is AskSia’s AI Education tutor — ask any INT6067 Research and Research Design question and get a clear, step-by-step explanation grounded in how INT6067 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
