Chapter 01
Choose a topic you can actually finish
The Real-World Problem: Most delayed theses stall at month three because the student chose a grand, philosophical topic without checking whether participants, medical records, or data files actually exist.
Most thesis problems begin with a topic that is too big. Test every candidate topic against three ruthless criteria before committing:
- Answerable: Can data or empirical evidence settle it, rather than just opinion?
- Measurable: Can you explicitly name the variables, scales, or qualitative themes you will collect?
- Doable: Can you finish it with your time, physical access, and budget?
The Narrowing Funnel (Step by Step)
Example narrowing walkthrough:
× Too broad: Hospital waste management and environmental health.
∼ Narrower: Clinical waste practices in operating theatres of private hospitals.
✓ Actionable: Waste segregation protocol compliance among scrub nurses in tertiary private hospitals of Lahore.
Topic scoring worksheet
Score each candidate topic from 1 (weak) to 5 (strong). Choose the topic with the highest total, then check the lowest score for any deal-breaker (such as lack of participant access):
| Evaluation Criterion | Topic A | Topic B | Topic C |
|---|
| Personal interest: will I still care in six months? | / 5 | / 5 | / 5 |
| Answerable with data or documents | / 5 | / 5 | / 5 |
| Variables or themes are clearly defined | / 5 | / 5 | / 5 |
| Guaranteed access to participants or hospital data | / 5 | / 5 | / 5 |
| Fits current budget and semester timeline | / 5 | / 5 | / 5 |
| Supervisor has proven expertise in this sub-field | / 5 | / 5 | / 5 |
| Total Feasibility Score (max 30) | / 30 | / 30 | / 30 |
One-page pitch for your supervisor
- Working title: Single concise line stating variables and target population.
- Problem: What is unknown or unsolved, and for whom, in two sentences.
- Question: Your draft research question.
- Method: Design, population and how you will collect data.
- Feasibility: Timeline, access, and ethics approval needed.
- The Ask: The one decision you need from your supervisor.
Common Mistake: Walking into your supervisor's office asking "What topic should I do?" Supervisors respect candidates who present two evaluated options with feasibility metrics.
Where M. Yousaf Helps
Topic Evaluation & Supervisor Pitch Review
Get a 1-on-1 review of your proposed topic, feasibility evaluation, and draft supervisor pitch before submission.
Message on WhatsApp →
Chapter 02
Turn it into a clear research question
The Real-World Problem: Students often write poetic or philosophical questions ("How can hospital culture be improved?") that cannot be answered with statistical tests or qualitative coding frames.
A rigorous empirical question explicitly identifies the independent variable, dependent variable, and target population. Use these frames and fill in the blanks:
Frame 1 • Comparison: Does [Dependent Variable] differ significantly between [Group A] and [Group B] among [Population]?
Frame 2 • Relationship: Is there a significant relationship between [Variable 1] and [Variable 2] among [Population]?
Frame 3 • Prevalence / Description: What is the prevalence of [Condition/Behavior] among [Population] across [Setting]?
Keep objectives matched: The Alignment Matrix
The single most common reason examiners fail proposals is methodological drift — when Chapter 3 uses tests that do not answer the questions in Chapter 1. Fill one row per objective:
| Objective | Hypothesis | Variable(s) & scale | Test or method | Target Table |
|---|
| 1. Compare test scores between groups | H1: Group A will score higher than B | Score (Continuous ratio scale) | Independent samples t-test | Table 4.1 |
| 2. Determine link between stress & sleep | H2: Significant negative correlation | Stress (Scale), Sleep (Hours) | Pearson r correlation | Table 4.2 |
| 3. Assess compliance association by shift | H3: Association between shift & compliance | Shift (Nominal), Compliance (Binary) | Chi-square test of independence | Table 4.3 |
Checks before you finalise
- Each objective starts with one measurable verb (compare, measure, identify, assess; avoid "understand" or "explore").
- The hypothesis can be proved wrong by your data.
- The population and setting are stated in the question.
- Every objective generates its own dedicated table or chart in your Results chapter.
Common Mistake: Stating five objectives in Chapter 1, but only presenting three tables in Chapter 4. Examiners cross-reference these page numbers immediately.
Where M. Yousaf Helps
Question Formulation & Alignment Matrix Audit
Review your research questions, operational objectives, and hypothesis-test alignment matrix to prevent methodological drift.
Message on WhatsApp →
Chapter 03
Search the literature properly
The Real-World Problem: Relying solely on Google Scholar page one, citing predatory journals, or copying citations from third-party bibliographies without reading the primary source.
Search academic databases with a reproducible plan and keep a log:
| Repository | Discipline Focus | Best For |
|---|
| Crossref | All disciplines (140M+ DOI records) | Verifying exact digital object identifiers & publication metadata. |
| OpenAlex | Multidisciplinary global database | Citation networks, institutional tracing, modern open science. |
| PubMed | Medicine, nursing, health sciences | Peer-reviewed clinical trials, MeSH terminology indexing. |
| arXiv | Physics, mathematics, computer science | Preprints and computational algorithms. |
Constructing Boolean Search Strings
Combine ideas with AND, synonyms with OR, and exact phrases in quotes:
("biomedical waste" OR "healthcare waste" OR "clinical waste") AND (segregation OR compliance OR practices) AND ("operating theatre" OR "operating room")
Systematic Search Log Template
| Date | Database | Exact Query String | Filters Applied | Hits | Retained |
|---|
| 12/10/26 | PubMed | ("clinical waste") AND (segregat*) | 2020-2026, English, Free Full Text | 142 | 18 |
| 14/10/26 | OpenAlex | ("hospital waste") AND ("operating theatre") | 2021-2026, Peer-reviewed | 89 | 12 |
| 16/10/26 | Crossref | "surgical waste management" | 2019-2026 | 45 | 7 |
Reference-manager setup (Zotero / Mendeley / EndNote)
- Pick one tool and stay with it throughout the thesis.
- Import via direct DOI / PMID lookup to avoid manual typing errors.
- Add a tag for each chapter that uses the source (e.g.
Ch2_Prevalence).
- Attach the real PDF to every entry. Never cite an abstract alone.
- Fix author names and years in the manager, never by hand in the Word draft.
Common Mistake: Copy-pasting citations generated by AI chatbots. These frequently invent fictitious authors and non-existent DOIs. Always verify the DOI on Crossref.
Where M. Yousaf Helps
Literature Search & Reference Manager Training
Master database queries, Boolean string syntax, Zotero reference library setup, and bibliographic verification.
Message on WhatsApp →
Chapter 04
Pick a design and a sample size
The Real-World Problem: Students pulling arbitrary numbers like "100 participants" out of thin air without mathematical justification or statistical power analysis.
Your question decides your method, not personal preference:
| Design Type | When to Use | Data Collected | Typical Tools |
|---|
| Quantitative Descriptive | Measuring prevalence, baseline characteristics, attitudes. | Numerical survey scores, demographic data. | Validated Likert scales, questionnaires. |
| Quantitative Quasi-Experimental | Testing pre-post interventions without random allocation. | Pre-test and post-test scores across cohorts. | Paired measurements, standardized tests. |
| Qualitative Phenomenological | Understanding lived experiences and perspectives. | Audio transcripts, reflective field notes. | Semi-structured interview guides. |
| Mixed Methods Explanatory | Using qualitative interviews to explain surprising statistical results. | Initial quantitative survey followed by focused interviews. | Questionnaire + purposive sub-sample. |
1. Survey Proportion Sample Size (Cochran's Formula)
For estimating prevalence or proportions in large or unknown populations:
n = (Z² * p * (1 - p)) / d²
• Z = 1.96 (for a standard 95% Confidence Interval)
• p = 0.50 (assumed proportion for maximum variability)
• d = 0.05 (margin of error at 5%)
Calculation: (1.96² * 0.5 * 0.5) / 0.05² = 384.16 → Minimum 385 participants.
2. Statistical Power Analysis via G*Power
For hypothesis testing (e.g. comparing two independent means with a t-test):
- Test family: t tests → Means: Difference between two independent means.
- Type of power analysis: A priori: Compute required sample size.
- Input Parameters: Effect size d = 0.50 (medium effect), α err prob = 0.05, Power (1-β) = 0.80, Allocation ratio N2/N1 = 1.
- Output Result: Sample size group 1 = 64, Sample size group 2 = 64 → Total sample size = 128.
Institutional Ethics Clearance Checklist
- Participant Information Sheet (PIS) written in plain, accessible language.
- Written Informed Consent Form with voluntary withdrawal clause.
- Data de-identification plan (storing survey keys separately from responses).
- Secure encrypted storage declaration and institutional approval letter.
Where M. Yousaf Helps
Methodology Design & G*Power Sample Sizing
Get exact sample size calculations, justification write-ups, and ethical protocol guidance tailored to your committee requirements.
Message on WhatsApp →
Chapter 05
Analyse your data without panic
The Real-World Problem: Jumping directly to running t-tests and regressions in SPSS on raw survey imports without checking missing values, reverse coding, or normality assumptions.
Before any test, clean the data. Most wrong results come from messy data, not hard statistics:
- Variable View Definition: Verify every item has explicit Values defined (1 = Strongly Disagree to 5 = Strongly Agree), appropriate Measure setting (Nominal, Ordinal, Scale), and clean Variable Names.
- Missing Values: Run
Analyze > Descriptive Statistics > Frequencies. Screen missing entries; exclude pairwise or impute mean only if missing rate is below 5%. Never hide missing data.
- Outliers: Generate boxplots:
Analyze > Descriptive Statistics > Explore. Inspect extreme values. Screen for data-entry typos (e.g. typing "55" instead of "5").
- Assumptions: Check normality (Shapiro-Wilk p > .05, skewness ±1.5) and spread (Levene's test p > .05) before interpreting parametric tests.
Parametric vs. Non-Parametric Match Guide
| Research Objective | Normal / Scale (Parametric) | Skewed / Ordinal (Non-Parametric) |
|---|
| Compare 2 independent groups | Independent samples t-test | Mann-Whitney U test |
| Compare same group twice (pre/post) | Paired samples t-test | Wilcoxon signed-rank test |
| Compare 3 or more groups | One-way ANOVA | Kruskal-Wallis H test |
| Relationship between 2 continuous items | Pearson correlation (r) | Spearman rank correlation (rho) |
| Association between 2 categorical items | Chi-square test of independence | Fisher's exact test (if cell < 5) |
Menu-by-Menu SPSS Execution Paths
- Independent t-test:
Analyze > Compare Means > Independent-Samples T Test. Outcome to Test Variable(s), grouping factor to Grouping Variable → Define Groups (1, 2).
- One-Way ANOVA:
Analyze > Compare Means > One-Way ANOVA. Select Post Hoc > Tukey, and Options > Homogeneity of variance test.
- Pearson Correlation:
Analyze > Correlate > Bivariate. Select Pearson, Two-tailed, Flag significant correlations.
- Chi-Square:
Analyze > Descriptive Statistics > Crosstabs. Under Statistics check Chi-square and Phi and Cramer's V; under Cells check Observed and Expected.
Ready-to-Adapt Result Reporting Templates
Independent t-test: An independent-samples t-test was conducted to compare scores between Group A (M = [M1], SD = [SD1]) and Group B (M = [M2], SD = [SD2]). There was a significant difference in scores; t([df]) = [t-val], p = .024, d = 0.52.
ANOVA: A one-way between-subjects ANOVA was conducted to compare the effect of [Factor] on [Outcome]. There was a significant effect; F([df1], [df2]) = [val], p = .008, η² = .14. Post-hoc comparisons using the Tukey HSD test indicated significant group differences.
Correlation: [Variable 1] and [variable 2] were positively correlated, r([df]) = [val], p = .014.
Chi-square: A chi-square test of independence showed a significant association between shift timing and protocol compliance, χ²([df], N = [N]) = [val], p = .031, Cramer's V = .22.
Reporting Rule: Report exact p-values (e.g. p = .032), unless SPSS shows .000, which must be reported as p < .001. Always report effect sizes alongside significance.
Where M. Yousaf Helps
SPSS & SmartPLS Analysis Support
Full data screening, statistical testing, APA table generation, and 1-on-1 explanation of your output (from PKR 8,000 / ~$45 USD).
Message on WhatsApp →
Chapter 06
Write, cite and format cleanly
The Real-World Problem: Treating the literature review like an annotated shopping list ("Author A said X. Author B said Y."), or mixing personal interpretations into the Results chapter before reaching the Discussion.
Structure carries the reader. For empirical theses, follow the IMRaD framework:
| Section | What You Must Deliver | Common Student Mistake |
|---|
| Introduction | Problem statement, identified research gap, objectives, significance. | Starting with broad platitudes; failing to name the exact gap. |
| Literature Review | Thematic synthesis across themes; conceptual framework. | Listing one study per paragraph with zero cross-synthesis. |
| Methodology | Design, sample justification, instruments, validity/reliability, analysis plan. | Too brief for an independent researcher to replicate. |
| Results | Objective-by-objective presentation of tables and findings. | Introducing opinions or discussing broader literature findings here. |
| Discussion | Comparing findings with prior studies; implications and limits. | Merely re-listing percentages without explaining why. |
| Conclusion | Direct answers to research questions; recommendations. | Introducing brand-new data or citations not previously presented. |
Citation Quick Reference (Journal Article Pattern)
| Style | Reference List Pattern |
|---|
| APA 7th | Author, A. A., & Author, B. B. (2024). Title of journal article. Journal of Healthcare Management, 18(2), 112–124. https://doi.org/10.1016/j.jhm.2024.01.005 |
| Vancouver | Author AA, Author BB. Title of journal article. J Healthc Manag. 2024;18(2):112-24. |
| IEEE | [1] A. A. Author and B. B. Author, "Title of journal article," J. Healthc. Manag., vol. 18, no. 2, pp. 112–124, 2024. |
| Harvard | Author, A.A. and Author, B.B. (2024) 'Title of journal article', Journal of Healthcare Management, 18(2), pp. 112–124. |
The Honest Paraphrasing Test
How to Paraphrase Legally:
1. Read the source paragraph twice until you understand its underlying argument.
2. Close the browser tab or paper completely.
3. Explain the finding in your own words onto a blank screen as if teaching a peer.
4. Open the source and compare sentence structures. If your draft mimics the original word order, rewrite it from scratch.
5. Add the parenthetical in-text citation immediately. Paraphrased ideas still belong to the original author.
Pre-Submission Formatting Checklist
- Consistent heading hierarchy (H1, H2, H3) throughout using Word Styles.
- Every table and figure is numbered, captioned, and explicitly referenced in paragraph text ("As shown in Table 4.2...").
- 100% two-way match: every citation in text appears in the reference list; every list item is cited.
- Write the abstract last: 200–250 words covering Background, Methods, Core Findings (with numbers), and Conclusion.
Where M. Yousaf Helps
Proofreading, Academic Editing & Formatting
Thorough line editing, tracked changes, marginal explanations, and university template compliance on your own draft.
Message on WhatsApp →
Chapter 07
Prepare for your viva defense
The Real-World Problem: Students viewing examiners as hostile adversaries trying to "catch them out", leading to defensive arguments, ungrounded justifications, and verbal rambling.
Examiners evaluate whether you comprehend the boundaries of what you did. They test three core fundamentals:
- Methodological Justification: Why did you select this specific approach over available alternatives?
- Intellectual Honesty: Can you candidly identify your study's limitations before the committee points them out?
- Core Contribution: Can you summarize what your data adds to the literature in two precise sentences?
The A-R-E Answering Framework
Answer (A) • One direct sentence: Answer directly without hedging or apologizing. ("I chose an independent samples t-test because the dependent variable was measured on a continuous scale and satisfied normality assumptions.")
Reason (R) • Academic logic: Give the cited rationale. ("According to Field (2018), when skewness and kurtosis fall between ±1.0, parametric tests provide superior statistical power.")
Example (E) • Dataset evidence: Anchor the answer in your actual thesis findings. ("In my dataset, the Shapiro-Wilk test yielded p = .18 for Group A and p = .24 for Group B, confirming acceptable normality.")
Stop talking, and let the examiner ask their next question.
Committee Question Bank
| Domain | Questions You Must Prepare and Rehearse Out Loud |
|---|
| Purpose & Gap | Why did you choose this topic? What exact gap does it fill? What is your primary contribution? |
| Methodology | Why this design? Why this specific sampling frame? How did you control for confounding variables? |
| Instruments | How was validity and reliability established? Did you pilot test the tool? How was consent recorded? |
| Analysis | Why did you choose this test over non-parametric alternatives? How did you treat missing data? |
| Results & Limits | Which finding surprised you? Which result do you trust least, and why? What are your limitations? |
| Future Impact | If you had six more months, what would you do differently? Who should act on these findings? |
The 7-Day Pre-Viva Countdown
- Day 7 to 5: Re-read your Abstract, Objectives, and Conclusions until you can recite them without slides.
- Day 4 to 3: Review every single table: know sample size, test statistic, p-value, and effect size by heart.
- Day 2: Conduct a full timed slide rehearsal with an academic colleague or coach who interrupts with questions.
- Day 1: Prepare your bound copy with colored adhesive tabs for instant chapter and table navigation. Rest.
Where M. Yousaf Helps
Mock Viva Defense Coaching
Experience an intensive 1-on-1 mock defense with direct, constructive feedback on your slide deck, verbal delivery, and answer structures.
Message on WhatsApp →