MH3511: pass the exams, not just read the notes
Your complete guide to Nanyang Technological University's data analysis with computer course. See where the marks are, work real practice questions, and study with an AI tutor that knows MH3511.
Sia generates MH3511 practice questions, walks through r basics and data in r step by step, and quizzes you on the material the exam weights most heavily.
Worked example
Two groups of 12 customers each rate a product on a 1-5 ordinal scale. You want to test whether the groups differ. Which test is appropriate, and why?
Identify the measurement scale: a 1-5 rating is ordinal, so means and t tests are not appropriate.
Choose the nonparametric two-independent-sample test: the Wilcoxon rank-sum test.
ANOVA would apply to three or more groups of interval data; here there are two groups of ordinal data.
The trap: Letting equal group sizes suggest a paired design, or letting 'two groups' suggest a t test regardless of scale. Design and measurement scale together choose the test. classic slip!
One exam decides 50% of your grade. Summative assessment. This whole page is built around that.
Overview
What MH3511 is, and where it sits
MH3511 Data Analysis with Computer is NTU's applied statistics course in R, taught by the Division of Mathematical Sciences. The OBTL+ document, implemented from AY2025-2026, motivates it through business decision-making: examining data to find patterns, and learning to plot, summarise, make inferences from and present various types of data.
Nine intended learning outcomes run from writing R programs and distinguishing measurement scales, through descriptive statistics and standard plots, to confidence intervals, hypothesis tests and Type I and II errors, inference on categorical data, nonparametric alternatives, and linear regression with assumption checking.
Assessment has five parts: a group written report 15%, two one-hour short-answer quizzes at 15% each, class participation tracked via Wooclap in lab sessions 5%, and a two-hour final examination of multiple-choice and short-answer questions 50%. The course is 3 AU with 50 contact hours of lectures and laboratories, offered in Semester 2, and requires MH2500 or BS1008. The course author is Yue Mu.
Always treat your own course outline and the exam timetable as authoritative.
Difficulty & time commitment
Is MH3511 hard, and how much time does it take?
MH3511 is manageable if you keep a weekly rhythm and treat the back half as the main event. The pattern is consistent: it starts gently and steepens, and the heaviest assessment is the part that separates grades.
The difficulty curve and the assessment weighting point the same way: the back half is harder and worth more. Front-loading effort there is the highest-return decision in the course.
Is this course for you
Who tends to do well, and who tends to struggle
You will likely do well if
- You attend the labs — eight Wooclap sessions is the whole participation mark.
- You want to be able to run and read a real analysis, not derive it.
- You can choose a test by asking what kind of data you have.
- You pull your weight in a group of five on the report.
You may struggle if
- You skip labs; the weekly assignment and the participation marks both live there.
- You run tests without checking assumptions; model checking is an ILO.
- You leave the group report late — it is 15% and rubric-marked.
- You treat the quizzes as low-stakes; together they are 30%.
- Keep an R script per week that reproduces the lecture examples from scratch.
- Build a decision table: data type and design on one axis, the right test on the other.
- Practise interpreting p-values and intervals in one plain sentence each.
- Check normality and equal-variance assumptions before every parametric test in your report.
Syllabus
The 13 topics, week by week
The exam-weight marker on each topic shows where the marks concentrate. The amber topics carry the highest exam weight.
T1 · R basics
Week 1; ILO 1Syntax, expressions, variables, vectors, matrices and data frames.
T2 · Data in R
Week 2; ILO 1Importing and subsetting data sets; loops.
T3 · Describing data
Week 3; ILOs 2-4Summary statistics, boxplots, histograms, normality checks, qq-plots, outliers, transformations.
T4 · Sampling and confidence intervals
Week 4; ILOs 5-7Sampling distributions, the central limit theorem, interval estimates.
T5 · Hypothesis testing
Week 5; ILOs 5-7Type I and Type II errors, p-values.
T6 · Categorical data: proportions
Week 6; ILOs 5-7Proportion estimates and tests; goodness-of-fit.
T7 · Contingency tables
Week 7; ILOs 5-7Two-way and paired two-way tables; quiz 1.
T8 · Two samples
Week 8; ILOs 5-7Independent and dependent samples; inference on mean differences.
T9 · Multiple samples
Week 9; ILOs 5-7ANOVA for more than two independent samples; dependent designs.
T10 · Nonparametric tests I
Week 10; ILOs 5-6, 8Quantile test, Wilcoxon rank-sum, Kruskal-Wallis; quiz 2.
T11 · Nonparametric tests II
Week 11; ILOs 5-6, 8Sign test, Wilcoxon signed-rank, Friedman test.
T12 · Correlation and regression
Week 12; ILOs 5-6, 9Correlation coefficient and its test; simple linear regression.
T13 · Regression inference
Week 13; ILOs 5-6, 9Inference on parameters, prediction, model checking.
How it's assessed
Assessment structure
| Component | Weight | Format & timing |
|---|---|---|
| Continuous assessment: written project report (groups of five) | 15% | Written project report prepared in groups of five. During the semester. Continuous assessment. |
| Continuous assessment: quiz 1, one-hour short-answer (ILOs 1-4) | 15% | One-hour short-answer quiz on ILOs 1-4 (R basics, measurement scales, summary statistics, plots). Week 7. Continuous assessment; no make-up. |
| Continuous assessment: quiz 2, one-hour short-answer (ILOs 5-7) | 15% | One-hour short-answer quiz on ILOs 5-7 (estimation, hypothesis testing, categorical data). Week 10. Continuous assessment; no make-up. |
| Final examination: two hours, MCQs and short-answer questions | 50% | Two-hour final examination: multiple-choice and short-answer questions on all nine ILOs. Examination period. Summative assessment. |
| Class participation tracked via Wooclap in lab sessions (8+ sessions = 5%, 6-7 sessions = 3%) | 5% | Class participation tracked via Wooclap during lab sessions: 8 or more sessions earns 5%, 6-7 sessions 3%. Weekly. Continuous assessment. |
- The five components sum to 100 and no examination hurdle is published. There is no make-up quiz; with a medical certificate a missed component is not counted. Participation marks depend on Wooclap-tracked lab sessions: eight or more for 5%, six or seven for 3%.
- Half the grade is a two-hour examination mixing multiple-choice and short-answer questions across all nine ILOs. The two quizzes split the first seven ILOs between them in weeks 7 and 10, so by the exam every topic except nonparametrics and regression has already been tested once.
This is an exam-cram course. With the exams at 50% of the grade and the final examination: two hours, mcqs and short-answer questions alone at 50%, your result is overwhelmingly decided by how well you perform under time pressure. Summative assessment.
Final exam timing: During the examination period. Confirm the exact date and venue on your exam timetable.
How to actually pass it
A weekly rhythm, two checklists, and the traps to avoid
The course rewards consistency over cramming, and practice over re-reading. Here is the loop that works, then what to have nailed before each exam.
The weekly loop
Before the mid-semester checklist
- Write an R script that imports, subsets and summarises data.
- Produce and interpret a boxplot, histogram and qq-plot.
- Build a confidence interval for a mean and a proportion.
- Run a hypothesis test and explain Type I and Type II errors.
Before the final heaviest topics
- Analyse a contingency table.
- Compare two or more samples with the right parametric test.
- Choose and run a nonparametric alternative.
- Fit a simple linear regression and check its assumptions.
The mistakes that cost marks
Parametric test on ordinal data. Measurement scale decides the test; the course's second ILO is exactly this.
Paired data treated as independent. Dependent samples need paired methods; the design, not the data size, decides.
Reporting without a plot. Every inference in the report should be accompanied by the plot that motivates it.
Teaching team
Who teaches MH3511
The bios below are factual. We do not rate lecturers; any star ratings are submitted by students who have taken MH3511.
Teaching team as listed in the course materials reviewed. AskSia does not rate lecturers; star ratings are submitted by students who have taken MH3511.
Formula & concept sheet
The vocabulary and formulas you must own
- Data frame
- R's table structure with named columns of possibly different types.
- Measurement scale
- Nominal, ordinal, interval or ratio: the level of a variable that decides valid analyses.
- qq-plot
- A plot comparing sample quantiles with a theoretical distribution to check normality.
- Central limit theorem
- The sample mean is approximately normal for large samples.
- Type I and Type II errors
- Rejecting a true null hypothesis, and failing to reject a false one.
- Goodness-of-fit test
- A chi-square test of whether observed counts match expected proportions.
- Contingency table
- A cross-tabulation of two categorical variables.
- ANOVA
- A test comparing means across more than two groups.
- Wilcoxon rank-sum test
- A nonparametric alternative to the two-sample t test.
- Model checking
- Verifying regression assumptions through residual analysis.
Set texts
The prescribed reading
The syllabus references map straight onto these.
A Handbook of Statistical Analyses Using R
.
Where it fits
Prerequisites, related courses & why it matters
Prerequisite: MH2500 or BS1008. 3 AU; 50 contact hours; offered in Semester 2.
Your MH3511 study toolkit
Study the course with Sia, not just read about it
Each tool already knows MH3511: your syllabus, your texts, and where the marks are. Grouped by how you study, from first contact to exam week.
FAQ
Frequently asked questions
Is MH3511 hard?
It rates moderate. It is applied and computational, with marks spread across quizzes, a group report, participation and a 50% exam with no hurdle.
What is the assessment breakdown?
Group written report 15%, two one-hour quizzes at 15% each, class participation 5% and a two-hour final examination 50%, per the OBTL+ document.
How is participation marked?
Via Wooclap during lab sessions: completing at least 8 sessions earns the full 5%; 6 or 7 sessions earns 3%.
What are the prerequisites?
MH2500 or BS1008.
Which software?
R throughout, from basic syntax in week 1 to regression model checking in week 13.
Who wrote the course?
The OBTL+ document names Yue Mu as course author.
Study MH3511 with Sia
Work through r basics, data in r, describing data and the rest of the course with a tutor that knows it and quizzes you on the topics the assessments weight most heavily.
Start studying with Sia
