SO5005 Quantitative and Qualitative Research Methods
The Complete Study & Assessment Guide · Semester 1, 2026/27SO5005 Overview
- The Hong Kong Polytechnic University
- Semester 1, 2026/27
- Postgraduate subject
- Statistics and research methods
- Coursework assessment
What this subject asks you to do
SO5005 treats research as a connected sequence of decisions. You begin with a problem, establish what is already known, isolate an uncertainty and turn it into a question.
- Assessment pattern Two presentations, two proposals and one statistical report each carry an equal share of the final mark.
- Quantitative habit Identify scale and dependence before selecting a statistical procedure.
- Qualitative habit Make sampling, interviewing and the path from excerpts to themes transparent.
- Before submission Check that the literature gap, research question, design, analysis and conclusion describe one coherent study.
How SO5005 is assessed
| Component | Weight | Format |
|---|---|---|
| Quantitative group presentation | 20% | Health-related quantitative research proposal presented in class |
| Qualitative group presentation | 20% | Health-related qualitative research proposal presented in class |
| Quantitative research proposal | 20% | Individual written research proposal |
| Qualitative research proposal | 20% | Individual written research proposal |
| Statistical report | 20% | Individual report with research question, justified test and results |
Current SO5005 dates
| Date | Item | Control |
|---|---|---|
| Quantitative group presentation | 2026-10-16 | |
| Statistical report | 2026-10-19 | |
| Quantitative research proposal | 2026-10-26 | |
| Qualitative group presentation | 2026-11-27 | |
| Qualitative research proposal | 2026-12-07 |
Dates are as published in the current course materials. Confirm exact deadlines and submission settings in the live LMS.
What SO5005 covers
The guide follows the research process from design and appraisal through quantitative analysis, qualitative inquiry and mixed-methods integration.
Research Process and Study Designs
questions, claims, observational and experimental designs, protocols and ethical alignmentStart with the claim
Research design links an answerable question to sampling, measurement, comparison and inference.
- Questions and claims
- Descriptive and analytical designs
- Cross-sectional and longitudinal time
- Experimental allocation
- Protocols and ethics
- Limits on causal inference
Literature Gaps, Questions and Proposals
evidence synthesis, literature gaps, answerable questions, significance and proposal structureSynthesise rather than list
A literature review groups studies around claims, compares populations, measurements and designs, and states what the evidence collectively establishes.
- Claim-led review paragraphs
- Comparing definitions
- Kinds of literature gap
- Answerable research questions
- Significance and originality
- Proposal alignment
Critical Appraisal and Research Ethics
validity, bias, confounding, applicability, consent, privacy and responsible researchValidity comes before precision
Critical appraisal tests whether a result deserves belief, how large and precise the effect is, and whether it transfers to a new setting.
- Internal validity
- Selection and information bias
- Confounding and random error
- Effect size and precision
- Consent and voluntariness
- Privacy and confidentiality
Variables, Descriptions and Distributions
measurement scales, counts, percentages, centre, spread, missingness and distribution shapeDescribe before testing
Variable scale determines which summaries and displays are meaningful. Counts and percentages describe categories; numerical measures require centre and spread.
- Categorical and numerical variables
- Continuous and discrete measurement
- Counts and denominators
- Mean and standard deviation
- Median and interquartile range
- Missing and extreme values
Sampling and Sample Size
target populations, frames, probability sampling, recruitment targets and attritionTrack three populations
The target population is named by the claim, the sampling frame is who can be reached, and the achieved sample is who contributes usable data. Gaps between them create coverage, non-response and attrition problems.
- Target population
- Sampling frames
- Probability sampling
- Purposeful and convenience routes
- Sample-size assumptions
- Attrition and retention
Hypotheses and Test Selection
null models, p-values, confidence intervals, errors, power and defensible test choiceChoose a test through design logic
Begin with the outcome scale, number of groups, dependence and estimand, then examine assumptions. The null hypothesis is a reference model.
- Null and alternative hypotheses
- P-value interpretation
- Confidence intervals
- Type I and Type II error
- Power
- Test-selection sequence
Paired and Independent Comparisons
paired differences, independent groups, t procedures, signed ranks and rank comparisonsDependence comes from design
Measurements from the same person, two eyes from one participant or deliberately matched units are paired. Observations from unrelated participants in separate groups may be independent.
- Sources of dependence
- Within-pair differences
- Paired t procedures
- Signed-rank procedures
- Independent group comparisons
- Direction and subtraction order
Correlation, Regression and Association
scatterplots, slopes, model fit, contingency tables, expected counts, odds and odds ratiosQuantify patterns in units
Correlation summarises direction and strength, while simple regression expresses an expected outcome through an intercept and slope. The slope is the expected outcome change for one predictor unit.
- Scatterplots
- Correlation
- Regression slopes
- R-squared
- Contingency tables
- Odds and odds ratios
Qualitative Design, Sampling and Interviews
purposeful recruitment, information-rich cases, neutral questions, probes and iterative designDesign for experience and meaning
Qualitative studies investigate how people understand, experience or produce a phenomenon in context. Purposeful recruitment seeks information-rich cases and relevant variation.
- Qualitative study purposes
- Purposeful sampling
- Variation and criterion sampling
- Interview sequencing
- Neutral questions
- Probes and adaptation
Qualitative Analysis, Rigour and Integration
coding, categories, themes, reflexivity, credibility, mixed-methods connection and inferenceBuild interpretation visibly
Coding labels relevant segments, categories organise related codes and themes make an interpretive claim about patterned meaning.
- Coding and memoing
- Categories and themes
- Contrary cases
- Reflexivity
- Qualitative rigour
- Mixed-methods integration
The design then determines who or what can supply evidence, the analysis summarises that evidence, and the final claim must remain inside the limits created by sampling, measurement and comparison.
The course uses health-related applications and expects both theoretical understanding and practical reasoning.
How the quantitative half fits together
The early weeks move through study design, literature review, critical appraisal and ethics before applied statistics. That sequence is important because a p-value cannot repair a vague question or a biased comparison.
The guide therefore makes you identify variable scales, paired or independent observations and relevant assumptions before choosing a test.
Descriptive summaries, confidence intervals and effect estimates lead the interpretation; statistical significance is only one part of the result.
How the qualitative half changes the logic
The later weeks cover qualitative design, sampling, interviews, analysis and paradigms. Qualitative sampling seeks information-rich cases rather than a miniature probability sample.
Interviews should invite experience without supplying the preferred answer. Analysis must preserve a visible route from excerpts through codes and categories to themes, with reflexivity and contrary cases used to test interpretation.
Where the two strands meet
The subject synopsis includes mixed methods. Integration is not achieved by placing a survey and interviews beside one another.
The strands must connect through sampling, explanation, comparison or a joint conclusion. Across both traditions, the same standard applies: every method choice needs a reason and every conclusion needs a traceable evidence path.
Worked example · free
Choose the unit of analysis before the test
- 2Identify the repeated measurements contributed by each participant.
- 2Define one directional within-participant difference.
- 2Explain why analysing ten independent eyes would misstate uncertainty.
Key terms
- Research Question
- A precise statement of the population, phenomenon or variables and the kind of answer a study seeks.
- Study Design
- The structure connecting a research question to sampling, measurement, comparison and inference.
- Sampling Frame
- The operational list or route through which members of a target population can be reached.
- Confounding
- A rival explanation created when a third factor relates to both an exposure and an outcome.
- Median
- The middle ordered value, used as a resistant measure of centre when a distribution is skewed.
- Confidence Interval
- A range produced by a procedure that quantifies uncertainty around an estimated quantity under its assumptions.
- P-value
- The probability, under a stated null model, of data at least as incompatible with that model as those observed.
- Paired Data
- Measurements linked because they come from the same or deliberately matched units.
- Odds Ratio
- The odds of an outcome in one group divided by the odds in a reference group.
- Purposeful Sampling
- Deliberate recruitment of cases capable of illuminating a qualitative research question.
- Reflexivity
- Systematic attention to how the researcher shapes access, interaction and interpretation.
- Theme
- An interpretive pattern of shared meaning built across coded qualitative material.
- Mixed Methods
- A design that connects quantitative and qualitative strands to produce an integrated inference.
SO5005 FAQ
How is the subject assessed?
The published structure has five equally weighted subcomponents. Quantitative and qualitative group presentations contribute twenty per cent each, quantitative and qualitative individual proposals contribute twenty per cent each, and an individual statistical report contributes the remaining twenty per cent. Together the numeric weights total one hundred per cent.
What should connect a research proposal from beginning to end?
The background should establish the problem, the literature review should synthesise what is known, and the gap should lead directly to the research question. The methodology then needs to answer that exact question, while the significance section explains why resolving the uncertainty matters. A different construct appearing halfway through is a warning that alignment has broken.
What does the statistical report need to show?
The instructions ask for the research question, the statistical test used with justification, and results beginning with descriptive findings. A strong report also makes the denominator, units, dependence structure and direction of each comparison visible. It interprets the estimate and uncertainty before limiting the conclusion to what the design can support.
How do I choose between paired and independent analysis?
Look at how observations were produced. Repeated measurements on the same person, two eyes from one participant or deliberately matched units are paired, so the analysis uses within-pair differences or another dependence-aware model. Separate groups of unrelated participants are independent when the recruitment and design support that assumption. Equal column lengths do not create pairing.
How should qualitative sampling be justified?
State who can illuminate the question, which kinds of variation matter and how recruitment will reach those cases. Purposeful sampling is defensible when its criteria follow from the study aim. The proposal should also explain how early analysis may refine later recruitment and what would show that additional interviews are contributing little new insight.
What makes a qualitative theme stronger than a topic label?
A topic label says what participants discussed, while a theme makes an interpretive claim about patterned meaning. The report should show how excerpts were coded, how codes formed categories, how candidate themes were checked against the full material and how contrary cases limited the final claim. Frequency alone does not determine conceptual importance.
When does a project genuinely use mixed methods?
A project becomes mixed methods when quantitative and qualitative strands connect in the design or inference. One strand may select cases for the other, explain an unexpected pattern, compare findings in a joint display or contribute to a combined conclusion. Two parallel result sections without a planned point of integration remain two separate analyses.
What should I check immediately before submission?
Recompute every reported number, confirm denominators, units, subtraction orders and reference groups, then read the conclusion against the design. Check that ethical safeguards and qualitative analytic decisions are concrete. Finally verify citations, tables, figure labels and any required disclosure of generative AI use, while confirming current submission settings on Canvas.
Questions students ask in each SO5005 chapter
- Can a cross-sectional study establish cause?
- A cross-sectional study can describe prevalence and association at one time, but exposure and outcome are measured together. Temporal order is therefore uncertain, and common causes may explain the pattern. A causal conclusion needs stronger design evidence than a single snapshot provides. Chapter 01 · Research Process and Study Designs
- How is synthesis different from summary?
- Summary reports what each paper says. Synthesis organises papers around a question, compares why their findings agree or differ and states what the body of evidence supports. It is the bridge from published work to the proposal's literature gap. Chapter 02 · Literature Gaps, Questions and Proposals
- Does a narrow confidence interval prove validity?
- No. A narrow interval indicates statistical precision under the analysis assumptions, but a biased sample or measurement process can produce a precise estimate of the wrong quantity. Appraise recruitment and measurement before interpreting interval width. Chapter 03 · Critical Appraisal and Research Ethics
- When should the median lead?
- Use the median when a numerical distribution is markedly skewed or contains influential extreme values that make the mean a poor description of a typical observation. Pair it with quartiles or the interquartile range and still explain the full shape. Chapter 04 · Variables, Descriptions and Distributions
- Does a larger sample remove sampling bias?
- A larger sample reduces random uncertainty when the design and model are suitable, but it does not automatically improve coverage or response. A huge convenience sample can estimate its respondents precisely while representing the target population poorly. Chapter 05 · Sampling and Sample Size
- What does a p-value actually mean?
- It is the probability, assuming the null model and analysis assumptions hold, of observing data at least as incompatible with that model as the data obtained. It does not state the probability that the null or alternative hypothesis is true. Chapter 06 · Hypotheses and Test Selection
- Are two columns automatically paired?
- No. Pairing comes from a meaningful link between observations, such as repeated measures on the same person or deliberate matching. Equal column lengths alone do not create dependence, and unrelated groups should not be analysed as pairs. Chapter 07 · Paired and Independent Comparisons
- Does correlation establish causation?
- No. Correlation records a pattern of co-variation. A causal interpretation requires temporal order, credible control of common causes, appropriate measurement and a design capable of isolating change. A strong coefficient cannot supply those design features. Chapter 08 · Correlation, Regression and Association
- How is qualitative sampling justified without probability selection?
- Justification follows information needs rather than population estimation. State which experiences can illuminate the question, what variation is analytically important and how recruitment will reach it. The final report should define the context instead of claiming statistical representativeness. Chapter 09 · Qualitative Design, Sampling and Interviews
- What separates a theme from a topic?
- A topic labels what was discussed, while a theme makes a patterned interpretive claim. Themes should connect multiple excerpts, explain variation and survive comparison with contrary material. Counting how often a word appears does not create interpretation. Chapter 10 · Qualitative Analysis, Rigour and Integration
How to prepare for the assessments
Begin each study session with one decision rather than one chapter title. For example, take a fresh scenario and decide what the question permits, whether observations are paired, which summary matches the distribution, or which qualitative sampling criterion is justified. Write the reason before checking the chapter.
For quantitative work, recompute every worked number and always print units, denominators, reference groups and subtraction order. For qualitative work, practise rewriting leading interview questions and turning topic labels into interpretive themes supported by an evidence trail. Once a week, build a one-page proposal chain from background through literature gap, question, design, analysis and bounded conclusion.
Use the published seven marking criteria as a final audit, but do not invent weights among them. Before any current assessment, compare dates and submission settings with Canvas, since operational details may be updated after the captured teaching materials. This routine prepares the actual coursework while keeping the methods connected instead of memorised as isolated definitions.
Your AI Statistics tutor for SO5005
Stuck on a hard SO5005 question? Sia is AskSia’s AI Statistics tutor — ask any SO5005 Quantitative and Qualitative Research Methods question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.
