Chapter 1 – Introduction to Health Research

1️⃣Core Concepts (Detailed)

Definition of Research – multiple viewpoints:

 Dictionary – careful investigation to find new facts.
 OECD – creative work done systematically to increase knowledge and create new
applications.
 Kerlinger – systematic, controlled, empirical investigation of hypotheses about
relations between phenomena.
 Theobald Smith – mindset of constant re-examination of assumptions.
 Redman & Mory – systematised efforts to gain new knowledge.
 Clifford Woody – defining/redefining problems → hypothesis → data → analysis →
conclusions → testing.

Discovery vs Invention:
 Discovery = finding something that exists but was unknown (can be by chance or
design).
o Chance I – blind luck (e.g., rare patient appears)
o Chance II – belief + action to prove (e.g., Ehrlich’s Salvarsan for syphilis)
o Chance III – clue + prepared mind (e.g., Fleming notices mould → penicillin)
o Chance IV – persistence + lateral thinking
 Invention = creating something that didn’t exist before (drug, machine).

Characteristics of good research:

 Clear problem statement
 Cause–effect relationship explored
 Generates principles/theories
 Evidence-based (observations, experiments, empiricism)
 Deep subject knowledge
 Objective & logical
 Careful recording/reporting
 Requires patience, no rushing
 Avoid personal bias

Objectives (Why we do research):

 Add to scientific knowledge
 Improve practice & patient outcomes
 Inform planning/programs
 Develop better diagnostic tests
 Support managerial decisions
 Encourage logical thinking habits

Types of Research:
 By purpose: Basic (fundamental), Applied, Clinical Trials
  OECD classification: Pure basic, Strategic basic, Applied, Experimental development
 By approach/nature: Historical, Correlation, Causal-comparative, Content analysis,
Quantitative, Qualitative, Biomedical, Behavioural, Health service, Survey,
Conceptual, One-time/Longitudinal, Field, Laboratory, Simulation, Translational.

Good research criteria:

 Clear purpose
 Detailed plan & design
 Honest, transparent reporting
 Appropriate data collection & analysis
 Systematic, logical, replicable

Mnemonic: “Curious Cats Chase All Interesting Rabbits”

(Clear, Cause–effect, Careful, Accurate, Impartial, Replicable)

Active Recall Q&A

1. Which “Chance” discovery type does Fleming’s penicillin belong to?
2. Give 2 reasons why research is important in healthcare.
3. Name 3 differences between Basic and Applied research.
4. What is translational research?
5. List 4 characteristics of a good research project.

Study Activity:
📝 Create a “Type of Research” Map – start with “Research” in the centre, branch to Purpose,
Approach, Area. Fill each with examples from healthcare.

Chapter 2 – Identification & Prioritization of Research

Problems
2️⃣Core Concepts (Detailed)

When is research needed? – 3 conditions:

1. Gap between current & ideal situation
2. Reasons for gap unclear
3. More than one possible answer/solution

Sources for research ideas:

 Personal experiences
 Literature
 Existing theories
 Previous studies

FINER criteria:
 Feasible – technically, financially, administratively possible
  Interesting – engages researcher & stakeholders
 Novel – fills knowledge gap, not reinventing wheel
 Ethical – no harm, respects rights
 Relevant – important to local/national priorities

Prioritisation criteria (expanded):

1. Relevance – magnitude, severity, affected population, perception
2. Avoid duplication – review if it’s been done before
3. Urgency – time-sensitive needs for decisions/interventions
4. Feasibility – resources, location, subject access
5. Applicability – likelihood results will be implemented
6. Ethical acceptability – consent, cultural sensitivity, benefits to participants
7. Political acceptability – local/national authority support

Nominal Group Technique (NGT) steps:

1. Individual idea listing
2. Display lists & discuss
3. First voting & ranking
4. Summarise results
5. Discuss again
6. Second voting & re-ranking

Mnemonic for prioritisation: “Real Ants Use Funny Actions Every Picnic”
(Relevance, Avoid duplication, Urgency, Feasibility, Applicability, Ethical, Political)

Active Recall Q&A:

1. What does “Novel” mean in FINER?
2. Why is political acceptability important in research topic selection?
3. Name 3 examples of research urgency.
4. What’s the difference between “Relevance” and “Applicability”?
5. List the steps in NGT in correct order.

Study Activity:
📊 Rating Scale Practice – Pick 3 possible research topics in public health. Score each on the 7
criteria (1–3 scale). Select the highest scorer.

Chapter 3 – Literature Search

3️⃣Core Concepts (Detailed)

Purpose:
 Avoid duplication
 Learn existing findings
 Understand methodologies used
 Justify your study
Sources of information:
 Books, monographs
 Research surveys/reports (government & NGO)
 Journal articles (original, review)
 Clinical guidelines, protocols
 International agencies (WHO, UNICEF, FAO, etc.)
 Online databases

Key databases:
 Medline/PubMed – global biomedical database
 EMBASE – more European + drug focus
 Cochrane Library – systematic reviews
 NLM MEDLARS – specialised medical databases
 WHO WHOSIS – international health statistics

Search strategy:
1. Formulate clear question
2. Identify keywords & synonyms
3. Use Boolean operators (AND, OR, NOT)
4. Use MeSH terms in PubMed
5. Apply filters (date, language, type)
6. Review & refine search

Other tools:
 Web directories (Yahoo Health, DMOZ Health)
 Google advanced search – limit to site:, filetype:, date range
 Grey literature (theses, conference abstracts, NGO reports)

PubMed search tips:

 Automatic term mapping
 Search field tags ([mh], [au], [pt], etc.)
 Subsets for specific topics (e.g., complementary medicine)
 History & Clipboard functions for saving searches

Mnemonic for sources: “Big Red Journals Give Insight Daily”

(Books, Reports, Journals, Guidelines, Internet, Databases)

Active Recall Q&A:

1. Name 3 advantages of using review articles in literature search.
2. Which database has stronger coverage of drug-related studies?
3. What’s the difference between AND, OR in Boolean searching?
4. Give 2 examples of grey literature sources.
5. What is MeSH and why is it used?

Study Activity:
💻 Live Search Drill – Take one research question and run it through:
 PubMed with MeSH
 Google Scholar
 Cochrane Library
Compare overlap in first 10 results.

Alright — let’s keep the same detailed + active learning format and move on to Chapters 4–6.
These cover Formulating objectives & hypotheses, Planning measurements, and Study
designs.

Chapter 4 – Formulation of Objectives, Research

Questions & Hypotheses
1️⃣Core Concepts (Detailed)

Research Objectives
 General objectives – broad aims of the study (overall purpose).
 Specific objectives – concrete, measurable steps to achieve the general objective.
o Often begin with action verbs: “To determine…”, “To compare…”, “To
assess…”.
 Should be SMART:
o Specific
o Measurable
o Achievable
o Relevant
o Time-bound

Research Questions
 Translate the problem into answerable questions.
 Guide study design, methods, and analysis.
 Should be clear, focused, and feasible.

Hypotheses
 Tentative statement predicting a relationship between variables.
 Null hypothesis (H₀) – assumes no relationship/effect.
 Alternative hypothesis (H₁) – assumes a relationship/effect exists.

Characteristics of a good hypothesis:

 Clear, specific, testable, measurable.
 Based on existing theory/literature.
 Limited in scope (so it’s manageable).

Types of hypotheses:
 Simple – relationship between 2 variables.
  Complex – relationship between >2 variables.
 Directional – predicts direction of relationship (positive/negative).
 Non-directional – predicts relationship but not its direction.
 Associative vs Causal – relationship vs cause–effect.
 Null – no relationship.
 Statistical – formulated for statistical testing.

Mnemonic for hypothesis types: “Some Cats Dance Near A New Street”
(Simple, Complex, Directional, Non-directional, Associative, Null, Statistical)

Active Recall Q&A

1. What is the main difference between a research objective and a research question?
2. Define null and alternative hypothesis.
3. Give an example of a directional hypothesis.
4. What does SMART stand for in objective setting?
5. Which type of hypothesis predicts only the existence of a relationship without
specifying direction?

Study Activity:
✍️Write one general objective, two specific objectives, and one hypothesis for a study on
childhood malnutrition in rural areas.

Chapter 5 – Planning the Measurements

2️⃣Core Concepts (Detailed)

Measurement – assigning numbers or labels to variables according to rules.

Scales of measurement:
 Nominal – categories without order (e.g., gender, blood type).
 Ordinal – ordered categories but intervals not equal (e.g., pain scale).
 Interval – ordered, equal intervals, no true zero (e.g., temperature in °C).
 Ratio – ordered, equal intervals, true zero (e.g., height, weight).

Validity – degree to which a tool measures what it intends to measure.

 Types:
o Content validity – covers full concept.
o Criterion validity – correlates with gold standard.
o Construct validity – matches theoretical concepts.

Reliability – consistency of a measurement.

 Types:
o Test–retest reliability – same results over time.
o Inter-rater reliability – agreement between observers.
o Internal consistency – items measure same concept.
Threats to validity (examples):
 Selection bias, measurement bias, confounding, observer bias, recall bias.

Strategies to deal with threats:

 Standardise procedures
 Train data collectors
 Calibrate instruments
 Use blinding where possible

Mnemonic for scales: “NOIR” (Nominal, Ordinal, Interval, Ratio)

Active Recall Q&A

1. Give one example each of nominal, ordinal, interval, and ratio data.
2. What is the difference between validity and reliability?
3. Which type of validity compares a tool to a gold standard?
4. Name two threats to measurement validity.
5. What is test–retest reliability?

Chapter 6 – Study Design Options in Medical & Health

Research
3️⃣Core Concepts (Detailed)

Broad types of study designs:

 Observational (non-experimental) – researcher does not intervene.
 Experimental (interventional) – researcher assigns intervention.

Observational designs:
 Descriptive – describe occurrence (case report, case series, cross-sectional).
 Analytical – explore associations/causes (case-control, cohort).

Case-control studies – start with outcome, look back for exposures.

 Pros: quick, inexpensive, rare diseases.
 Cons: recall bias, selection bias.

Cohort studies – start with exposure, follow for outcomes.

 Pros: temporal relationship, multiple outcomes.
 Cons: costly, time-consuming.

Cross-sectional studies – measure exposure and outcome at one point in time.

 Pros: cheap, quick, prevalence data.
 Cons: no cause–effect inference.
Nested case-control – case-control within a cohort.

Experimental designs:
 Randomized Controlled Trials (RCTs) – gold standard for intervention efficacy.
 Types: parallel, crossover, factorial, cluster.

Principles for RCTs:

 Random allocation, blinding, control group, adherence monitoring.

Clinical trials – special type of RCT for drugs/therapies; phases I–IV.

Mnemonic for observational studies: “Cool Cats Cross Neighborhoods”

(Case-control, Cohort, Cross-sectional, Nested case-control)

Mnemonic for RCT principles: “Really Big Cats Are Cool”

(Randomisation, Blinding, Control group, Adherence, Compliance monitoring)

Active Recall Q&A

1. Which study design is best for rare diseases?
2. What’s the main limitation of a cross-sectional study?
3. Name two advantages of cohort studies.
4. What are the four phases of clinical trials?
5. Which type of trial gives each participant both interventions in sequence?

Study Activity:
🗂 Make a comparison table of Case-control vs Cohort vs Cross-sectional with headings:
Direction, Timeframe, Pros, Cons, Best use.

Chapter 7 – Research on Diagnostic Tests

1️⃣Core Concepts (Detailed)

Purpose:
 Evaluate accuracy, reliability, and clinical usefulness of a diagnostic test.
 Compare new test to gold standard (reference test).

Key definitions:
 True Positive (TP): Test correctly identifies disease.
 True Negative (TN): Test correctly identifies absence of disease.
 False Positive (FP): Test says disease is present, but it’s not.
 False Negative (FN): Test says no disease, but it’s there.
Accuracy measures:
 Sensitivity: TP / (TP + FN) → ability to detect disease when present.
 Specificity: TN / (TN + FP) → ability to correctly identify healthy individuals.
 Positive Predictive Value (PPV): TP / (TP + FP) → probability person with positive
test has disease.
 Negative Predictive Value (NPV): TN / (TN + FN) → probability person with
negative test is disease-free.

Other measures:
 Likelihood ratios – combine sensitivity & specificity.
 ROC curve – plots sensitivity vs 1-specificity; area under curve (AUC) indicates
accuracy.

Bias sources in diagnostic studies:

 Spectrum bias: Tested group doesn’t represent full range of disease.
 Verification bias: Not all patients get gold standard test.
 Observer bias: Knowledge of results influences interpretation.

Mnemonic for accuracy measures: “SnNout, SpPin”

 High Sensitivity → Negative result rules OUT disease.
 High Specificity → Positive result rules IN disease.

Active Recall Q&A

1. Define sensitivity and specificity.
2. What is PPV?
3. Which type of bias occurs if not all subjects get the gold standard?
4. What does the area under the ROC curve represent?
5. How does “SnNout” help remember clinical decision-making?

Study Activity:
📊 Create a 2×2 table with 50 TP, 30 TN, 10 FP, 10 FN. Calculate sensitivity, specificity,
PPV, and NPV.

Mind Map – Diagnostic Test Research

(I’ll make this visual when we compile your final PDF guide)
 Gold standard → TP, TN, FP, FN → Sensitivity, Specificity, PPV, NPV
 Bias → Spectrum, Verification, Observer
 Evaluation tools → Likelihood ratios, ROC curve

Chapter 8 – Qualitative Research

2️⃣Core Concepts (Detailed)

Purpose:
 Explore behaviours, beliefs, experiences, social contexts.
  Understand “why” and “how” questions.

Key features:
 Data = words, images, observations (not numbers).
 Flexible, open-ended design.
 Seeks depth over breadth.

Main methods:
 In-depth interviews – one-on-one, open-ended.
 Focus groups – guided discussion with 6–12 people.
 Participant observation – researcher immersed in setting.
 Document analysis – written/visual materials.

Sampling:
 Purposive (selecting info-rich cases), snowball, convenience.
 No fixed sample size → until data saturation reached.

Data analysis:
 Thematic analysis – identify patterns/themes.
 Grounded theory – build theory from data.
 Content analysis – count occurrences of categories.

Validity in qualitative research:

 Credibility (truth value)
 Transferability (applicability)
 Dependability (consistency)
 Confirmability (neutrality)

Ethics:
 Informed consent, confidentiality, cultural sensitivity.

Mnemonic for trustworthiness: “Cats Take Delicious Cookies”

(Credibility, Transferability, Dependability, Confirmability)

Active Recall Q&A

1. Name two qualitative data collection methods.
2. What is meant by data saturation?
3. Which trustworthiness criterion refers to neutrality?
4. What’s the difference between thematic and content analysis?
5. Give one advantage of focus groups.

Study Activity:
🗂 Take a transcript of a short interview and identify at least three emerging themes.

Mind Map – Qualitative Research

  Purpose → explore behaviours/contexts
 Methods → Interviews, Focus groups, Observation, Documents
 Sampling → Purposive, Snowball
 Analysis → Themes, Grounded theory, Content analysis
 Trustworthiness → Credibility, Transferability, Dependability, Confirmability

Chapter 9 – Variables
3️⃣Core Concepts (Detailed)

Definition:
 Any attribute, quality, or quantity that can vary.

Types:
1. By role in study:
o Independent variable – manipulated/causal factor.
o Dependent variable – outcome of interest.
o Confounding variable – affects relationship between independent & dependent
variables.
2. By nature:
o Qualitative (categorical) – nominal, ordinal.
o Quantitative (numerical) – discrete, continuous.
3. By measurement scale:
o Nominal, Ordinal, Interval, Ratio (NOIR).

Operational definition:
 Exact way a variable is measured or categorized in the study.

Examples:
 Independent: Diet type.
 Dependent: Blood sugar level.
 Confounder: Physical activity.

Mnemonic for variable types: “I Depend on Cats”

(Independent, Dependent, Confounding)

Active Recall Q&A

1. What’s the difference between independent and dependent variables?
2. Give an example of a confounding variable in a smoking–lung cancer study.
3. Which scale of measurement is blood pressure?
4. What is an operational definition?
5. Name two types of quantitative variables.

Study Activity:
📋 From a published research article, identify one independent variable, one dependent
variable, and one possible confounder.

Mind Map – Variables

 Role → Independent, Dependent, Confounding
 Nature → Qualitative (nominal, ordinal), Quantitative (discrete, continuous)
 Measurement → NOIR
 Operational definition → how measured in study

Chapter 10 – Sampling Methods

1️⃣Core Concepts (Detailed)

Definition:
 Process of selecting a subset of a population to represent the whole.

Why sample?
 Saves time, money, resources.
 Allows more in-depth data collection.
 Feasible for large populations.

Population vs Sample:
 Population – entire group of interest.
 Sample – subset selected from the population.
 Sampling frame – list or method from which the sample is drawn.

Sampling methods:
1. Probability sampling (random, equal chance):
o Simple random – every individual has equal chance; use random numbers,
lottery.
o Systematic – select every kth element from a list after random start.
o Stratified – divide population into strata (e.g., age, gender) and sample from
each.
o Cluster – select groups/clusters first, then individuals within.
o Multistage – combination of methods in stages.
2. Non-probability sampling (non-random, unequal chance):
o Convenience – easiest to access participants.
o Purposive – selected for specific characteristics.
o Quota – fixed number from each category.
o Snowball – existing participants recruit others.

Sample size determination:

 Depends on study design, variability, precision needed, confidence level, and power.
 Larger samples → more accurate estimates (up to a point).
Mnemonic for probability sampling: “Some Students Study Class Methods”
(Simple random, Systematic, Stratified, Cluster, Multistage)

Mnemonic for non-probability: “Cool People Quit Slowly”

(Convenience, Purposive, Quota, Snowball)

Active Recall Q&A

1. What’s the main difference between probability and non-probability sampling?
2. Give one advantage of stratified sampling.
3. What is a sampling frame?
4. Name two situations where snowball sampling is useful.
5. Which method selects every kth element?

Study Activity:
📋 For a study on childhood asthma in a city, choose a sampling method and justify your
choice in 2–3 sentences.

Mind Map – Sampling

 Purpose → represent population, save time
 Probability → Simple, Systematic, Stratified, Cluster, Multistage
 Non-probability → Convenience, Purposive, Quota, Snowball
 Sample size → design, variability, confidence, power

Chapter 11 – Data Collection Tools

2️⃣Core Concepts (Detailed)

Why tools matter:

 Ensure accuracy, consistency, and comparability of data.

Common data collection tools:

1. Questionnaires – structured, semi-structured, open-ended.
2. Interviews – face-to-face, phone, online.
3. Observation checklists – direct or participant observation.
4. Measurement instruments – scales, laboratory tests, devices.
5. Document review – patient records, reports, policy documents.

Questionnaire design tips:

 Keep language simple and clear.
 Logical flow (general → specific).
 Avoid leading or double-barrelled questions.
 Pre-test (pilot) to check understanding and timing.

Interview tips:
  Use an interview guide.
 Build rapport.
 Avoid judgmental tone.
 Record responses accurately.

Observation tips:
 Define what you are looking for.
 Use standard forms.
 Train observers for consistency.

Pre-testing/Piloting:
 Helps identify confusing questions, technical problems, and estimate time required.

Mnemonic for designing questionnaires: “Short Lovely Apples Prefer Sunshine”

(Simple language, Logical flow, Avoid bias, Pilot test, Short length)

Active Recall Q&A

1. Name three types of data collection tools.
2. Why is piloting a questionnaire important?
3. Give one advantage and one disadvantage of interviews.
4. What’s the main difference between structured and semi-structured questionnaires?
5. Why train observers before data collection?

Study Activity:
📝 Write 5 questions for a survey on physical activity habits. Ensure they follow good design
principles.

Mind Map – Data Collection Tools

 Tools → Questionnaires, Interviews, Observation, Measurement, Documents
 Design principles → Simple, Logical, Avoid bias, Pilot
 Pre-test → identify problems, estimate timing

Chapter 12 – Data Analysis

3️⃣Core Concepts (Detailed)

Purpose:
 Organise, summarise, and interpret data to answer research questions.

Two main types:

1. Descriptive analysis – summarises data (mean, median, mode, range, SD, frequencies,
percentages, charts).
2. Inferential analysis – makes predictions or inferences about a population (hypothesis
testing, confidence intervals, regression).
Common descriptive statistics:
 Measures of central tendency: mean, median, mode.
 Measures of dispersion: range, variance, standard deviation.
 Data presentation: tables, bar charts, pie charts, histograms.

Inferential statistics examples:

 t-test – compare means between 2 groups.
 ANOVA – compare means across >2 groups.
 Chi-square test – association between categorical variables.
 Correlation – strength of relationship between variables.
 Regression – predict outcome from predictor variables.

Key concepts in hypothesis testing:

 P-value – probability of observing results by chance.
 Confidence interval (CI) – range where true value likely lies.
 Type I error (α) – rejecting null when it’s true.
 Type II error (β) – failing to reject null when it’s false.

Mnemonic for hypothesis errors: “I Accept, II Ignore”

(Type I = Accepting false positive, Type II = Ignoring true effect)

Active Recall Q&A

1. What’s the difference between descriptive and inferential statistics?
2. Give one example of a measure of central tendency and one of dispersion.
3. Which test compares means between three groups?
4. What is a Type I error?
5. What does a 95% CI represent?

Study Activity:
📊 Take a small dataset (even 10–15 entries). Calculate mean, median, range, and make a bar
chart.

Mind Map – Data Analysis

 Descriptive → mean, median, mode, range, SD, charts
 Inferential → t-test, ANOVA, chi-square, correlation, regression
 Hypothesis testing → p-value, CI, Type I/II errors

Chapter 1 – Introduction to Health

Research
(In-depth study guide)
1️⃣What is Research?
Definition:
Systematic, logical, and objective process of collecting, analyzing, and interpreting
information to answer a question or solve a problem.
 It adds to existing knowledge.
 It guides action in healthcare.

Two key ways knowledge develops:

1. Discovery – finding something that already exists but was unknown.
o Chance I: Pure accident (e.g., Alexander Fleming’s penicillin discovery).
o Chance II: Determined pursuit to prove an idea.
o Chance III: Recognising a clue due to prior knowledge/preparedness.
o Chance IV: Persistence + lateral thinking.
2. Invention – creating something entirely new (drug, device, diagnostic method).

💡 Memory hook:
Think D for Discovery = Detecting what’s already there,
and I for Invention = Inventing something new.

2️⃣Characteristics of Good Research

 Clearly defined purpose.
 Based on logical reasoning.
 Systematic and step-by-step.
 Based on reliable data (not assumptions).
 Objective, free from personal bias.
 Reproducible — others can repeat and get similar results.

💡 Mnemonic: “Clear Logic Supports Real Objective Results”

(Clear purpose, Logical reasoning, Systematic, Reliable data, Objective, Reproducible)

3️⃣Objectives of Health Research

 Add to scientific knowledge.
 Improve healthcare practice.
 Influence policy and planning.
 Evaluate healthcare delivery.
 Advance diagnostic and treatment methods.
 Promote logical, critical thinking.

4️⃣Types of Research
By purpose:
 Basic research – builds fundamental knowledge (lab studies, theory-building).
  Applied research – solves practical, real-world problems.

By setting:
 Field research – in community/real-life settings.
 Laboratory research – controlled conditions.

By data type:
 Quantitative – numbers/statistics.
 Qualitative – words, descriptions, meanings.

By time sequence:
 Cross-sectional – one point in time.
 Longitudinal – follow over time.

By intervention:
 Observational – no intervention, just observation.
 Experimental – researcher manipulates variables.

Clinical Trials – test interventions in human participants before general use.

 Phases I–IV (safety → efficacy → comparison → post-marketing surveillance).

💡 Mnemonic for main types: “Big Apples Can Have Lovely Orange Cakes”
(Basic, Applied, Clinical, Historical, Longitudinal, Observational, Cross-sectional)

5️⃣Major Areas in Health Systems Research

1. Health policy (laws, regulations).
2. Health environment (sanitation, pollution).
3. Health administration (management, workforce).
4. Community health (epidemiology, prevention).
5. Individual & family health needs.
6. Direct healthcare services (treatment, rehabilitation).

💡 Mnemonic: “Please Eat And Cook In Durban”

(Policy, Environment, Administration, Community, Individual, Direct services)

6️⃣The Research Process (Overview)

1. Identify & define the problem.
2. Review literature.
3. State objectives/hypothesis.
4. Choose study design.
5. Select sample.
6. Collect data.
7. Analyse data.
8. Interpret & report results.
💡 Visual trick: Imagine a ladder — each rung is one step. You climb from “problem” to
“publication.”

7️⃣Ethics in Health Research

(introduced here, expanded in later chapters)

 Respect for persons (informed consent).
 Beneficence (maximize benefits, minimize harm).
 Justice (fair selection of subjects).

8️⃣Common Challenges in Health Research

 Lack of funding.
 Poor study design.
 Insufficient sample size.
 Ethical issues.
 Data management errors.

Active Recall Questions

1. Differentiate between discovery and invention in research.
2. List 6 characteristics of good research.
3. Give one example each of basic and applied research in healthcare.
4. Name the 6 major areas of health systems research.
5. Which phase of a clinical trial focuses on safety?

Study Activities
 Activity 1: Choose a medical innovation (e.g., MRI machine) and classify it:
discovery or invention? Justify.
 Activity 2: Draw the Research Process Ladder and label each step in your own words.
 Activity 3: Make a mind map linking the types of research with real-life examples in
your country’s healthcare.

Mind Map – Chapter 1

(will be visualised in the final compiled PDF)
 Definition → Systematic, logical, objective
 Discovery → Chance I–IV
 Invention → new creation
 Characteristics → Clear, logical, systematic, reliable, objective, reproducible
 Objectives → Knowledge, practice, policy, evaluation, diagnostics, thinking
 Types → Basic, Applied, Clinical, Historical, Observational, Quantitative,
Qualitative, Cross-sectional, Longitudinal
  Major areas → Policy, Environment, Administration, Community, Individual, Direct
services

Chapter 2 – Identification & Prioritization

of Research Problems
1️⃣What is a Research Problem?
Definition:
A question, issue, or knowledge gap that a study aims to address through systematic
investigation.

💡 Think of it as the starting point of the research journey — without a clearly defined
problem, the rest of the study lacks direction.

2️⃣Sources of Research Problems

Research problems can arise from:
1. Personal experience – issues observed in your own practice.
2. Professional literature – gaps or contradictions in published work.
3. Theory – testing, refining, or expanding existing theories.
4. Society/community needs – pressing public health concerns.
5. Government policy & health priorities – aligned with national or global goals.
6. Technological developments – need for evaluation or improvement.

💡 Mnemonic: “Please Let The Students Go Teach”

(Personal experience, Literature, Theory, Society needs, Government priorities, Technology)

3️⃣Criteria for a Good Research Problem

A good problem should be:
 Relevant – important to health or policy.
 Novel – fills a gap in knowledge.
 Feasible – can be studied within time, resources, and skills available.
 Ethical – respects rights and safety of participants.
 Clear & Specific – well-defined and not too broad.

💡 Mnemonic: “Really Nice Friendly Ethical Cats”

(Relevant, Novel, Feasible, Ethical, Clear)
4️⃣Identifying a Research Problem – Step-by-Step
1. Scan sources – literature, health statistics, community needs.
2. Select a broad topic of interest.
3. Narrow it down by:
o Population of interest
o Geographical scope
o Time frame
4. Define in clear, concise terms.
5. Check feasibility (cost, skills, logistics).

5️⃣Prioritizing Research Problems

When many possible problems exist, priority is given to those that:
 Have high relevance to health needs.
 Address urgent issues or emerging diseases.
 Benefit large numbers of people.
 Are likely to produce actionable results.
 Fit with available resources & expertise.
 Are fundable.

Common priority-setting methods:

 Hanlon Method – scores problems by size, seriousness, effectiveness of intervention,
and feasibility.
 Delphi Technique – group consensus via multiple rounds of anonymous feedback.
 Nominal Group Technique – structured group discussion and ranking.

💡 Mnemonic for priority criteria: “Really Urgent Big Action Fits Funds”
(Relevance, Urgency, Benefit, Actionable, Feasible, Fundable)

6️⃣From Problem to Research Question

 Convert problem into clear, answerable question.
 Use PICO format (especially for clinical research):
o Population
o Intervention (or exposure)
o Comparison
o Outcome

Example:
Problem → “High rates of uncontrolled diabetes in rural adults.”
PICO Question → “In rural adults with diabetes (P), does community-based lifestyle
education (I), compared to standard clinic visits (C), improve blood sugar control (O)?”

7️⃣Ethical Considerations in Choosing Problems

 Ensure potential benefits outweigh risks.
  Avoid exploitation of vulnerable populations.
 Respect cultural sensitivities.
 Maintain scientific integrity (no plagiarism, fabrication).

8️⃣Common Pitfalls in Problem Selection

 Being too broad or too narrow.
 Ignoring existing solutions in literature.
 Choosing a problem beyond available resources.
 Selecting a problem without personal or institutional interest/support.

Active Recall Questions

1. List 6 common sources of research problems.
2. What are the 5 criteria for a good research problem?
3. What does PICO stand for?
4. Which priority-setting method uses anonymous expert feedback over multiple
rounds?
5. Give an example of a problem that is relevant but not feasible.

Study Activities
 Activity 1: Think of 3 research problems in your field. Apply the Really Nice
Friendly Ethical Cats criteria and rank them.
 Activity 2: Take one of your problems and reframe it into a PICO question.
 Activity 3: Research how your country’s health department prioritises health research
topics.

Mind Map – Chapter 2

 Research Problem → starting point
 Sources → Personal, Literature, Theory, Society, Government, Technology
 Good criteria → Relevant, Novel, Feasible, Ethical, Clear
 Identification steps → Scan, Select, Narrow, Define, Check feasibility
 Prioritisation → Relevance, Urgency, Benefit, Actionable, Feasible, Fundable
 Methods → Hanlon, Delphi, Nominal group
 PICO → Population, Intervention, Comparison, Outcome

Chapter 3 – Review of Literature

1️⃣What is a Literature Review?
Definition:
A systematic search, evaluation, and synthesis of existing research related to a particular
topic or problem.
Purpose:
 Understand what is already known.
 Identify gaps and inconsistencies.
 Avoid duplication of work.
 Refine research problem and hypotheses.
 Provide context and justification for your study.

💡 Think of it as the “background check” for your research — you don’t start until you know
what’s already out there.

2️⃣Types of Literature Reviews

Type Description Example
Broad, descriptive, non-systematic “Overview of childhood asthma
Narrative
summary of literature. trends in Africa.”
Follows a clear, replicable method to “Systematic review on the efficacy
Systematic
find, select, and appraise studies. of CBT in depression.”
Meta- Combines statistical results from multiple “Pooled effect size of statins on
analysis studies. cardiovascular events.”
Scoping Maps the existing literature, often on
“Scoping review of AI in radiology.”
review emerging topics.

💡 Memory hook: “Never Start Making Soup”

(Narrative, Systematic, Meta-analysis, Scoping)

3️⃣Steps in Conducting a Literature Review

1. Define the topic & scope
o Based on your research question (use PICO if applicable).
2. Search for relevant literature
o Use databases (e.g., PubMed, Scopus, Web of Science, CINAHL, Google
Scholar).
o Use keywords, Boolean operators (AND, OR, NOT), and MeSH terms.
3. Screen & select studies
o Apply inclusion/exclusion criteria.
o Remove duplicates.
o Prioritise recent and high-quality studies.
4. Extract data
o Study details, sample, methods, key findings, limitations.
5. Synthesize & organise
o Group by themes, methods, chronology, or concepts.
6. Critically appraise
o Assess reliability, validity, bias, and generalisability.
7. Write the review
o Structure: introduction → themes → gaps → conclusion.
o Cite correctly (APA, Vancouver, etc.).
💡 Mnemonic: “Don’t Search Small Elephants Standing Calmly Waiting”
(Define, Search, Screen, Extract, Synthesise, Critique, Write)

4️⃣Searching Effectively – Tips

 Use Boolean operators:
o AND → narrows search (“diabetes AND hypertension”).
o OR → broadens search (“adolescent OR youth”).
o NOT → excludes terms (“stroke NOT hemorrhage”).
 Use truncation: child** retrieves child, children, childhood.
 Use quotation marks for exact phrases (“physical activity”).

5️⃣Critical Appraisal Skills

Ask:
 Is the study design appropriate?
 Was the sample representative?
 Are results consistent across studies?
 Were outcomes measured reliably?
 Are there conflicts of interest?

💡 Mnemonic for appraisal: “Some Small Rats Were Cute”

(Sample, Study design, Results consistency, Measurement reliability, Conflicts)

6️⃣Avoiding Common Pitfalls

 Relying only on Google.
 Ignoring grey literature (theses, reports, conference papers).
 Including outdated studies without justification.
 Failing to critically analyse (just summarising without evaluation).

7️⃣Using Citation Managers

 Tools like Mendeley, Zotero, EndNote save time and keep references organised.
 Allow quick formatting into any style (APA, Vancouver, Harvard).

Active Recall Questions

1. Name the 4 main types of literature reviews.
2. What does the Boolean operator AND do in a search?
3. Give two examples of databases used for health research.
4. What are two purposes of a literature review?
5. Which step comes right after “Extract data” in the review process?

Study Activities
  Activity 1: Pick a research topic. Use 3 databases and apply Boolean operators to find
at least 10 relevant studies.
 Activity 2: Make a table with 5 recent studies — include authors, year, design, sample
size, and main finding.
 Activity 3: Choose one article and write a 3-sentence critical appraisal.

Mind Map – Chapter 3

 Purpose → identify gaps, avoid duplication, refine question
 Types → Narrative, Systematic, Meta-analysis, Scoping
 Steps → Define, Search, Screen, Extract, Synthesise, Critique, Write
 Search tips → Boolean operators, truncation, phrases
 Critical appraisal → Sample, Design, Results, Reliability, Conflicts

Chapter 4 – Hypothesis, Objectives &

Variables
1️⃣Research Objectives
Definition:
Clear statements of what the research intends to achieve. They guide study design, data
collection, and analysis.

Types of objectives:
 General (broad aim): The overall purpose.
Example: To assess the effectiveness of a community-based
nutrition program in reducing childhood obesity.
 Specific: Narrow, focused statements breaking down the general aim into measurable
parts.
Example: To compare BMI changes in participants and non-
participants after 6 months.

💡 Rule: Each specific objective should be SMART:

 Specific
 Measurable
 Achievable
 Relevant
 Time-bound

2️⃣Hypothesis
Definition:
A clear, testable statement predicting the relationship between variables.

Characteristics of a good hypothesis:

 Clear & precise
 Testable (measurable variables)
 Based on existing evidence
 States expected relationship

Types of hypotheses:
Type Meaning Example
No relationship/difference There is no difference in blood pressure between
Null (H₀)
exists salt-restricted and normal diets.
Alternative A relationship/difference Salt restriction reduces blood pressure compared
(H₁) exists to a normal diet.
One-tailed Predicts direction Salt restriction reduces blood pressure.
States difference but no
Two-tailed Salt restriction affects blood pressure.
direction

💡 Memory hook:
Think of Null as “Nothing happening,” and Alternative as “An effect exists.”

3️⃣Relationship Between Objectives & Hypotheses

 Objectives describe what you want to do.
 Hypotheses predict what you expect to find.
 Not all studies have hypotheses (e.g., exploratory or descriptive studies may just have
objectives).

4️⃣Variables
Definition:
Any characteristic or attribute that can be measured and can vary between people or over
time.

Types:
1. By role in study:
o Independent variable (IV): The factor manipulated or categorized (cause).
o Dependent variable (DV): The outcome measured (effect).
o Confounding variable: A factor that distorts the true relationship between IV
and DV.
2. By data type:
o Qualitative (categorical): Nominal (unordered: gender, blood type), Ordinal
(ordered: pain scale).
o Quantitative (numerical): Discrete (count: number of children), Continuous
(measure: weight).
💡 Trick to remember IV vs DV:
Input → Independent, Discovered → Dependent.

5️⃣Operational Definitions
Precise description of how a variable will be measured in your study.

Example: “Obesity” will be defined as BMI ≥ 30 kg/m² measured

using a calibrated scale and stadiometer.

6️⃣Common Mistakes in Formulating Hypotheses &

Objectives
 Being too vague (“to study diabetes”).
 Mixing multiple ideas in one hypothesis.
 Using variables that can’t be measured reliably.
 Making objectives that don’t align with data collection tools.

Active Recall Questions

1. What does the SMART rule stand for in writing objectives?
2. Differentiate between null and alternative hypotheses.
3. Give an example of an independent variable in a study on exercise and blood sugar.
4. What is a confounding variable?
5. Why are operational definitions important?

Study Activities
 Activity 1: Take a health topic and write 1 general objective, 3 specific objectives,
and a matching hypothesis.
 Activity 2: Identify IV, DV, and a possible confounder for your chosen study.
 Activity 3: Write an operational definition for one of your variables.

Mind Map – Chapter 4

 Objectives → General, Specific (SMART)
 Hypotheses → Null, Alternative, One-tailed, Two-tailed
 Variables → Independent, Dependent, Confounding; Qualitative (nominal, ordinal),
Quantitative (discrete, continuous)
 Operational definitions → precise measurement description

Alright — here’s your full, detailed, exam-ready guide for:

Chapter 5 – Study Design

1️⃣What is a Study Design?
Definition:
The overall strategy and structure used to integrate the study’s components in a logical way,
ensuring you can answer your research question effectively.

💡 Think of it as the blueprint of your research — it tells you how you will collect, measure,
and analyse data.

2️⃣Why is Study Design Important?

 Determines validity of findings.
 Helps control bias.
 Maximises efficiency in time and resources.
 Allows replication by others.

3️⃣Broad Classification of Study Designs

A. Observational (no researcher-controlled intervention)
B. Experimental (researcher intervenes or manipulates)

4️⃣Observational Study Designs

1. Descriptive Studies

Purpose: Describe occurrence of a disease, behaviour, or condition without analysing cause.

Examples:
 Case report – detailed report on a single patient.
 Case series – several similar cases reported together.
 Cross-sectional study – data collected at a single point in time.
 Ecological study – uses population-level data, not individuals.

Advantages: Quick, inexpensive, good for prevalence data.

Limitations: Cannot establish causation.

2. Analytical Studies

Purpose: Investigate associations and possible causes.

Types:
 Case-control – compares people with the disease (cases) to those without (controls) to
find past exposures.
o Good for rare diseases, quick, cheap.
o Limitation: Recall bias.
  Cohort – follows a group over time to see who develops the disease.
o Good for rare exposures, can calculate incidence.
o Limitation: Time-consuming, expensive.

💡 Memory hook for observational types: “Cool Cats Can Eat”

(Case-control, Cohort, Cross-sectional, Ecological)

5️⃣Experimental Study Designs

Definition:
Researcher actively assigns exposure/intervention to study participants.

Randomised Controlled Trial (RCT)

 Participants randomly assigned to intervention or control group.
 Controls confounding, considered gold standard for causality.

Phases of clinical trials:

 Phase I: Safety and dosage (small group).
 Phase II: Efficacy and side effects (larger group).
 Phase III: Compare to standard treatment (large multi-centre).
 Phase IV: Post-marketing surveillance.

Quasi-experimental Designs
 Intervention without full randomisation.
 Example: Before-and-after study in one group without a control group.

6️⃣Special Designs
 Crossover design: Each participant receives both intervention and control at different
times.
 Factorial design: Studies effects of two or more interventions simultaneously.
 Community trial: Whole communities, not individuals, receive intervention.

7️⃣Key Design Considerations

1. Randomisation: Prevents selection bias.
2. Blinding:
o Single blind → participants don’t know allocation.
o Double blind → participants & researchers both unaware.
o Triple blind → participants, researchers, and data analysts unaware.
3. Control groups: Allow comparison to measure effect.
4. Sample size: Should be adequate to detect meaningful difference.

💡 Trick to remember blinding levels:

1 eye covered = single blind, 2 eyes covered = double blind, both eyes + mouth covered
(can’t talk about it) = triple blind.

8️⃣Strengths & Weaknesses Summary Table

Design Strengths Weaknesses
Case report/series Cheap, quick No causation, anecdotal
Cross-sectional Prevalence data Can’t show causality
Case-control Good for rare diseases, quick Recall bias, can’t calculate incidence
Cohort Good for rare exposures, incidence Expensive, slow
RCT High evidence level Costly, ethical limits
Quasi-experimental Practical Higher risk of bias

9️⃣Choosing the Right Design

Ask:
 What is my research question?
 Is my goal to describe, associate, or test causation?
 Do I have the ability to intervene?
 What are my time and budget constraints?

Active Recall Questions

1. What is the main difference between observational and experimental studies?
2. Name one advantage and one disadvantage of a case-control study.
3. Which study design is considered the gold standard for establishing causality?
4. What is the purpose of blinding in research?
5. Which study design would you choose for investigating a rare exposure?

Study Activities
 Activity 1: Choose a health problem and outline how you would study it using case-
control and cohort designs — compare differences.
 Activity 2: Write an example research question and decide if it needs a descriptive,
analytical, or experimental design.
 Activity 3: Make a flowchart that starts with “research question” and leads to the
correct study design.

Mind Map – Chapter 5

 Definition → blueprint for research
 Observational → Descriptive (case report, case series, cross-sectional, ecological),
Analytical (case-control, cohort)
 Experimental → RCT, quasi-experimental, crossover, factorial, community trial
 Key elements → Randomisation, blinding, control, sample size
 Choice factors → Question, intervention ability, resources, time
Chapter 6 – Sampling
1️⃣What is Sampling?
Definition:
The process of selecting a portion (sample) of a population so that information from the
sample can be used to draw conclusions about the whole population.

💡 Analogy: Sampling is like tasting a spoonful of soup to judge the flavour — you don’t eat
the whole pot, but you still get a good idea if it’s done right.

2️⃣Why is Sampling Important?

 Saves time and money.
 Allows research on large populations.
 Can provide accurate results if done correctly.
 Minimises effort while maximising information.

3️⃣Key Concepts in Sampling

 Population: The entire group of individuals with the characteristics of interest.
 Sample: A subset of the population chosen for the study.
 Sampling frame: The list of individuals from which the sample will be drawn.
 Element: A single member of the population.
 Sampling unit: The element actually selected during sampling.
 Representativeness: How closely the sample matches the population.

💡 Memory hook for key terms: “Please Send Sally Every Single Recipe”
(Population, Sample, Sampling frame, Element, Sampling unit, Representativeness)

4️⃣Types of Sampling Methods

A. Probability Sampling

Each element has a known chance of being selected — reduces bias.

1. Simple Random Sampling
o Every element has equal chance.
o Example: Drawing names from a hat, random number generator.
2. Systematic Sampling
o Select every kth individual from a list after a random start.
o Example: Every 10th patient entering a clinic.
3. Stratified Sampling
o Population divided into subgroups (strata) → random sample from each.
o Ensures representation of important groups (e.g., age, gender).
 4. Cluster Sampling
o Population divided into clusters (e.g., villages, schools) → random selection of
clusters → sample everyone in them or a subset.

B. Non-Probability Sampling

Chance of selection is unknown — more bias risk, but practical in many field studies.
1. Convenience Sampling
o Participants chosen based on availability.
o Example: First 50 patients visiting a clinic.
2. Purposive (Judgmental) Sampling
o Researcher chooses based on specific purpose/criteria.
3. Quota Sampling
o Like stratified, but selection within groups is non-random until a quota is
filled.
4. Snowball Sampling
o Existing participants recruit others — useful for hard-to-reach populations.

💡 Mnemonic for probability sampling: “Some Smart Students Clap”

(Simple, Systematic, Stratified, Cluster)

💡 Mnemonic for non-probability sampling: “Cool People Quit Swimming”

(Convenience, Purposive, Quota, Snowball)

5️⃣Sample Size Determination

Why it matters:
Too small → low power, risk of false negatives.
Too large → waste of resources, may find statistically significant but clinically meaningless
results.

Factors influencing sample size:

 Study design.
 Expected effect size.
 Variability in population.
 Desired confidence level (usually 95%).
 Acceptable margin of error.
 Statistical power (usually 80%).

6️⃣Sampling Errors
 Sampling error: Random differences between sample and population due to chance
selection.
 Non-sampling error: Biases from poor design, faulty measurement, or data entry
errors.
7️⃣Steps in Sampling Process
1. Define population.
2. Decide on sampling frame.
3. Choose sampling method.
4. Determine sample size.
5. Select the sample.
6. Collect data.

8️⃣Strengths & Weaknesses Overview Table

Method Strengths Weaknesses
Simple Random Unbiased, easy analysis Needs complete list of population
Systematic Easy to do Risk of periodicity bias
Requires knowledge of population
Stratified Ensures subgroup representation
structure
Cluster Cost-effective for large areas Higher sampling error
Convenience Quick, easy High bias risk
Purposive Focused Subjective
Quota Ensures quotas Non-random, bias
Snowball Finds hidden populations Bias from network effects

Active Recall Questions

1. Define “sampling frame.”
2. Name two probability and two non-probability sampling methods.
3. What is the main advantage of stratified sampling?
4. Give one situation where snowball sampling would be appropriate.
5. What does statistical power refer to in sample size determination?

Study Activities
 Activity 1: Pick a research topic and outline two different sampling methods you
could use — compare pros and cons.
 Activity 2: Using a population of 1,000 people, plan how you’d select a 100-person
sample using systematic sampling.
 Activity 3: List possible sources of sampling error in your fieldwork setting.

Mind Map – Chapter 6

 Definition → selecting subset to represent population
 Key terms → Population, Sample, Sampling frame, Element, Unit,
Representativeness
 Probability → Simple, Systematic, Stratified, Cluster
 Non-probability → Convenience, Purposive, Quota, Snowball
 Sample size factors → design, effect size, variability, confidence, margin of error,
power
 Errors → sampling, non-sampling
Chapter 7 – Data Collection Methods
1️⃣What is Data Collection?
Definition:
The process of systematically gathering information relevant to the research question, using
standardised tools and procedures.

💡 Why it matters: Good data collection ensures validity (measures what it’s supposed to) and
reliability (produces consistent results).

2️⃣Types of Data
A. Primary Data
– collected directly by the researcher for the first time

Examples: interviews, observations, questionnaires, laboratory results.

Advantages:
 Specific to your objectives.
 Up-to-date and relevant.

Disadvantages:
 Time-consuming, costly.

B. Secondary Data
– collected by others, used for a new purpose

Examples: census data, hospital records, published research.

Advantages:
 Quick, inexpensive.

Disadvantages:
 May be outdated or not match your variables.

💡 Memory hook: “Primary = Personally collected, Secondary = Someone else collected.”

3️⃣Methods of Data Collection
1. Questionnaires
 Written set of questions for self-completion or interviewer-assisted.

Types of questions:
 Closed-ended → fixed responses (yes/no, multiple choice).
 Open-ended → allow detailed, free-text answers.

Advantages:
 Standardised, easy to analyse.

Limitations:
 Literacy needed, risk of non-response.

2. Interviews
 Researcher asks questions directly.

Types:
 Structured → fixed set of questions in same order.
 Semi-structured → guided but flexible.
 Unstructured → conversational, exploratory.

Advantages:
 Allows clarification, richer data.

Limitations:
 Interviewer bias, time-consuming.

3. Observation
 Watching and recording behaviour or events.

Types:
 Participant observation → researcher is part of the group.
 Non-participant → researcher observes without involvement.

Advantages:
 Records real-life behaviour.

Limitations:
 Observer bias, Hawthorne effect (people change behaviour if watched).
4. Document/Record Review
 Analysis of existing documents (e.g., patient charts, policy reports).

Advantages:
 No burden on participants.

Limitations:
 Missing or incomplete data.

5. Physical/Lab Measurements
 Using equipment to measure variables (e.g., blood pressure, BMI, lab tests).

Advantages:
 Objective, precise.

Limitations:
 Requires equipment and trained personnel.

4️⃣Validity & Reliability

Validity: Accuracy of measurement.
 Face validity → appears to measure what it should.
 Content validity → covers all aspects of the concept.
 Construct validity → matches theoretical concepts.
 Criterion validity → correlates with a gold standard.

Reliability: Consistency of measurement.

 Test–retest reliability → stable over time.
 Inter-rater reliability → agreement between observers.
 Internal consistency → items measure same concept (Cronbach’s alpha).

💡 Mnemonic for validity types: “Funny Cats Chase Cars”

(Face, Content, Construct, Criterion)

5️⃣Pretesting & Piloting Tools

 Pretest: Try out on small group to check clarity and timing.
 Pilot study: Small-scale run of the whole study to detect problems in methods.

6️⃣Ethical Considerations in Data Collection

 Informed consent.
 Privacy and confidentiality.
  Cultural sensitivity.
 Right to withdraw without penalty.

Active Recall Questions

1. Differentiate between primary and secondary data.
2. Name two advantages and two limitations of interviews.
3. What is the Hawthorne effect?
4. Give one example each of face and criterion validity.
5. Why is a pilot study important?

Study Activities
 Activity 1: Pick a health topic and decide which data collection method would be best
— justify your choice.
 Activity 2: Draft 5 questionnaire questions — mix closed- and open-ended types.
 Activity 3: Explain how you would check the reliability of your data collection tool.

Mind Map – Chapter 7

 Types of data → Primary, Secondary
 Methods → Questionnaire, Interview, Observation, Record review, Lab measurement
 Validity → Face, Content, Construct, Criterion
 Reliability → Test–retest, Inter-rater, Internal consistency
 Pretest & pilot → small trials to refine methods
 Ethics → consent, privacy, cultural respect

Chapter 8 – Data Processing &

Presentation
1️⃣What is Data Processing?
Definition:
A sequence of steps to check, clean, code, classify, and prepare data so it’s ready for analysis.

💡 Think of it as washing, cutting, and sorting vegetables before cooking — the cleaner and
more organised they are, the better the final dish (analysis).

2️⃣Stages of Data Processing

1. Editing
 What: Checking for completeness, consistency, and clarity of collected data.
 When: Immediately after data collection.
  Example: Checking if all questionnaire items are filled in.

2. Coding
 What: Assigning numbers or symbols to responses for easier analysis.
 Example: Gender → Male = 1, Female = 2.

3. Classification
 What: Grouping data into meaningful categories.
 Example: Age groups → 0–5, 6–12, 13–18, etc.

4. Tabulation
 What: Summarising data in tables for clarity.
 Types:
o Simple table → shows one variable.
o Cross table → compares two or more variables.

💡 Mnemonic for stages: “Every Cat Can Tap”

(Editing, Coding, Classification, Tabulation)

3️⃣Data Presentation
Why it matters:
 Makes complex data clear, concise, and interpretable.
 Highlights trends, comparisons, and relationships.

A. Textual Presentation
 Writing results in paragraphs.
 Best for small datasets or emphasising specific values.

B. Tabular Presentation
 Data arranged in rows and columns.
 Includes table title, headings, and source if needed.

C. Graphical Presentation

Common types:
Chart/Graph Best for Example
Bar chart Comparing categories Number of smokers vs non-smokers
Pie chart Showing proportions % of budget by department
Histogram Distribution of continuous data Age distribution of patients
Line graph Trends over time Monthly malaria cases
Scatter plot Relationship between two variables Height vs weight correlation
💡 Trick: If your data shows change over time → line graph, if parts of a whole → pie chart, if
frequency distribution → histogram.

4️⃣Qualitative vs Quantitative Data Presentation

 Qualitative → bar charts, pie charts, thematic summaries.
 Quantitative → tables, histograms, scatter plots, line graphs.

5️⃣Guidelines for Good Presentation

 Choose the correct form (table, chart, text).
 Keep it simple and uncluttered.
 Label clearly (titles, axis names, units).
 Use appropriate scales.
 Indicate source of data.

6️⃣Common Mistakes to Avoid

 Overloading with too much data in one figure.
 Using wrong type of chart.
 Missing labels or scales.
 Misleading proportions in graphs.

Active Recall Questions

1. What are the four main stages of data processing?
2. Give an example of coding in data analysis.
3. Which chart is best to show trends over several years?
4. What’s the difference between a histogram and a bar chart?
5. Name two mistakes to avoid when presenting data.

Study Activities
 Activity 1: Take a small dataset (e.g., 10 students’ ages) and go through the four data
processing steps.
 Activity 2: Present the same dataset in three different ways — table, pie chart, and bar
graph.
 Activity 3: Find a published health study and critique one of its graphs for clarity and
accuracy.

Mind Map – Chapter 8

 Data Processing → Editing, Coding, Classification, Tabulation
 Presentation types → Textual, Tabular, Graphical
 Graph types → Bar, Pie, Histogram, Line, Scatter
 Good presentation → simple, labelled, correct scale, source noted
Chapter 9 – Data Analysis & Interpretation
1️⃣What is Data Analysis?
Definition:
The process of organising, summarising, and examining data to identify patterns,
relationships, or trends and answer the research question.

💡 Analogy: If data collection is gathering puzzle pieces, analysis is putting them together to
see the picture.

2️⃣Two Main Types of Data Analysis

A. Descriptive Statistics
–
“Describe” the data
 Purpose: Summarise and organise data.
 Examples:
o Measures of central tendency → Mean, Median, Mode.
o Measures of dispersion → Range, Variance, Standard Deviation (SD).
o Frequencies & percentages.

When to use: Early stage, to understand data distribution before deeper testing.

B. Inferential Statistics
–
“Infer” from the data
 Purpose: Make conclusions about a population based on a sample.
 Examples:
o Hypothesis testing (t-test, Chi-square, ANOVA).
o Confidence intervals.
o Correlation & regression.

When to use: When you want to test predictions or generalise findings.

💡 Memory hook: Describe → tell the story, Infer → predict the sequel.

3️⃣Steps in Data Analysis

1. Data preparation – clean and process (Chapter 8).
 2. Descriptive analysis – summarise key variables.
3. Check assumptions – normality, independence, variance equality.
4. Inferential tests – choose correct test based on data type and design.
5. Interpret results – statistical + practical significance.

4️⃣Choosing the Right Statistical Test

Data type Two groups > Two groups Relationship
Numeric (normal) t-test ANOVA Pearson correlation
Numeric (non-normal) Mann–Whitney U Kruskal–Wallis Spearman correlation
Categorical Chi-square/Fisher’s exact Chi-square Logistic regression

💡 Trick:
 “t” for two groups.
 “A” in ANOVA for above two groups.

5️⃣Data Interpretation
Statistical significance:
 Usually p < 0.05 → unlikely due to chance.
 BUT: statistically significant doesn’t always mean clinically important.

Confidence interval (CI):

 Range of values within which the true population value is likely to lie (usually 95%
CI).
 Narrow CI → more precise estimate.

6️⃣Common Analysis Errors to Avoid

 Using wrong statistical test for data type.
 Ignoring assumptions of the test.
 Focusing only on p-value without effect size.
 Over-interpreting small sample results.

7️⃣Example
Research question: Does a new diet reduce cholesterol more than a standard diet?
 Data type: Numerical (cholesterol level).
 Groups: 2 → New diet vs Standard diet.
 Test: Independent t-test (if data normal).
 Interpretation: p = 0.03 → Statistically significant difference, CI shows likely
reduction range of 5–15 mg/dL.

Active Recall Questions

1. What’s the difference between descriptive and inferential statistics?
 2. Which test would you use to compare cholesterol levels in three diet groups?
3. What does a narrow 95% CI indicate?
4. Why can a statistically significant result still be clinically unimportant?
5. Give one example of a non-parametric test.

Study Activities
 Activity 1: Take any dataset (e.g., student test scores) and calculate mean, median,
mode, and SD.
 Activity 2: Choose a research question and decide which statistical test you’d use and
why.
 Activity 3: Interpret a given p-value and CI for a mock study.

Mind Map – Chapter 9

 Definition → organising & examining data to answer questions
 Types → Descriptive (mean, median, mode, SD), Inferential (t-test, Chi-square,
ANOVA, correlation)
 Steps → prepare, describe, check assumptions, test, interpret
 Choosing tests → depends on data type & groups
 Interpretation → p-value, CI, effect size

Chapter 10 – Report Writing

1️⃣What is a Research Report?
Definition:
A structured document that communicates the objectives, methods, results, and conclusions
of a research study in a clear, logical, and concise way.

💡 Think of it as your study’s “story” — but told with evidence, not fiction.

2️⃣Purposes of a Research Report

 Share findings with the scientific community.
 Provide a record for future reference.
 Enable others to replicate or build on your work.
 Influence policy, practice, or further research.

3️⃣Types of Research Reports

 Thesis/Dissertation → Detailed, often for academic degree.
 Journal Article → Condensed for publication.
 Technical Report → Prepared for an organisation or funder.
 Conference Paper → Focused summary for presentation.
4️⃣Structure of a Research Report (IMRaD Format)
I – Introduction
 Background, rationale, problem statement.
 Literature review (brief for articles, detailed for theses).
 Objectives & hypotheses.

M – Methods
 Study design, setting, population, sampling.
 Data collection tools & procedures.
 Ethical considerations.

R – Results
 Present findings logically.
 Use tables, figures, and text together.
 No interpretation here — just facts.

a – and (this is the linking word — reminds you there’s more to come)

D – Discussion
 Interpret results, compare with other studies.
 Explain unexpected findings.
 Implications for practice and policy.
 Limitations of the study.

Conclusion & Recommendations

 Brief summary of key points.
 Practical or research recommendations.

💡 Mnemonic: I Make Reports and Discussions → Introduction, Methods, Results,

Discussion.

5️⃣Presentation Tips
 Use clear, concise language — avoid jargon unless explained.
 Follow the target journal or institution’s format guidelines.
 Number tables and figures, and give descriptive titles.
 Use past tense for methods and results.
 Proofread for grammar, spelling, and accuracy.

6️⃣Common Mistakes to Avoid

 Mixing results with discussion.
 Overloading with raw data.
 Ignoring negative results.
 Not citing sources correctly.
  Using vague conclusions.

7️⃣Writing Style Essentials

 Clarity: Be direct, avoid unnecessary complexity.
 Brevity: Keep it as short as possible, but no shorter.
 Objectivity: Focus on evidence, not opinion.
 Logical flow: One section should lead naturally to the next.

8️⃣Abstract & Title

 Title: Concise, informative, specific.
 Abstract: 150–300 words summarising background, objectives, methods, key results,
and conclusions. Written last but placed first.

Active Recall Questions

1. What does IMRaD stand for?
2. Where should limitations of the study be discussed?
3. Name two common mistakes to avoid in a research report.
4. Why is the abstract often written last?
5. What’s the difference between results and discussion?

Study Activities
 Activity 1: Take a published research article and label the IMRaD sections.
 Activity 2: Draft a 200-word abstract for a study you’ve done or read.
 Activity 3: Rewrite a vague conclusion into a specific, evidence-based one.

Mind Map – Chapter 10

 Definition & purpose → communicate findings
 Types → thesis, article, technical report, conference paper
 IMRaD → Introduction, Methods, Results, Discussion
 Other parts → title, abstract, conclusions, references
 Tips → clarity, brevity, objectivity, format adherence
 Mistakes → mixing sections, overloading, ignoring negative results