BSMA1002 · 4 credits · Foundation
Statistics for
Data Science 1
Six weeks of describing and counting, which feel easy — and then, from week 7, probability, which is the first genuinely abstract thing the Foundation level asks of you.
- Lectures
- 149 · 51 h
- Prerequisite
- None
- Bonus marks
- up to 5
- Unlocks
- Statistics 2
What this course actually asks
Two courses wearing one nameStatistics 1 is taught by Usha Mohan of the Department of Management Studies at IIT Madras, and it is really two courses joined in the middle. Weeks 1 to 4 are descriptive statistics — summarise a dataset, draw it honestly, measure how two variables move together. Weeks 5 to 12 are counting and probability, ending in named distributions.
The first half rewards care. The second half rewards a different skill entirely: being able to set up a problem before solving it. "Three cards are drawn without replacement" is a modelling question dressed as arithmetic, and the arithmetic is never the hard part.
The week-7 cliff is real. People coast through six comfortable weeks, then hit probability and discover they have not built a study habit yet. If you are taking four courses, this is the one to stay ahead in.
It feeds Statistics 2 directly, and nearly everything in the Diploma's machine learning courses assumes it. Expectation and variance, in weeks 10 and 11, are used constantly later.
The twelve weeks
Topics quoted from the official course pageIntroduction and type of data. Types of data, descriptive and inferential statistics, scales of measurement.
Describing categorical data. Frequency distributions, best practices for graphing categorical data, mode and median for a categorical variable.
Describing numerical data. Frequency tables, mean, median and mode, quartiles and percentiles, range, variance, standard deviation, IQR, five-number summary.
Association between two variables. Contingency tables and relative frequencies; scatterplot, covariance, Pearson correlation, point bi-serial correlation.
Counting. Addition rule, multiplication rule, factorials.
Permutations and combinations.
Probability. Basic definitions, events, properties of probability.
Conditional probability. Multiplication rule, independence, law of total probability, Bayes' theorem.
Random variables. Random experiment, sample space, discrete and continuous random variables, probability mass function, cumulative distribution function.
Expectation and variance of a discrete random variable.
Binomial and Poisson. Bernoulli trials, i.i.d. random variables, binomial random variable and its expectation and variance, Poisson distribution.
Continuous random variables. Area under the curve, properties of a pdf, uniform distribution, exponential distribution.
Week 3, on your own numbers
Centre, spread, and the five-number summaryPaste any list of numbers. Everything week 3 asks for appears at once, with the histogram and the boxplot that go with it. Change one value at the top end and watch how much the mean moves and how little the median does — that is the entire argument for reporting both.
Two conventions that cost marks
Which standard deviation? Dividing by n gives the population value; dividing by n − 1 gives the sample value. Both are shown above. A question that says "these are all the observations" wants n; one that says "a sample of" wants n − 1.
Which quartiles? This tool uses the median-of-halves rule, excluding the overall median when the count is odd — the convention in the prescribed Weiss text. Other books split differently and get different numbers. Check which one your course uses before an exam.
How it is graded
May 2026 grading document · confirm against your course pageThe standard formula, plus something the other courses do not have:
+ bonus for extra activity, capped at 5
The five bonus marks
Statistics 1 runs four optional "extra activities" across the term. They are worth 5 marks in total — 3.75 for the weekly activity itself and 1.25 at the instructors' discretion for quality.
- They are added only if you pass the course.
- Each activity is peer reviewed. If you do not complete at least five peer reviews, you get zero for that activity — even if you submitted it.
Five marks is the difference between a B and an A at the boundary. Doing the activity and then skipping the reviews is the most avoidable loss in the whole Foundation level.
The eligibility gate
Average of the best 5 of 8 — the first seven weekly assignments plus Mock Quiz 1 — must be 40/100 or more, and you must attend at least one quiz.
Where the marks actually go
My read, from this course and from what people ask aboutSubmitting an activity and skipping the reviews
Fewer than five peer reviews means zero for that activity. It is written in the grading document and it still happens every term.
Permutation where a combination was wanted
Week 6 lives or dies on one question: does the order matter? Decide that out loud before you write a single factorial, every time.
Conditioning the wrong way round
P(A|B) and P(B|A) are different numbers, and week 8's word problems are written so that the natural reading gives you the wrong one. Write the events down as symbols before computing.
Confusing correlation's scale
Covariance has units and no bound; Pearson correlation is unitless and lives in [−1, 1]. Week 4 questions check that you know which one changes when you change the units of measurement.
Everything worth opening
Official first, then the free things that genuinely helpEvery link here was opened and checked on 23 September 2026.
Official
- Course page BSMA1002 on study.iitm.ac.in The week list, the instructor, and the two downloadable volumes.
- Lectures Statistics for Data Science 1 — full playlist All 149 official lectures on the programme's own channel.
For the probability half (weeks 5–12)
- Tool Seeing Theory · Brown University Probability and distributions as things you can drag. The Bayes chapter alone makes week 8 click. Start here before re-watching a lecture.
- Video StatQuest · Statistics Fundamentals Short, plain explanations of exactly this syllabus — histograms, distributions, expectation, variance. Good when a lecture has not landed.
- Free book OpenIntro Statistics A complete free textbook covering weeks 1 to 12, with exercise solutions. The closest free substitute for the prescribed books.
For practice
- Practice Khan Academy · Statistics and probability Drill sets for counting, conditional probability and random variables. Use it to find your gap, not to learn the theory.
- Tool Explained Visually Short interactive explainers; the conditional probability and correlation pages match weeks 4 and 8 directly.
What to do in week one
Term starts 2 October 2026- Find the extra activity scheduleFive marks, gated behind peer reviews you have to remember to do. Put both dates — submission and review — in your calendar the moment they are announced.
- Download both volumesDescriptive Statistics and Probability and Probability Distributions, straight from the course page.
- Do not coast through weeks 1–4They are easy, which is exactly why people arrive at week 7 without a study routine.
- Open Seeing Theory in week 5Two weeks before you need it. Counting and probability make far more sense after you have played with them.
- Write down "does order matter?"Stick it above your desk. It is most of weeks 5 and 6.
- Check your course page's grading sectionEverything here is from the May 2026 document. The September 2026 one was not published when this was written.