Mid-term final notes

This subject's mid-term uses Bloom's Taxonomy "K-levels" printed right next to each question's marks — you'll see them on the real paper as K1 (3), K2 (4), K3 (5), K4 (8), K5 (10). Knowing what each level actually wants is worth more than any other single piece of exam prep here.

Level	Verb style	Marks seen	What the examiner wants
K1 — Remember	Describe, List, State, Define	3	A clean definition/list, straight from memory. No analysis needed — don't overwrite this one.
K2 — Understand	Identify, Explain, Define	4	Definition plus why/how, in your own words — one short paragraph or a labelled list.
K3 — Apply	Contrast, Evaluate, Illustrate	5	Use the concept on a case: a comparison table, or apply the idea to a concrete example.
K4 — Analyze	Identify + formulate, Sketch a plan	8	Break a real scenario into parts and build something from it — a problem statement, a plan, a design.
K5 — Evaluate/Create	Elaborate, Design a framework	10	The biggest answer on the paper: structured, multi-part, synthesising several ideas into one coherent output.

Scoring pattern that works across K-levels: short definition first (1–2 lines) → structured body (numbered points, a table, or named stages — never one big paragraph) → a concrete example or mini-case tied to the question's own scenario if it gives one → for K4/K5, close with a short "why this works" or "limitations" line, since synthesis questions reward judgement, not just recall.

Where this document's 20 topics come from: the official lecture-wise syllabus (Lectures 1–20, grouped into the three real units — Research Problem, Literature Studies, Technical Writing). Every topic below folds in the actual questions that test it, pulled from both the real Nov 2025 mid-term paper and the full 36-question practice set — so each topic is "learn the idea, then see exactly how it gets asked."

The idea, in plain words

Research methodology is the systematic science of how to do research — the overall framework and the reasoning behind why a researcher picks one way of investigating a problem over another. It's easy to confuse this with a research method, but they're different layers:

• Research methodology = the "why/how we choose" — the logic and strategy behind the whole study.
• Research method = the "what we actually do" — one specific technique, like running a survey, conducting an experiment, or doing a case study.

So methodology is the bigger picture that justifies picking a particular method; the method is one tool chosen because the methodology says it's the right one for this problem.

Objectives of research

Every research study is ultimately aiming at one (or more) of these four objectives:

• Exploratory — to gain familiarity with a new phenomenon or get fresh insight into it, often when little is known yet.
• Descriptive — to accurately portray the characteristics of a group, situation, or phenomenon as it currently exists.
• Diagnostic — to determine how frequently something occurs, or whether it's associated with something else.
• Hypothesis-testing — to test whether a causal relationship actually exists between two or more variables.

Significance of research methodology

Why does learning this matter at all? Research methodology is significant for:

• Sound decision-making — governments and businesses rely on properly conducted research to make policy and strategic decisions instead of guessing.
• Scholars and degree-seekers — anyone pursuing a higher degree or academic career needs a rigorous methodology to produce defensible, credible work.
• Professionals — staying current in a field often means being able to read, evaluate, and apply new research correctly.
• Everyday problem-solving — the same systematic thinking (define the problem, gather evidence, evaluate alternatives, decide) is useful well outside formal academic research.

Taken together: methodology is what turns a casual guess into a defensible, repeatable, trustworthy conclusion. It's the foundation the rest of this unit is built on — every later topic (sources of a research problem, data collection, analysis, ethics) is really a deeper look at one piece of this overall framework.

Exam questions this topic answers

No question in the real mid-term paper or the practice set targets this lecture directly — it's the foundational orientation topic the rest of the syllabus builds on, not a standalone exam answer. If a question does ask something like "What is research methodology?" or "Distinguish between research method and research methodology," the answer is exactly the distinction and the objectives/significance points above: define methodology, contrast it against method with the example given, then list the four objectives and the significance points as supporting detail.

The idea, in plain words

A research problem exists when there's a gap between what is currently known and what needs to be known — or when there's some difficulty or question that genuinely needs systematic investigation to resolve. It's not just "a topic you find interesting"; it has to be a real, structured problem that research can actually be designed around.

The five components of a true research problem

For something to count as a proper research problem (not just a vague curiosity), it needs all five of these:

1. An individual or group with the problem — someone who actually faces the difficulty or needs the answer.
2. Objectives to be achieved — a clear goal the research is trying to reach.
3. Alternative means of reaching those objectives — more than one possible way to get there (if there's only one obvious way, there's nothing to investigate).
4. Genuine doubt about which alternative is best — real uncertainty about which option is correct or most effective; without this doubt, there's no question left to research.
5. An environment/context — the setting or conditions the problem exists in, which shapes which alternatives are even realistic.

Types of research

Research doesn't fall into one single list of categories — it's described along several independent dimension-pairs, and a given study can be classified on several of these at once.

• Basic / fundamental vs applied — basic research builds theory or knowledge for its own sake (e.g. a physicist studying a subatomic particle with no immediate product in mind); applied research solves a specific practical problem (e.g. a company researching how to make a battery charge faster).
• Descriptive vs analytical — descriptive research portrays "what is" (e.g. a survey describing current smartphone usage habits among teenagers); analytical research uses existing facts to critically evaluate "why" (e.g. analysing existing sales data to explain why a product's sales dropped).
• Quantitative vs qualitative — quantitative research is numerical/statistical (e.g. measuring the percentage of customers who prefer feature A over B); qualitative research is non-numerical and focused on meaning or quality (e.g. interviewing customers about why they feel loyal to a brand).
• Conceptual vs empirical — conceptual research is abstract and theory-building, often done by thinkers reasoning from existing ideas (e.g. developing a new theoretical model of motivation); empirical research is grounded in observed data or direct experience (e.g. running an experiment to test whether that motivation model actually predicts behaviour).

A different three-way split: theoretical, applied, and action research

Separately from the pairs above, research problems are sometimes classified into three types based on who is doing the research and why:

• Theoretical research — builds general knowledge or theory with no immediate practical goal; typically done in academic settings to advance understanding itself.
• Applied research — aims to solve one specific, real-world problem for a person or organisation.
• Action research — carried out by practitioners themselves (teachers, managers, nurses) to improve their own immediate practice while they study it. It's usually cyclical: plan → act → observe → reflect, then repeat with adjustments.

Each fits a different situation: theoretical research suits foundational academic work where building knowledge is the point; applied research suits a concrete organisational problem that needs a specific fix; action research suits a practitioner who is embedded in the situation and wants to improve their own practice directly, in real time, while generating research knowledge from doing so.

Exam questions this topic answers

Real mid-term, Q2 (K1, 3 marks): "Identify the different types of research and provide an example of each."

Answer plan: at K1/3 marks this wants a short, clean list, not an essay. List the four dimension-pairs — basic/applied, descriptive/analytical, quantitative/qualitative, conceptual/empirical — and give exactly one real-world example per type (the examples above work directly), keeping each to a single line.

Practice set, Unit 2 Q5: "Classify the different types of research problems (theoretical, applied, and action research) and analyze the situations in which each type would be most appropriate."

Answer plan: this question specifically wants the three-way split, not the dimension-pairs above — define theoretical, applied, and action research as given above, then analyse the fitting situation for each: theoretical for foundational academic work with no immediate practical aim, applied for solving a specific organisation's concrete problem, and action research when the researcher is a practitioner embedded in the situation who wants to improve their own practice directly while studying it (note the plan → act → observe → reflect cycle as a concrete detail that shows deeper understanding).

Real mid-term, Q7 (K4, 8 marks): "Identify a specific research problem and formulate a detailed research problem statement. Your statement should address all key elements required in a well-defined research problem statement."

Answer plan: this is a K4/8-mark synthesis question — pick one concrete, specific problem (don't stay abstract) and write out a full problem statement that explicitly labels each required element as you go, drawing directly on the five components above plus the characteristics from topic 4 and the scope/objectives from topic 6:

1. Background/context — briefly set the scene (e.g. "Online learning platforms have grown rapidly, but engagement data shows high drop-off rates mid-course.")
2. The problem/gap itself — state the specific difficulty or unanswered question clearly (e.g. "It is unclear which specific course-design factors most strongly predict whether undergraduate students disengage from an online course before completion.")
3. Significance — why it's worth answering (e.g. helps institutions redesign courses to reduce dropout, improves student outcomes).
4. Scope and boundaries — what's included/excluded (e.g. limited to undergraduate students, one institution, one semester — see topic 6).
5. Objectives / research questions — specific, measurable goals (e.g. "To identify the three course-design factors most correlated with mid-course disengagement among undergraduates at [institution] in [semester].").
6. The individual/group affected and the alternatives/doubt — name who faces the problem (students, instructors) and note that there's genuine uncertainty about which design factor(s) matter most (satisfying components 1, 3, and 4 from the five-component list above).

Writing it this way — one example, every element explicitly labelled — is what distinguishes a full K4/8-mark answer from a vague paragraph: the examiner is checking that you know all the pieces a real problem statement needs, not just that you can describe a topic.

The idea, in plain words

A research problem doesn't just appear out of nowhere — it comes from somewhere real. Knowing the common sources of research problems is useful both for finding your own problem and for explaining, in an exam, where good problems typically come from.

Where research problems actually come from

• Personal or professional experience — real difficulties a researcher encounters firsthand in their own field or job, which they then turn into a formal question to investigate.
• Review of existing literature — reading what's already been published often reveals gaps, contradictions, or questions nobody has fully answered yet.
• Intuition and creative thinking — a researcher's own "why does this happen?" curiosity, sometimes before any literature or data points to it.
• Discussions with experts, mentors, and peers — conversations that surface problems a researcher wouldn't have noticed alone, especially from people closer to the practical side of a field.
• Current social, economic, or technological issues — real-world developments (a new technology, an economic shift, a social trend) that raise new questions needing study.
• Replication of an existing study in a new context — repeating a known study with a new population, a new place, or at a new point in time, to see if the original findings still hold.
• Sponsoring-agency or funding-call requirements — many research problems are shaped (or directly specified) by the organisation funding the study, which defines the problem area it wants investigated.

Exam questions this topic answers

Practice set, Unit 2 Q1: "Analyze the various sources of a research problem and justify how prior research and intellectual curiosity contribute to identifying a good research problem."

Answer plan: first list all the sources above so the "analyze the various sources" part is fully covered. Then specifically justify the two sources the question names:

• Prior research (review of literature) — this is the single most productive and most common professional source, because a careful literature review doesn't just confirm what's known; it actively surfaces gaps, contradictions between studies, and questions earlier researchers raised but never answered. A researcher reading broadly in a field is almost guaranteed to find at least one legitimate unanswered question this way.
• Intellectual curiosity — a researcher's own genuine "why does this happen?" question, arising from observation rather than reading, often leads to problems that are genuinely novel — the kind of question literature alone wouldn't surface, because nobody has thought to write about it yet. Curiosity is what drives a researcher to notice something odd in the first place, before any formal literature search even begins.

Conclude by noting that these two sources work well together in practice: curiosity generates the initial spark of a question, and the literature review then confirms whether it's genuinely unanswered and worth pursuing.

The idea, in plain words

Not every question is worth researching. A good research problem is one that's actually possible to investigate, worth investigating, and will produce a result someone can trust and use. Before committing months to a study, a researcher checks a problem against a short list of characteristics — if it fails most of them, it gets reshaped or dropped.

The characteristics

• Novelty / originality — it adds something new: a gap in existing knowledge, a fresh angle on an old question, or a new context (new population, new technology, new time period) for a known question. A problem that just repeats a settled answer isn't worth a study.
• Clarity and specificity — the problem is stated precisely enough that two different researchers would understand it the same way. "Study social media" is not a research problem; "Does daily Instagram use above 2 hours correlate with self-reported anxiety in college students?" is.
• Feasibility — it can actually be completed with the time, budget, data access, equipment, and skills available. A brilliant question that needs five years and a lab nobody has access to isn't feasible for most researchers.
• Measurability / researchability — the variables involved can be observed, measured, or tested empirically. Pure value judgements ("is modern art better than classical art?") aren't researchable in the scientific sense; they can't be settled with data.
• Relevance / significance — the answer matters to the field, to practice, or to society. It should fill a genuine gap, not just satisfy curiosity with no wider use.
• Ethical soundness — the study can be conducted without harming participants, violating privacy, or requiring data that can't be ethically obtained.

Why these characteristics matter (the "why" the exam wants)

• Novelty and relevance protect against wasted effort — research that only repeats what's already known doesn't advance the field and is hard to get published or funded.
• Clarity keeps the whole study focused: a vague problem leads to a vague methodology, which leads to results nobody can act on.
• Feasibility is a practical filter — the best research problem in the world is worthless if it can never be completed within the real constraints a researcher has.
• Measurability is what separates a research problem from a philosophical or opinion-based question — without measurable variables, there's no way to collect evidence or draw a defendable conclusion.
• Ethical soundness protects participants and the researcher's own credibility; an ethically compromised study can be rejected outright regardless of how good its other qualities are.

Put together: a problem with all of these is one a researcher can actually start, finish, defend, and have taken seriously.

Exam questions this topic answers

Real mid-term, Q1 (K1, 3 marks): "Describe the characteristics of a good research problem and explain why these characteristics are important."

Answer plan: list the six characteristics above (novelty, clarity, feasibility, measurability, relevance, ethical soundness) in one or two words each with a one-line description, then pair each with its "why it matters" line from the section above. At K1/3 marks, keep each point to a single sentence — this is a recall-and-list question, not an essay; a clean bulleted list scores full marks faster than paragraphs.

Practice set, Unit 2 (research problem) Q2: "Examine the criteria for selecting a good research problem and relate how factors such as feasibility, originality, and researcher's competence influence the outcome of a study."

Answer plan: this is the same characteristic list, but at a higher K-level (Examine/relate = K3 territory), so go beyond naming them — for each of feasibility, originality, and researcher's competence specifically, explain the consequence of getting it wrong: low feasibility → the study stalls or is abandoned partway; low originality → wasted effort, rejection by reviewers/examiners; a mismatch between the problem and the researcher's own competence (statistical skill, domain knowledge, access to the field) → poor-quality execution even if the problem itself was good. Note that "researcher's competence" is its own factor, distinct from the problem's own qualities — a good problem can still fail in the hands of the wrong researcher.

The idea, in plain words

Picking a research problem is easy to get wrong in predictable ways. Knowing the common errors helps a researcher avoid wasting months on a problem that was never going to work — and it's a direct flip side of topic 4 (Characteristics of a Good Research Problem): most of these errors are simply what happens when one of those good characteristics is missing.

Common errors in selecting a research problem

• Too broad or too vague — the problem can't actually be investigated within real time, budget, or data constraints because it tries to cover too much ground.
• Not researchable — the problem is really a value judgement ("is X better than Y?"), not something data can settle empirically.
• Already thoroughly studied — the problem lacks originality; it just repeats a settled answer instead of adding anything new.
• Beyond the researcher's competence — the problem demands skills, domain knowledge, or resources (equipment, funding, field access) the researcher doesn't actually have.
• Ambiguous problem statement — the problem itself is worded unclearly, so different people reading it would understand it differently.
• Feasibility never checked — time, cost, and data-access constraints weren't verified before committing to the problem, only discovered partway through.
• Personal bias shaped the choice — the problem was picked because of the researcher's own preference or agenda rather than a genuine need or gap.
• Doesn't match the researcher's real interest — even a technically sound problem hurts motivation and the quality of the work over a long study if the researcher isn't genuinely interested in it.

Exam questions this topic answers

Practice set, Unit 2 Q6: "Examine the common errors while selecting a research problem, and suggest measures to minimize such errors."

Answer plan: list each error above, pairing it with one concrete fix:

• Too broad/vague → narrow the scope with clear boundaries early (define population, time period, and variables up front).
• Not researchable → reframe the question so it's measurable and empirical, not a value judgement.
• Already studied → do a proper literature review first to confirm the gap genuinely exists.
• Beyond competence → honestly assess the required skills and resources before committing, and seek co-researchers or training if there's a gap.
• Ambiguous statement → write and test the problem statement for clarity, ideally getting a second reader to check it means one single thing.
• Feasibility not checked → run a feasibility check (time, cost, data access) before finalising the problem, not after starting.
• Personal bias → consult peers/mentors to sanity-check that the problem reflects a genuine need, not just the researcher's own preference.
• Doesn't match real interest → choose among feasible, valid problems the one the researcher is genuinely motivated by, since long studies need sustained motivation.

Practice set, Unit 2 Q3: "Differentiate between quantitative and qualitative errors in selecting a research problem, and analyze their impact on the validity of research findings."

Answer plan: this question's wording isn't rigidly standardised in the literature, so state plainly that what follows is one clear, reasoned framing of the distinction, then present it:

• Quantitative errors — errors about scope and measurability: the problem is too broad, too narrow, has unmeasurable variables, or the researcher has insufficient sample size or data access. Impact on validity: these tend to make results statistically unreliable or inconclusive — the numbers either can't be collected properly or don't carry enough statistical power to support a real conclusion.
• Qualitative errors — errors about judgement and framing: ambiguous wording, personal bias shaping the problem, ethical problems with the study, or poor alignment between the stated problem and the real underlying need. Impact on validity: these tend to produce results that may look statistically fine on the surface but are meaningless, misdirected, or biased in interpretation — the data collection process itself worked, but it was answering the wrong or a distorted question.

Close by noting the key practical difference: quantitative errors usually show up as a study that can't produce usable numbers at all, while qualitative errors are more dangerous because they can produce numbers that look valid but are actually answering a poorly framed or biased version of the real problem.

The idea, in plain words

Once a research problem is chosen, two things have to be nailed down before any data collection starts: how far the study goes (scope) and what it's actually trying to achieve (objectives). Think of scope as the fence around a field, and objectives as the specific spots inside that fence the researcher is walking toward. Skip either one and the study drifts — too much gets studied, or nobody can say what "finished" looks like.

Scope

Scope defines the boundaries of the study — what is included, and just as importantly, what is explicitly excluded. It's usually set along a few dimensions:

• Population — who or what is being studied (e.g. "college students," not "all students everywhere").
• Time period — the window the data or the phenomenon covers (e.g. "the last three academic years").
• Geography — the place or region the study is limited to (e.g. "colleges in one city," not nationwide).
• Variables — exactly which factors are being looked at, and which related factors are deliberately left out of this particular study.

A clearly stated scope keeps the study focused and feasible. It's the main defence against scope creep — the slow, tempting drift of adding "just one more angle" until the project is too big to finish with the time, budget, and data actually available.

Objectives

Objectives are the specific, measurable goals the research is trying to reach. A study usually states:

• One general objective — the overall aim, often close to the research problem's title.
• Several specific objectives — concrete, worded with action verbs like "To examine X," "To determine Y," "To compare A and B." Each one should be something a reader can check off as achieved or not once the study is done.

Objectives give the study a measurable finish line. A vague goal like "to understand customer behaviour" can never really be shown to be complete; "to determine whether delivery time affects repeat purchase rate" can.

How scope and objectives work together

Scope and objectives aren't two separate boxes to tick — they operate together as the control system for the whole study:

• Scope is the fence: it tells the researcher (and anyone reading the study later) what is in bounds and what is deliberately left out.
• Objectives are the destinations inside that fence: they tell the researcher exactly what to aim for within the boundaries scope has already set.
• Every later methodological choice — who gets sampled, which data collection method is used, which analysis technique is applied — is designed to meet the stated objectives without spilling outside the defined scope.

Example. Suppose the research problem is about remote-work productivity.

• Scope: software employees (population) at mid-sized Indian IT firms (geography), surveyed over the 2024–25 financial year (time period), looking only at self-reported productivity and hours worked (variables) — explicitly excluding freelancers and excluding any analysis of mental health outcomes.
• Objectives: the general objective is "to examine the effect of remote work on employee productivity"; specific objectives might be "to determine the average change in self-reported output after shifting to remote work" and "to compare productivity between employees with under and over two years of tenure."

Notice how every objective stays strictly inside the scope's fence — none of them quietly reaches for data outside the stated population, time period, or variables.

Exam questions this topic answers

Practice set, Unit 2 (research problem) Q4: "Compare the scope and objectives of a research problem, and examine how they collectively guide the direction of a research study."

Answer plan: define scope (boundaries — population, time period, geography, variables) and objectives (specific, measurable goals — one general plus several specific) clearly and separately first, since the question explicitly asks to compare them. Then use the remote-work example above (or build a similarly matched pair) to show one scope statement and the objectives that sit inside it. Finish with the "fence + destinations" explanation: scope sets what the study is and isn't allowed to touch, objectives set what it's specifically trying to find out within that space, and together they determine every later methodological decision — sampling, data collection method, and analysis technique are all chosen to hit the objectives without exceeding the scope.

The idea, in plain words

Once a research problem and its scope/objectives are set, the researcher has to decide how to go about actually investigating it. There are three broad approaches, and the right one depends on whether the question needs hard numbers, rich human detail, or both.

Quantitative approach

Investigates the problem through numerical data and statistical analysis.

• Uses larger samples, aiming for results that can be generalised to a wider population.
• Tests a specific, pre-defined hypothesis.
• Relies on objective, standardised instruments (e.g. structured questionnaires with fixed-choice answers) so that results from different respondents can be directly compared and statistically analysed.
• Good fit when the research question is something like "how much," "how many," or "is there a measurable relationship between X and Y."

Qualitative approach

Investigates the problem through non-numerical data — interviews, direct observation, case studies, open-ended discussion.

• Uses small samples, because the goal is depth, not breadth.
• Explores meaning, experience, and context in detail; it's interpretive rather than purely objective — the researcher's own careful reading of the material is part of the method.
• Not meant to produce results that generalise to everyone; it's meant to understand a phenomenon richly, in its own context.
• Good fit for "why" and "how" questions, or when the researcher doesn't yet know enough about the phenomenon to even write a fixed-choice questionnaire about it.

Pragmatic / mixed-methods approach

Combines quantitative and qualitative methods within the same study.

• The choice isn't made out of philosophical loyalty to one paradigm — it's made because the specific research question genuinely needs both the breadth that quantitative data gives and the depth that qualitative data gives to be properly answered.
• Typical pattern: a quantitative survey establishes what is happening at scale, and a smaller set of qualitative interviews explains why it's happening, or vice versa.

Exam questions this topic answers

Practice set, Unit 3 (literature/data) Q7: "Evaluate the quantitative, qualitative, and pragmatic approaches to investigating a research problem."

Answer plan: describe all three approaches using the detail above — for each, cover what kind of data it uses, sample size, and the kind of question it suits. Then actually evaluate them against a few criteria, since the question says "evaluate," not just "describe":

• Generalisability — quantitative is strong here (large samples, statistical testing); qualitative is weak (small samples, context-specific); pragmatic can achieve some of both.
• Depth of insight — qualitative is strongest (rich, contextual understanding); quantitative is weakest on its own (numbers without the "why"); pragmatic captures both breadth and depth.
• Cost and time — quantitative is often faster to run at scale but needs careful instrument design upfront; qualitative is slower per respondent but needs fewer of them; pragmatic costs the most overall since it runs two research processes.
• Suitability — conclude that there's no universally "best" approach: the right choice depends entirely on whether the research question needs breadth (go quantitative), depth (go qualitative), or genuinely needs both to be properly answered (go pragmatic/mixed-methods).

The idea, in plain words

Data is the raw facts, figures, or observations a researcher gathers to investigate the research problem — it's the evidence everything else (analysis, interpretation, conclusions) is built on. Data can be split along more than one dimension, but the exam-relevant split is about where the data comes from: did the researcher collect it themselves for this exact study (primary), or is it something someone else already collected for a different purpose (secondary)? There's also the numbers-vs-descriptions split: quantitative data is numerical (e.g. "62% of respondents agreed"), while qualitative data is descriptive, non-numerical information about qualities or experiences (e.g. "respondents said the new interface felt confusing").

Primary data

Collected firsthand by the researcher, specifically for the study at hand. Common methods:

• Questionnaires / surveys — structured questions sent to a sample of respondents.
• Interviews — structured (fixed questions, easy to compare across respondents) or unstructured (open conversation, more depth but harder to compare).
• Direct observation — watching and recording behaviour as it happens, without relying on self-report.
• Experiments — manipulating a variable under controlled conditions to observe the effect.

Trade-off: primary data is more relevant and current — it's built exactly around this study's own questions — but it's costlier and more time-consuming to collect than reusing existing data.

Secondary data

Data that was already collected by someone else, for a different original purpose — books, journal articles, government reports, existing databases, past survey datasets.

Trade-off: secondary data is cheaper and faster to access since the collection work is already done, but it may not perfectly fit the new research question (wrong population, wrong time period, wrong variables measured), and its original quality or bias can't always be verified by the new researcher.

Choosing a primary data collection method

Even once "go primary" is decided, the specific method still matters, and the right one depends on the scenario:

• Questionnaire — cheap, scales to large samples easily, but gives less depth per respondent (fixed questions can't chase an interesting answer further).
• Interview — gives rich, detailed information and lets the researcher probe follow-up answers, but is time-costly per respondent and harder to run at scale.
• Observation — captures what people actually do (real behaviour), but can't capture the motives or reasons behind that behaviour — the researcher has to infer those, or pair observation with another method.

The right choice depends on the specific research scenario's goals: breadth and speed point toward questionnaires, depth and nuance point toward interviews, and verifying actual behaviour (as opposed to what people say they do) points toward observation.

Exam questions this topic answers

Real mid-term, Q4 (K2, 4 marks): "Define data and its types with appropriate examples."

Answer plan: at K2/4 marks this wants a clear definition-plus-examples paragraph, not a long essay. First define data (the raw facts, figures, or observations gathered as evidence for a research problem). Then give the primary/secondary split with one example each (primary: a researcher's own customer survey; secondary: a government census report reused for a new study), and the quantitative/qualitative split with one example each (quantitative: age or income figures; qualitative: a respondent's written description of their experience). Four types, four short examples — that's the whole answer at this mark weight.

Real mid-term, Q5 (K3, 5 marks): "Contrast between primary and secondary data collection methods in research. Give examples of situations where each method would be most suitable."

Answer plan: build a short contrast covering source, cost, time, relevance, and control over quality:

• Source — primary is collected firsthand by the researcher; secondary was collected by someone else for a different purpose.
• Cost and time — primary is costlier and slower to collect; secondary is cheaper and faster since it already exists.
• Relevance — primary fits the exact research question perfectly since it was designed for it; secondary may only approximately fit.
• Control over quality — the researcher fully controls primary data's quality and method; secondary data's original quality and any bias in it can't always be verified.

Then give one concrete "when to use" example for each, as the question explicitly asks: primary is most suitable for a brand-new question nobody has studied yet — e.g. a company directly surveying its own customers about a new product feature. Secondary is most suitable for a quick trend analysis using data that already exists in the right form — e.g. using existing government census data to study population trends, where collecting brand-new data would be redundant and wasteful.

Practice set, Unit 3 (literature/data) Q8: "Assess the suitability of different primary data collection methods for a given research scenario."

Answer plan: briefly compare the three primary methods as above — questionnaire (cheap, scalable, less depth), interview (rich detail, time-costly), observation (captures real behaviour, can't capture motives or reasons) — then note explicitly that the right choice depends on the specific research scenario's own goals: if the priority is reaching a large sample quickly and cheaply, choose questionnaires; if the priority is deep understanding of a smaller group's reasoning, choose interviews; if the priority is verifying actual behaviour rather than self-reported behaviour, choose observation. A strong answer picks one example scenario and justifies which method fits it best, rather than just repeating the three definitions.

The idea, in plain words

Once data is collected, it's still just raw material — data analysis is the process of organising, summarising, and examining that data to find patterns in it. There are two broad families of analysis, and they answer two different questions.

Descriptive analysis

Summarises what the data shows, for the sample that was actually collected. It answers: "what's the pattern in this sample?"

Typical tools:

• Mean, median, mode — central-tendency measures that summarise a "typical" value.
• Frequency counts — how often each value or category occurs.
• Charts and tables — visual summaries that make the pattern easy to see at a glance.

Descriptive analysis never claims anything about people or cases outside the sample itself — it's purely a summary of the data in hand.

Inferential analysis

Uses the sample data to make a generalisation or prediction about a larger population beyond just the sample collected. It answers: "what's likely true beyond this sample?"

Typical tools:

• Hypothesis testing — checking whether an observed effect in the sample is likely to be real in the wider population, or just due to chance.
• Confidence intervals — a range within which the true population value is likely to fall.
• Correlation — whether two variables move together.
• Regression — modelling how one variable predicts or affects another.

Inferential analysis is what lets a study say something about people or cases it never actually collected data from — but that leap is only valid if the sample was properly selected and the relevant statistical assumptions hold.

How they relate

Descriptive analysis is the necessary first step — it's how a researcher actually understands their own dataset before trying to claim anything bigger about it. Inferential analysis is the step that turns a sample-level summary into a defensible claim about the wider population the sample was drawn from. Skipping straight to inferential analysis without first understanding the data descriptively is how misleading or poorly-checked conclusions happen.

Exam questions this topic answers

Practice set, Unit 3 (literature/data) Q11: "Judge the effectiveness of descriptive versus inferential analysis in drawing research conclusions."

Answer plan: since the question says "judge," give a real verdict, not just two definitions. Descriptive analysis is effective for simply understanding and reporting the sample itself — mean, median, frequency counts, and charts give a clear, honest picture of what was actually collected — but it can't, by itself, support any claim about people or cases outside that sample. Inferential analysis is needed whenever the research wants to generalise its findings to a wider population — hypothesis tests, confidence intervals, correlation, and regression are what make that leap possible — but inferential analysis is only as trustworthy as the sampling behind it: proper random/representative sampling and the right statistical assumptions have to hold, or the inference is invalid even if the maths is done correctly. Judge that most rigorous research needs both, used in sequence: descriptive analysis first, to actually understand what the data shows, and inferential analysis second, to draw the real, generalisable research conclusion the study is aiming for. Neither one alone is sufficient — descriptive without inferential can't generalise, and inferential without first understanding the data descriptively risks drawing conclusions nobody has actually sanity-checked.

The idea, in plain words

Analysis gives you numbers and patterns. Interpretation is the step after that: figuring out what those numbers actually mean, in plain language, and tying them back to the original research question. Two studies can run the exact same statistical test and still "interpret" the result differently depending on how carefully they connect it back to the real-world question they set out to answer. Interpretation is where a pile of statistics turns into an actual finding.

What interpretation means, specifically

• It explains the pattern the analysis found — not just "mean = 42," but what a mean of 42 actually implies about the thing being studied.
• It connects back to the research objectives and hypothesis — every interpreted finding should answer, directly, one of the questions the study set out to ask. A result that doesn't map back to the original problem is interesting trivia, not a research finding.
• It's the bridge between the data and the conclusions/recommendations section of a report — nothing after this point should introduce a claim the data itself doesn't actually support.

Precautions while interpreting data

• Don't over-generalise — a finding only applies to what the sample actually represents. Claiming a result holds for "all students" when the sample was 50 students from one college is a classic over-generalisation error.
• Check for alternative explanations and confounding variables — before accepting the obvious explanation, ask whether something else could have produced the same pattern.
• Distinguish correlation from causation — two things moving together does not mean one causes the other. This is the single most common interpretation mistake in research.
• Watch for the researcher's own bias — when a result is ambiguous, it's tempting to read it in the direction you expected or hoped for. Good interpretation stays honest about what's actually ambiguous.
• Cross-check against existing literature — if a finding is wildly inconsistent with what prior studies found, that's a signal to double-check the analysis before reporting it as a breakthrough.

Exam questions this topic answers

Practice set, Unit 3 (literature/data) Q9: "Critique the techniques and precautions used in interpreting research data."

Answer plan: list the five precautions above, then critique each one specifically rather than just restating it — for example, "avoid over-generalising" is easy to say but hard to enforce in practice unless the sample's scope was clearly defined before data collection started; "distinguish correlation from causation" requires the researcher to actively design the study (or at least the analysis) to rule out alternative explanations, not just remember the rule while writing up results; "watch for researcher bias" is hardest of all to self-police, which is why peer review and independent replication exist as outside checks. Close by noting that good interpretation always ties numbers back to the specific research question — presenting statistics in isolation, without connecting them to what the study was actually trying to find out, is the core failure interpretation exists to prevent.

The idea, in plain words

A research instrument is the actual tool used to collect or measure data — the questionnaire a respondent fills in, the checklist an observer ticks off, the sensor that records a reading. The instrument is where the real world turns into usable data, so if the instrument is flawed, everything downstream — analysis, interpretation, conclusions — inherits that flaw. No amount of clever statistics can fix data that was badly measured to begin with.

Common types of instruments

• Questionnaires / surveys — a fixed set of written questions, usually self-administered.
• Interview schedules — a structured or semi-structured list of questions a researcher asks in person or over a call, allowing follow-up.
• Rating scales — items where a respondent picks a point on a scale (e.g. 1–5 agreement), used to quantify opinions or attitudes.
• Observation checklists — a predefined list of behaviours or events a researcher watches for and records as they occur.
• Tests — structured tasks used to measure ability, knowledge, or performance.
• Equipment / sensors — physical devices used to take measurements directly, mainly in experimental research.

The two essential quality criteria

• Validity — does the instrument actually measure what it claims to measure? A test meant to measure "anxiety" that actually measures "general mood" has a validity problem.
• Reliability — does the instrument give consistent results if used again under the same conditions? If the same respondent would answer very differently on a second attempt with nothing having changed, the instrument isn't reliable.

An instrument can be reliable without being valid (it's consistent, but consistently measuring the wrong thing), but it can't be truly valid without also being reasonably reliable — inconsistent measurement can't reliably capture the right thing either.

Guidelines for building a good instrument

• Use clear, unambiguous wording — every respondent should be able to read an item the same way.
• Pilot-test it on a small group before full deployment, to catch confusing items early.
• Make sure every item maps to a specific research objective — no item should be in there "just in case."
• Keep it as short as the objectives allow — long instruments cause respondent fatigue, which lowers the quality of answers toward the end.
• Make sure it's appropriate for the target respondents — their literacy level, language, and context, so the wording itself doesn't become a barrier.

Exam questions this topic answers

Real mid-term, Q3 (K2, 4 marks): "Define the necessary instrumentation for data interpretation."

Answer plan: define "research instrument" as above, list the common types (questionnaire, interview schedule, rating scale, observation checklist, test, equipment/sensor), and then make the connection the question is really asking for: the instrument is what produces the raw data that later gets analysed and interpreted — a flawed or poorly designed instrument produces data that can't be meaningfully interpreted no matter how good the later analysis is. At K2/4 marks, a clear definition-plus-list-plus-one-connecting-sentence is the right length — no need for a long essay.

Practice set, Unit 3 Q10: "Evaluate how the choice of research instrument affects the reliability of data collected."

Answer plan: explain that a poorly worded or inconsistent instrument directly lowers reliability — different respondents may interpret ambiguous items differently even though the underlying real conditions are identical, which produces inconsistent data that has nothing to do with any real difference between respondents. A well-piloted, validated instrument is precisely what makes repeat measurements comparable across respondents and across time. Conclude that instrument quality isn't a side detail — it's a direct, upstream cause of how reliable the eventual data actually is.

Practice set, Unit 3 Q12: "Critically assess the guidelines for developing a valid and reliable research instrument."

Answer plan: walk through the guidelines above (clear wording, pilot-testing, mapping items to objectives, keeping it short, matching respondents' context) and critique each one rather than just listing it — for example, pilot-testing adds real time and cost to a study, but skipping it is almost always a false economy because it's usually the only way to catch ambiguous wording before it contaminates the real dataset; "keep it short" has to be balanced against "cover every objective," so in practice researchers often have to trade some depth for completion rates. Conclude that these guidelines work best as a checklist applied together, not any single one in isolation.

The idea, in plain words

Before starting any new research, a researcher reads what's already been written on the topic. This is the literature review. Its job is simple to state but easy to do badly: find out what's already known, spot the gaps, make sure you're not accidentally repeating someone else's study, and build the theoretical foundation the new study will stand on.

Why do a literature review at all

• Identify what's already known — so the new study adds something, instead of re-discovering a settled answer.
• Spot gaps and contradictions — these gaps are very often where a genuinely good research problem comes from (see "3. Sources of Research Problem").
• Avoid duplicating existing work — wasted effort if someone already answered the exact same question.
• Build the theoretical foundation — the concepts, definitions, and prior findings the new study will build its own argument on top of.

Two main approaches

• Narrative review — a broad, descriptive summary of relevant work. Flexible structure, no fixed protocol, written more like a guided tour of the topic. Faster to produce, but more subjective and harder for someone else to reproduce exactly, since the choice of what to include is left to the reviewer's own judgement.
• Systematic review — follows a strict, predefined, and fully documented protocol for how studies are searched for, selected, and evaluated. Far more rigorous and reproducible, because another researcher following the same documented protocol should arrive at the same set of included studies. This is the standard expected in evidence-based fields (like medicine) where the review itself is treated as a piece of research in its own right.

The process, step by step

1. Define the review question — what exactly is this review trying to find out.
2. Search databases with defined keywords — using academic databases and a clear, repeatable search strategy.
3. Screen and select relevant sources against clear inclusion/exclusion criteria.
4. Critically appraise each source's quality — methodology, sample size, potential bias, relevance.
5. Synthesise findings across sources — look for patterns, agreements, and contradictions between studies, not just summaries of each one in isolation.
6. Identify the gap the new research will fill — the whole point of doing all of the above.

Exam questions this topic answers

Practice set, Unit 3 (literature/data) Q1: "Evaluate the different literature study approaches and justify which is most suitable for a systematic review."

Answer plan: describe both the narrative and systematic approaches as above, then answer the "justify" part directly: for a systematic review specifically, the systematic approach is the only suitable choice — almost by definition, since a systematic review requires the documented, reproducible search-and-selection protocol described above. Contrast this against why a narrative approach would be the weaker choice there: its flexible, judgement-driven selection process is exactly what a systematic review is designed to eliminate, so using it would undermine the rigour the systematic review is meant to provide.

Practice set, Unit 3 Q2: "Critique the process of conducting an effective literature review and propose improvements to it."

Answer plan: go through the six process steps above and critique each — for example, a keyword search alone can easily miss relevant work that uses different terminology for the same concept (propose: searching multiple databases, and adding snowball/citation-chasing search — following references forward and backward from known key papers — to cover what keyword search misses); screening against inclusion/exclusion criteria can still be subjective (propose: having two reviewers screen independently and compare); critical appraisal quality depends entirely on the reviewer's own expertise (propose: using a standard appraisal checklist rather than ad hoc judgement).

Practice set, Unit 3 Q5: "Judge the strengths and limitations of narrative review versus systematic review approaches."

Answer plan: narrative review — strength: flexible, fast, good for broadly exploring a topic that doesn't yet have a tightly defined question; limitation: subjective, hard for someone else to reproduce, carries a real risk of selective citation (unconsciously favouring sources that support a particular view). Systematic review — strength: rigorous, reproducible, specifically designed to minimise selection bias; limitation: time-consuming and resource-heavy, and its strict protocol can sometimes be too narrow for a genuinely exploratory topic where the right question hasn't been pinned down yet. Judge that the right choice depends on where the research actually is: exploring a new area favours narrative, confirming or synthesising evidence on a well-defined question favours systematic.

Real mid-term, Q8 (K4, 8 marks): "Sketch a detailed plan for conducting effective literature studies approaches. Your plan should demonstrate an understanding of the key steps and techniques in conducting literature studies."

Answer plan: this is a K4/8-mark "sketch a plan" question, so go one level more concrete than the six-step process above — name an actual technique at each stage, not just the stage name:

1. Define the review question — write it as a specific, answerable question (not a broad topic).
2. Search — use multiple academic databases (not just one), with a documented keyword list and Boolean search operators (AND/OR/NOT) to combine terms precisely.
3. Select — apply explicit, written inclusion/exclusion criteria (e.g. publication date range, study type, language) agreed before screening starts, so selection isn't made up as you go.
4. Critically appraise — score each included source against a fixed quality checklist (methodology soundness, sample size, potential bias, relevance to the question).
5. Track citations — use forward and backward citation-chasing (snowballing) from key papers to catch relevant work the keyword search alone missed (see topic 13 for why this matters).
6. Synthesise — organise findings into a synthesis matrix or thematic table so patterns, agreements, and contradictions across sources are visible at a glance, not just summarised one source at a time.
7. Identify and state the gap — the explicit output of the whole plan, which becomes the justification for the new research problem.

Naming a concrete technique at each step (Boolean operators, a written criteria list, a quality checklist, snowballing, a synthesis matrix) is what turns this into a genuine "plan" rather than a restatement of the process — that's specifically what the question is testing at K4.

The idea, in plain words

Doing a literature review means gathering the relevant sources. Literature analysis is what happens next: actually working through those sources critically and in a structured way, rather than just listing what each paper found one after another. A good literature analysis reads like an argument built from the existing evidence, ending in a clear statement of what's still missing.

The typical structure, stage by stage

1. Introduction — states the scope and purpose of the review: what question it's covering and why, so the reader knows what to expect.
2. Body, organised thematically or chronologically — groups the existing studies around themes or tracks how understanding of the topic developed over time, rather than just listing sources in whatever order they were found.
3. Critical appraisal of each source — judging methodology quality, sample size, potential bias, and relevance to the question at hand, not just summarising what each source concluded.
4. Synthesis across sources — pulling the individual appraisals together to identify patterns, contradictions between studies, and — most importantly — the specific gap the new research can address.
5. Citation tracking — following references backward (who did this paper cite — the foundational work the field is built on) and forward (who has since cited this paper — the newest work building on it). This catches both the classics and the most current developments that a simple keyword search might miss entirely.
6. Conclusion — explicitly links everything reviewed back to the new research problem: here's what's known, here's the gap, here's where this new study fits.

Why critical appraisal matters

Critical appraisal is the quality filter on the whole review. Without it, a literature review is just a list of "paper said X" statements with no judgement about whether any of them are actually trustworthy. With it, only credible, methodologically sound sources end up shaping the direction of the new study — a weak or biased prior study doesn't get to quietly set the foundation for new work.

Why citation tracking matters

A keyword search alone can miss a lot — papers that use different terminology, or very recent papers that haven't accumulated the "right" keywords in their abstract yet. Citation tracking surfaces both the most influential prior work (the papers everyone else cites, found by tracking backward) and the most current related research (found by tracking forward from a known key paper), filling in exactly the gaps a plain keyword search leaves behind.

Exam questions this topic answers

Practice set, Unit 3 (literature/data) Q3: "Assess the typical structure of a literary analysis and justify the importance of each stage."

Answer plan: walk through the six stages above in order, giving one line justifying each: introduction orients the reader; thematic/chronological organisation keeps the body from being a disconnected list; critical appraisal filters for quality; synthesis is where the actual insight comes from, not just the summary; citation tracking fills gaps a keyword search leaves; the conclusion is what actually earns the review its place in the new study by tying it to the new problem.

Practice set, Unit 3 Q4: "Evaluate the role of critical appraisal and citation tracking in an effective literature review."

Answer plan: as explained above — critical appraisal exists to raise the quality of the review (filtering out weak or biased sources before they can shape the new study), while citation tracking exists to raise the completeness of the review (catching influential and current work that a keyword search alone misses). Evaluate that a review missing either one is incomplete in a different way: missing appraisal risks building on weak foundations; missing citation tracking risks missing either the field's foundational work or its newest developments.

Practice set, Unit 3 Q6: "Critically examine how literature analysis contributes to formulating a strong research problem."

Answer plan: the synthesis stage of literature analysis is specifically where gaps, contradictions, and unanswered questions in the existing body of work get surfaced — and those gaps are themselves usually the direct source of a well-grounded, original research problem. This ties literature analysis directly back to "3. Sources of Research Problem," where "review of existing literature" is listed as one of the most productive sources of a research problem: literature analysis is the concrete process that turns that general source into an actual, specific, defensible problem statement.

The idea, in plain words

Plagiarism is presenting someone else's words, ideas, data, images, or even your own previously published work as if they were new — without giving proper credit. It isn't only "copy-pasting a paragraph" — rewording someone's argument without citing them still counts.

The four types

• Direct / verbatim plagiarism — copying text word-for-word from a source with no quotation marks and no citation. The most obvious and most heavily penalised type.
• Paraphrasing / mosaic plagiarism — rewriting someone's sentence or idea in your own words but still presenting it as your own, without citing where the idea came from. Called "mosaic" because it's often built by stitching together reworded fragments from several sources.
• Self-plagiarism — reusing substantial parts of your own earlier published work (a past assignment, thesis chapter, or paper) in a new piece without disclosing that it's been published before. Still dishonest, because it misrepresents the new work as entirely original.
• Accidental plagiarism — unintentional: forgetting a citation, paraphrasing too close to the original wording, or not realising a fact needed a source. Unintentional doesn't mean consequence-free — institutions usually still penalise it, just typically less severely than deliberate plagiarism.

Many institutions (including UGC-regulated ones in India) formalise this with a similarity-index tier system from plagiarism-detection software: roughly, a low similarity score (under ~10%) is treated as incidental and ignored, while progressively higher bands trigger escalating penalties — resubmission, a one-year bar on submitting work, up to cancellation of registration for the most severe, repeated cases. The exact thresholds vary by institution, but the pattern — "a little overlap is normal, a lot is a serious offence" — is the point to remember.

Six ways to avoid it

1. Cite every source you draw from — both direct quotes and ideas you've paraphrased.
2. Use quotation marks for any text copied word-for-word, however short.
3. Paraphrase properly — don't just swap a few synonyms; genuinely rewrite the idea in your own structure and words, and still cite the original source.
4. Run a plagiarism-detection check (e.g. Turnitin) on your own draft before submitting it, so you catch accidental overlaps yourself.
5. Keep a working bibliography as you research, so you never lose track of where an idea came from and end up guessing or skipping a citation later.
6. Get permission / give credit for data, images, or unpublished material that isn't yours — including disclosing when you're reusing your own prior work.

Consequences of plagiarism

• Academic: failing the assignment, course, or having a degree revoked if discovered after the fact; formal disciplinary proceedings.
• Professional: retraction of published papers, loss of research credibility, damage to a researcher's reputation that can follow them for years.
• Legal: copyright infringement claims if the copied material was protected — this is where plagiarism and IPR (intellectual property rights) directly overlap, since using someone's copyrighted text, images, or data without permission can be a legal violation, not just an academic one.
• Institutional: the university or publisher's own credibility and accreditation can be affected by a pattern of unpunished plagiarism.
• To the original author: they lose the credit and citation count that is often the entire "currency" of academic work — plagiarism is a real harm to a specific person, not a victimless rule-break.

Example: a student copies three paragraphs from a journal article into their report with no citation — caught by a similarity check, the report is failed and the student faces a formal hearing. A researcher reuses data from their own earlier published study without disclosing it in a new paper — the paper is later retracted when a reviewer notices the overlap, and the researcher's other work comes under scrutiny too.

Exam questions this topic answers

Real mid-term, Q6 (K3, 5 marks): "Evaluate the potential consequences of plagiarism in a research paper. Support your answer with relevant examples."

Answer plan: cover the five consequence categories above (academic, professional, legal, institutional, harm to the original author) — at K3/5 marks, "Evaluate... support with examples" wants more than a list, so attach one concrete example to at least two or three of the categories (use the two examples above, or adapt them). Close with one line tying consequences back to why plagiarism is taken so seriously: it breaks the basic trust that research results are genuinely the author's own verified work.

Practice set, Unit 2 Q7: "Evaluate the six ways to avoid plagiarism and propose measures an institution could adopt to prevent it."

Answer plan: list the six prevention methods above, then add an institutional layer on top — mandatory plagiarism-detection checks before submission, plagiarism-awareness training for new students/researchers, clear written policy with defined similarity-index thresholds and penalties, and a named academic-integrity officer or committee to review flagged cases fairly.

Practice set, Unit 2 Q9: "Critique the four types of plagiarism based on their severity and formulate a strategy to detect them."

Answer plan: rank the four types by severity — deliberate direct/verbatim plagiarism is typically treated most severely (clear intent, easiest to prove), self-plagiarism and mosaic plagiarism next (intent is often present but harder to prove conclusively), accidental plagiarism least severely (no intent, but still a lapse in care). For detection: similarity-index software catches direct and much mosaic plagiarism; self-plagiarism needs cross-referencing against an author's own publication history (software alone often misses this); accidental plagiarism is best caught by careful self-review against a citation checklist before submission.

Practice set, Unit 2 Q10: "Construct a scenario illustrating accidental plagiarism and propose suitable corrective measures."

Answer plan — example scenario: a student paraphrases a textbook definition closely while taking notes, then later drops that paraphrase into their report, having forgotten it wasn't originally their own wording and forgetting to attach a citation. Corrective measures: always cite at the moment of writing a paraphrase (not "add citations later"), keep a reference manager (Zotero, Mendeley) running throughout the research process so every note is pre-linked to its source, and run a similarity check on the finished draft as a final safety net before submission.

Practice set, Unit 2 Q12: "Formulate a comprehensive plagiarism and research ethics policy suitable for a university or research institute."

Answer plan: structure it as named sections — (1) Definition of plagiarism and its types (as above); (2) Prevention, the six methods above, made mandatory; (3) Detection, required similarity-index checks with a specified threshold before any submission is accepted; (4) Tiered penalties, scaled to severity (minor similarity → resubmission; moderate → formal warning/grade penalty; severe or repeated → suspension or expulsion); (5) Appeals process, so flagged cases get a fair review rather than an automatic penalty; (6) Training, mandatory academic-integrity orientation for all new students/researchers. This mirrors the broader research ethics framework the rest of this unit covers (see "15. Research Ethics").

The idea, in plain words

Research ethics is the set of moral principles that govern how a researcher designs, conducts, and reports a study. It isn't a side concern layered on top of "real" research — it's what makes research results trustworthy in the first place. A study with brilliant methodology but a broken ethical foundation (fabricated data, harmed participants, an undisclosed conflict of interest) can be thrown out entirely, no matter how sound its statistics look.

The core principles

• Honesty — accurate reporting of methods, data, and results; never fabricating (inventing data that was never collected) or falsifying (altering real data to get a desired result).
• Objectivity — minimising personal bias in how the study is designed, how the data is analysed, and how results are interpreted.
• Integrity — consistency between the principles a researcher claims to follow and what they actually do in practice.
• Confidentiality — protecting participants' identity and personal data, so taking part in a study doesn't expose someone to unwanted risk or exposure.
• Informed consent — participants genuinely understand what the study involves and voluntarily agree to take part, before it starts, without being pressured or misled.
• Non-maleficence — avoiding causing harm (physical, psychological, social, or financial) to participants.
• Respect for intellectual property — giving proper credit for others' words, ideas, and data. This principle directly overlaps with "14. Plagiarism" — plagiarism is, at its core, a violation of this specific ethical principle.
• Accountability — the researcher takes responsibility for the work and its consequences, rather than deflecting blame if something goes wrong.

Why research ethics matters

Ethical conduct protects the people taking part in a study from real harm. Beyond that, it's the precondition for the wider scientific and public community being able to trust published findings at all — a reader has no way to independently verify that a researcher collected and reported data honestly, so the whole system of published research runs on an assumption of ethical conduct. When that assumption is violated — through fabricated data, undisclosed harm to participants, or a hidden conflict of interest — it doesn't just damage that one study; it can invalidate otherwise competent research entirely, because the reader can no longer trust any part of it.

Exam questions this topic answers

Practice set, Unit 2 (research problem/ethics) Q8: "Design a set of research ethics guidelines and justify the importance of each principle included."

Answer plan: turn the principle list above into a short numbered guideline document, with one justification line per principle — e.g. (1) Honesty: report data exactly as collected, because fabricated results are worthless and dangerous if acted on; (2) Objectivity: minimise bias at every stage, because a biased design or analysis can produce a conclusion that was never really supported by the evidence; (3) Confidentiality: protect participant data, because without this guarantee people won't participate honestly, if at all; (4) Informed consent: get genuine, voluntary agreement before starting, because participants have a right to know what they're agreeing to; (5) Non-maleficence: avoid harming participants, because the research exists to produce knowledge, not to cause damage along the way; (6) Respect for intellectual property: credit others' work properly, because taking credit for someone else's contribution is a form of theft; (7) Accountability: the researcher owns the outcomes of their own work, good or bad.

Practice set, Unit 2 Q11: "Evaluate how research ethics influences the credibility and trustworthiness of research findings."

Answer plan: state plainly that ethics is a precondition for credibility, not a separate add-on bolted onto good methodology. Without demonstrable ethical conduct — genuine consent, honest reporting, no fabrication, no undisclosed conflicts of interest — published results can't be trusted even if the methodology looks technically sound on paper, because the reader has no way to verify that the data itself is real and uncorrupted. Evaluate with the flip side too: a single exposed ethics violation (a retraction over fabricated data is the classic case) tends to cast doubt backward over a researcher's entire body of work, not just the one flawed study — which is exactly why the research community treats ethical violations as more serious than most methodological weaknesses.

Real mid-term, Q9 (K5, 10 marks) — OR option: "Design a framework that integrates research ethics principles, and real-life case studies for new researchers."

This is a K5/10-mark synthesis question — the single biggest-weight answer in the whole exam — so it needs a genuinely structured, multi-part framework, not one paragraph gesturing at "ethics matters."

Answer plan — a four-part training framework:

1. Foundation: the core principles. Lay out the eight principles from the section above (honesty, objectivity, integrity, confidentiality, informed consent, non-maleficence, respect for intellectual property, accountability) as the framework's foundation layer — every other part of the framework refers back to these.

2. Learning layer: one real-life-style case per principle. Pair each principle with a realistic case a new researcher can learn from, so the principle isn't just an abstract rule:

   • Honesty — a researcher fabricates or touches up data to get a cleaner result; the paper is later retracted when another team can't reproduce the finding.
   • Objectivity — a study funded by a company is designed and analysed in a way that quietly favours that company's product, without disclosing the conflict of interest.
   • Confidentiality — participant data from a health study is stored insecurely and leaks, exposing people's private medical information.
   • Informed consent — participants in a psychological study are misled about what's actually being tested, and only told the truth (if at all) after the fact, without a chance to have genuinely agreed upfront.
   • Non-maleficence — an experiment causes participants real psychological distress that wasn't disclosed or planned for beforehand.
   • Respect for intellectual property — a researcher copies large sections of another paper's literature review without citation (the direct overlap with "14. Plagiarism").
   • Integrity — a researcher publicly claims a rigorous peer-reviewed process while privately skipping proper review steps to publish faster.
   • Accountability — when a study's result turns out to be wrong, the researcher blames a co-author or lab assistant instead of taking responsibility for the published work.

3. Decision checklist. A short, practical checklist a new researcher runs through before starting any study: Have participants given genuine informed consent? Is there a plan to protect their confidentiality? Could this study cause harm, and has that been minimised? Is there any conflict of interest that needs disclosing? Is every source and prior contribution properly credited? Is the data being reported exactly as collected, with no pressure to "improve" it?

4. Escalation and reporting path. What a new researcher should actually do if an ethical concern comes up mid-study: pause and document the concern; raise it with a supervisor or the institution's research ethics committee/institutional review board rather than deciding alone; follow the institution's formal reporting procedure if the concern involves misconduct by someone else; and know that raising a genuine concern early is always treated far better, by any reasonable institution, than letting a violation surface later during peer review or after publication.

Presented together — principles, cases, checklist, escalation path — this is a complete, usable framework a new researcher could actually follow, which is what a K5/10-mark "design a framework" question is asking for.

The idea, in plain words

Technical writing is writing whose job is to pass on technical or specialised information to a particular audience, clearly enough that they can understand it or act on it — a user manual, a research report, a proposal, a set of specifications. It's the opposite of creative or literary writing: there's no room for ambiguity, mood, or multiple interpretations. If a reader can misunderstand a technical document, the document has failed at its one job.

What counts as technical writing

Technical writing covers a range of document types, not just one format:

• Manuals — step-by-step instructions for using or operating something.
• Reports — technical reports, research reports, progress reports.
• Proposals — documents written to get approval or funding for a piece of work.
• Specifications — precise descriptions of requirements, dimensions, or standards a product or system must meet.
• White papers — in-depth explanations of a problem and a proposed solution, often used to persuade a technical or business audience.
• Professional emails and memos — when they're conveying technical or project information inside an organisation, they follow the same clarity standards as the longer documents above.

The 4 C's of technical writing

These four words are the shorthand the whole subject is built around:

• Clear — the meaning comes across in one reading, with no possible second interpretation. Short sentences, specific words, and a logical order all serve this.
• Concise — says what's needed and nothing more. Padding, repetition, and unnecessary qualifiers waste the reader's time and can bury the actual point.
• Correct — every fact, figure, unit, and technical term is accurate. In technical writing, a small factual error (a wrong number, a mislabelled diagram) can cause a real-world mistake, not just an embarrassment.
• Complete — includes everything the reader needs to understand or act on the content, with nothing important left out or assumed. A manual missing one step is a manual that fails the reader at exactly the point they needed it most.

Basic principles of good technical writing

• Know your audience — write to their actual level of background knowledge; don't over-explain to experts or under-explain to beginners.
• Use simple, precise language — plain words over fancy ones, and the exact right word over a vague approximation.
• Avoid unnecessary jargon — use technical terms only where the audience needs them and understands them; define any that might not be familiar.
• Organise logically — structure the document with clear headings and sections so a reader can follow the flow, or jump straight to the part they need.
• Use visuals where they help — a diagram, table, or chart can communicate a relationship or a set of numbers far faster than a paragraph of prose; use one whenever it genuinely does the job better than text would.
• Revise and proofread carefully — technical writing is rarely right on the first draft; careful revision is what actually produces the clarity and correctness the 4 C's demand.

Qualities needed in the writer

Good technical writing depends on the person writing it having:

• Strong subject knowledge — you can't explain clearly what you don't understand well yourself.
• Clarity of thought — clear writing starts with clear thinking; a muddled idea produces a muddled sentence no matter how carefully it's edited.
• Objectivity — reporting what is actually true/observed, not what the writer wants to be true.
• Attention to detail — technical content is full of small facts (numbers, units, steps) where a single slip changes the meaning.
• Ability to structure information for someone else — the writer has to set aside their own familiarity with the subject and organise the content the way a reader who doesn't share that context would need it, not the way it happens to live in the writer's own head.

Purpose of technical writing

The purpose of technical writing is to inform, instruct, or persuade a specific audience precisely and without ambiguity — very often so that the reader can go on to take a specific action correctly. For example, a user manual's whole purpose is to let someone operate a device correctly without needing to call anyone or ask anyone else for help; a technical proposal's purpose is to let a reviewer judge and approve a plan of work without needing to ask the writer to explain it further.

Exam questions this topic answers

Practice set, Unit 1 (technical writing) Q1: "Define technical writing and list the different types of documents it includes."

Answer plan: give the definition above (writing that conveys technical/specialised information clearly to a specific audience for a practical purpose, as opposed to creative/literary writing), then list the document types: manuals, reports, proposals, specifications, white papers, and professional emails/memos in a technical context.

Practice set, Unit 1 Q2: "State the four characteristics (4 C's) of technical writing."

Answer plan: name all four — Clear, Concise, Correct, Complete — with one line each explaining what it means in practice, using the descriptions above (clear = one reading, no ambiguity; concise = nothing unnecessary; correct = every fact/figure accurate; complete = nothing important missing).

Practice set, Unit 1 Q3: "List the basic principles of good technical writing."

Answer plan: list the six principles above — know your audience, use simple/precise language, avoid unnecessary jargon, organise logically with headings, use visuals where they help, revise and proofread carefully.

Practice set, Unit 1 Q4: "Name the qualities required for effective technical writing."

Answer plan: list the writer qualities above — strong subject knowledge, clarity of thought, objectivity, attention to detail, and the ability to structure information logically for a reader who doesn't share the writer's own context.

Practice set, Unit 1 Q5: "State the purpose of technical writing."

Answer plan: state the purpose line above — to inform, instruct, or persuade a specific audience precisely and without ambiguity, usually so they can take a specific action — then give one example to make it concrete, such as a user manual's purpose being to let someone operate a device correctly without needing to ask anyone else.

The idea, in plain words

A research report is the finished, written account of a completed piece of research — what the problem was, what was done, what was found, and what it means. Writing one well isn't just about getting the content right; it's about organising it so a reader (an examiner, a supervisor, another researcher) can follow the work, trust it, and reuse it.

Types of research reports

• Technical report — written for a specialist or scientific audience. It's formal and detailed: full methodology, complete data, statistical detail, and technical terminology the audience is assumed to already understand.
• Popular report — written for a general, non-specialist audience. It's simplified: it focuses on the findings and what they mean in practice, uses minimal jargon, and usually drops or shrinks the full methodology section since that audience doesn't need (or want) that level of technical detail.

Steps in writing a research report

1. Logical analysis of the subject and a rough outline — work out the main sections the report needs and the logical order they should go in, before writing any actual prose.
2. Preparation of the final, detailed outline — turn the rough outline into a fuller, more specific plan: what exactly goes under each heading.
3. Preparation of the rough draft — write the first full version, following the detailed outline, without worrying too much yet about polish.
4. Rewriting and polishing the draft — revise for clarity, accuracy, flow, and the 4 C's (clear, concise, correct, complete) from the technical-writing topic — this step is usually done more than once.
5. Preparing the final bibliography — compile the complete, correctly formatted list of every source cited in the report.
6. Writing the final, clean draft — produce the finished version that incorporates all the revisions and the finished bibliography, ready for submission.

Layout of a research report

A standard research report is organised into sections, each with its own job for the reader:

• Title page — names the study, the author, the institution, and the date.
• Acknowledgements — credits anyone (supervisors, funders, participants) who helped the work happen.
• Table of contents — lets the reader navigate straight to any section.
• Abstract / executive summary — a short, self-contained summary of the whole report: problem, method, key findings, and conclusion, so a reader can grasp the essentials without reading further.
• Introduction — states the research problem, its background, and why it matters.
• Review of literature — summarises existing relevant work and shows the gap this study addresses.
• Methodology — explains exactly how the research was conducted: design, sample, data collection, and analysis method, in enough detail that another researcher could judge or repeat it.
• Results / findings — presents what was actually found, usually with tables, charts, or statistics, without yet interpreting what it means.
• Discussion — interprets the findings, connects them back to the research problem and existing literature, and discusses their implications.
• Conclusions and recommendations — states the final answer to the research problem and any practical recommendations that follow from it.
• References / bibliography — the full, properly formatted list of every source cited.
• Appendices — supporting material that's too detailed for the main body (raw data, full questionnaires, extra tables) but useful for a reader who wants to verify or dig deeper.

Precautions while preparing a research report

• Stay concise yet complete — include everything necessary, nothing that isn't.
• Organise logically, following the standard layout above, so the reader is never lost.
• Stay free of personal bias — report what the data actually shows, not what the researcher hoped or expected to find.
• Cite properly throughout — every claim or data point drawn from elsewhere needs a citation (this directly overlaps with avoiding plagiarism, covered in topic 14).
• Proofread thoroughly — careless errors undermine the reader's trust in the report's accuracy.
• Write for the actual intended audience — match the level of technical detail to who will actually read it (technical vs. popular report, as above).
• Never state a generalisation the data doesn't actually support — conclusions should stay within what the evidence genuinely justifies.

Exam questions this topic answers

Practice set, Unit 1 Q6: "Identify the different types of technical reports and name their general outline."

Answer plan: identify the two report types — technical report (specialist audience, full methodology and data) and popular report (general audience, findings-focused, minimal jargon) — then give the general outline shared by both: title page, acknowledgements, table of contents, abstract/executive summary, introduction, review of literature, methodology, results, discussion, conclusions and recommendations, references, appendices (noting a popular report typically shrinks or drops the full methodology section).

Practice set, Unit 1 Q7: "Explain the different steps involved in writing a research report."

Answer plan: walk through the six steps above in order — logical analysis and rough outline, final detailed outline, rough draft, rewriting and polishing, final bibliography, final clean draft — explaining briefly what each step accomplishes and why the order matters (you can't polish a draft that doesn't exist yet; you can't finalise a bibliography before you know what's actually been cited in the final text).

Practice set, Unit 1 Q8: "Describe the layout of a research report and outline the contents included in each section."

Answer plan: go through the full layout above, section by section, giving one line on what each section specifically contains: title page (study name, author, institution, date), acknowledgements (credits to helpers/funders), table of contents (navigation), abstract (self-contained summary), introduction (problem and background), review of literature (existing work and the gap), methodology (design, sample, data collection, analysis), results (what was found), discussion (interpretation and implications), conclusions and recommendations (final answer and practical suggestions), references (full source list), appendices (supporting detail).

Practice set, Unit 1 Q9: "Discuss the statement 'Report writing is more an art that hinges upon practice & experience,' and narrate the various steps involved in writing such a report."

Answer plan: agree with the statement and explain why — the structural rules (layout, citation format) can be taught in an afternoon, but the actual judgement a good report needs (what to include versus leave for an appendix, how to phrase a finding precisely without overstating it, how to balance completeness against concision) only sharpens with repeated practice, feedback, and experience — two researchers can follow the identical outline and produce reports of very different quality. Then narrate the six writing steps above as the "steps involved."

Practice set, Unit 1 Q10: "Explain the different types of reports and describe the general outline of a technical report."

Answer plan: explain both report types (technical vs. popular, as above), then focus specifically on the technical report's outline: title page, acknowledgements, table of contents, abstract, introduction, review of literature, full methodology, results, discussion, conclusions and recommendations, references, appendices — emphasising that the technical report keeps the full methodology and technical detail that a popular report would simplify away.

Practice set, Unit 1 Q11: "Describe the difference between a technical report and a popular report, giving their respective outlines."

Answer plan: give a direct comparison across: audience (specialists vs. general readers), depth of methodology/data shown (full and detailed vs. minimal or summarised), language/jargon level (technical terminology assumed vs. plain language), and typical outline differences (a popular report often drops or shrinks the full methodology section and leads with findings/implications instead, while the technical report keeps the complete standard layout). Then give each report's own outline side by side using the layout list above, trimmed for the popular report.

Practice set, Unit 1 Q12: "Discuss the precautions a researcher should keep in mind while preparing a good research report."

Answer plan: discuss each precaution above — stay concise yet complete, organise logically following the layout, stay free of personal bias, cite properly throughout, proofread thoroughly, write for the actual audience, and never generalise beyond what the data supports — giving one sentence on why each matters for the report's quality and credibility.

The idea, in plain words

A research proposal is written before the actual research begins. Its job is to convince a supervisor, an academic committee, or a funding body that the proposed study is worth doing, and that the researcher has a credible, workable plan for actually carrying it out. Where a research report (topic 17) documents work that's already finished, a proposal is a pitch and a plan for work that hasn't started yet.

Why developing a proposal matters

A proposal does three distinct jobs at once:

• It secures approval or funding — a supervisor, committee, or funding agency reads the proposal to decide whether to let the study proceed or to pay for it. Without a convincing proposal, the research often doesn't get the green light at all.
• It forces the researcher to clarify their own thinking before committing time — writing down the problem, objectives, and method in enough detail to convince someone else usually exposes gaps or vague spots in the researcher's own plan that would otherwise only surface halfway through the actual study, when they're much more expensive to fix.
• It becomes the working roadmap for the research itself — once approved, the proposal doesn't get thrown away; it's the document the researcher refers back to throughout the study to stay on track, and it's often what the final report is later checked against.

Key elements of a research proposal

• Problem statement — a clear statement of exactly what research problem the study will address and why it needs addressing.
• Objectives — the specific, stated goals the research aims to achieve.
• A brief literature review — enough of a summary of existing relevant work to show the gap the proposed study is meant to fill, demonstrating the researcher already knows the field.
• Proposed methodology — the plan for how the research will actually be carried out: design, sample, data collection, and analysis approach.
• Expected outcomes / significance — what the study is likely to find or contribute, and why that matters to the field, to practice, or to society.
• A timeline — a schedule showing when each stage of the research will be carried out.
• A budget, where relevant — an estimate of the costs involved, for proposals going to a funding body.
• References — the sources already cited in the literature review and elsewhere in the proposal.

How this topic connects to the rest of the unit

Developing a research proposal sits as the conceptual bridge between two neighbouring topics in this unit. It comes right after topic 17 (Writing a Research Report), which documents research that's already been done — a proposal is effectively the mirror image, written for research that's about to be done. And it leads directly into topic 19 (Format of a Research Proposal), which lays out the detailed section-by-section structure a proposal actually follows in practice. The skill of writing a good proposal also draws directly on the same technical-writing principles covered in topic 16: a proposal still has to be clear, concise, correct, and complete, written for the specific audience (the committee or funder) who will decide whether to approve it, and organised logically so that audience can quickly judge the problem, the plan, and its feasibility.

The idea, in plain words

If topic 18 was about why a research proposal exists and what it needs to contain, this topic is about the actual section-by-section structure a proposal follows — the specific order a reviewer expects to read things in, and what each section is there to let that reviewer judge.

The typical section order, and what each section is for

• Title — concise and descriptive, telling the reviewer at a glance what the study is about. A vague or overly clever title makes the reviewer work harder before they've even started reading.
• Introduction / Background — states the context and the problem the study addresses. Its role is to let the reviewer quickly understand why this problem, why now before they get into any detail.
• Objectives / Research questions — the clear, measurable goals the research aims to achieve. Its role is to give the reviewer a concrete standard to judge the rest of the proposal against: does the proposed methodology actually look capable of achieving these stated objectives?
• Literature review — a brief review demonstrating the gap the study will address. Its role is to show the reviewer the researcher already knows the existing field well enough to be sure this isn't just repeating settled work.
• Research methodology — the design, sample, data collection, and analysis plan. This is the section that specifically exists so a reviewer can judge whether the plan is actually feasible and rigorous, not just whether the topic sounds interesting — it's usually the section a strong reviewer scrutinises hardest.
• Significance / expected contribution — states what the study is expected to find or contribute, and why that matters. Its role is to justify, for the reviewer, that the result (if achieved) is actually worth the resources being requested.
• Timeline / work plan — often shown as a schedule or Gantt-style chart. Its role is to let the reviewer judge whether the proposed work can realistically be completed in the time available or requested.
• Budget, where relevant — its role is to let a funding body judge whether the requested resources match the scope of the proposed work.
• References / Bibliography — the full list of sources cited, letting the reviewer verify the literature review's claims and see the proposal is properly grounded in existing work.

Why the format itself matters

The whole point of following a standard, predictable format is that it lets a reviewer evaluate a proposal efficiently and fairly — they know exactly where to look to check the problem, the objectives, the feasibility of the method, and the realism of the timeline, rather than having to hunt for that information inside an unstructured document. A proposal that skips or buries one of these sections (most commonly a weak or vague methodology section) is the most common reason a proposal gets rejected or sent back for revision, even when the underlying research idea itself is sound.

The idea, in plain words

A written proposal, report, or thesis is one thing — but it also usually has to be presented out loud, for example in a viva or a defense, and that oral presentation gets assessed separately from how the written document itself is graded. A researcher can write an excellent report and still present it poorly, or vice versa; this topic is specifically about what makes the presentation of the work good, not the document.

Typical assessment criteria for a presentation

• Clarity of delivery — can the audience actually follow what's being said, in real time, without the speaker's own notes to lean on? This mirrors the "Clear" principle from technical writing (topic 16), just applied to spoken delivery instead of a written page.
• Depth of subject knowledge shown, especially when answering questions — a presenter who only knows their prepared slides, and can't engage with a question that goes slightly off-script, demonstrates shallower understanding than one who can.
• Quality and relevance of visual aids — slides, diagrams, and charts should support and clarify what's being said, not distract from it or repeat it word-for-word; this is where the technical-writing principle of using visuals only where they genuinely communicate better than prose (topic 16) directly carries over into presentation design.
• Logical structure / organisation of the talk — the presentation should move through a clear, sensible order rather than jumping around, so the audience always knows where they are in the argument.
• Time management — staying within the allotted time while still covering everything important; running out of time partway through the findings, or finishing with ten minutes of silence left, both count against the presenter.
• Ability to justify and defend methodology choices under questioning — a panel will usually probe why a particular method, sample, or analysis was chosen; being able to defend those choices convincingly is often weighted as heavily as the original presentation itself.
• Audience / panel engagement — maintaining the audience's attention and responding naturally to their reactions and questions, rather than reading a script at them.
• Composure in handling unexpected questions — staying calm and thinking clearly under an unanticipated or difficult question, rather than freezing or becoming defensive.

Tying the presentation back to the written work

An effective presentation doesn't invent a separate narrative from the underlying proposal or report — it mirrors the same structure: context → problem → methodology → findings (or expected outcomes, for a proposal) → conclusion. The presentation is essentially the same argument the written document makes, delivered live and compressed to fit the time available, so a presenter who knows their own written document well should already know the shape their talk needs to take.