ECMT1010: pass the exams, not just read the notes
Your complete guide to University of Sydney's introduction to economic statistics unit. See where the marks are, work real practice questions, and study with an AI tutor that knows ECMT1010.
Sia generates ECMT1010 practice questions, walks through inference for means: one- step by step, and quizzes you on the material the exam weights most heavily.
Worked example
In a random sample of 400 recent graduates, 144 say they used a paid tutor at some point. The CLT conditions hold (n times p-hat = 144 and n times (1 − p-hat) = 256, both at least 10). Using the normal approximation, which is the correct 95% confidence interval for the population proportion p who used a paid tutor? (Use z* = 1.96 and SE = root[ p-hat(1 − p-hat)/n ].)
Point estimate: p-hat = 144/400 = 0.36.
Margin of error at 95%: ME = z* times SE = 1.96 times 0.024 = 0.047.
Confidence interval: 0.36 plus or minus 0.047 = (0.313, 0.407). So we are 95% confident the interval 31.3% to 40.7% captures the true proportion (option index 0).
The trap: Option (0.312, 0.408) uses the 2 SE shortcut instead of the exact z* = 1.96, which is acceptable in the unit but is not the requested 1.96-based interval. Option (0.336, 0.384) forgets the division by n inside the square root for the margin (it standardises p-hat itself rather than the SE). Option (-0.581, 1.301) takes SE = root[ p-hat(1 − p-hat) ] with no /n at all, giving an impossible interval that runs below 0 and above 1: a probability can never exceed 1, which is the quick sanity check that flags this error. classic slip!
One exam decides 50% of your grade. FORMAT is grounded from the captured S2 2024 final paper; the WEIGHT is an estimate subject to confirmation. This whole page is built around that.
Overview
What ECMT1010 is, and where it sits
ECMT1010 is the University of Sydney's first-year economic-statistics gateway, taught out of the School of Economics. It builds the full introductory-statistics toolkit for economists in one 6-credit-point unit: describing data, estimating with confidence intervals and the bootstrap, hypothesis testing with randomization and p-values, the normal distribution and the Central Limit Theorem, formula-based inference for proportions and means, simple linear regression, then a probability-theory and random-variables block that ends on the mathematical-statistics results that bridge to ECMT1020. The set text is the free Lock5 book, Statistics: Unlocking the Power of Data, with analysis done in StatKey and Excel.
What makes the unit distinctive, and genuinely confusing for many students, is its two-pronged structure. A modern simulation track (bootstrap confidence intervals and randomization tests, all run in StatKey) runs alongside a classical formula and CLT track (z and t inference by hand from a provided formula sheet), and then Weeks 11 to 13 take a hard turn into probability axioms, Bayes' rule, expected value and variance rules, and the estimator-versus-estimate theory. The unit's own Week 2 page calls it one of the harder undergraduate units at the University and recommends 6 to 8 hours of private study a week on top of the lecture and workshop.
The exam pain is specific and known from the past papers: students rarely lose marks on arithmetic. They lose them on choosing the right test (one mean? a difference in means? paired? a single proportion? a difference in proportions? a regression slope?), on stating the hypotheses in population parameters, on reading the supplied t and N(0,1) tables for the right degrees of freedom and tail, and on writing the one-sentence conclusion in context. Because the exam hands you the formula sheet and the distribution tables, every mark sits in the setup, the decision rule, and the interpretation, not in formula recall.
Official outline: sydney.edu.au · ECMT1010 outline. Always treat the official outline and the exam timetable as authoritative.
Difficulty & time commitment
Is ECMT1010 hard, and how much time does it take?
ECMT1010 is manageable if you keep a weekly rhythm and treat the back half as the main event. Across student reviews 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 unit.
Is this unit for you
Who tends to do well, and who tends to struggle
You will likely do well if
- You are comfortable with HSC-level algebra and reading values off a table: rearranging the z and t formulas, substituting cleanly, and pulling the right critical value for the right degrees of freedom and tail.
- You drill the "name the right test" decision (one mean, a difference in means, paired, one proportion, a difference in proportions, a regression slope) until it is automatic, because that single choice drives most of the marks.
- You actually run StatKey for the bootstrap and randomization questions so the simulation track is as familiar as the formula track, rather than only watching the lectures.
- You keep up week by week, especially the probability-theory turn in Weeks 11 to 13 (axioms, Bayes' rule, expected value and variance rules, the estimator-versus-estimate theory), which feels like a different unit and still appears on the final.
You may struggle if
- You treat the unit as plug-and-chug arithmetic; the marks are in the setup, the test choice and the one-sentence conclusion in context, not in the number you compute.
- You let the two tracks blur together and cannot say when a question wants a bootstrap or randomization simulation versus a formula-based z or t calculation.
- You leave the Weeks 11 to 13 probability-theory block (Bayes' rule, variance rules, the estimator theory) to the last week, even though it is examinable and conceptually the hardest part.
- You skip writing out hypotheses in population parameters and defining notation, which is where structured short-answer marks are quietly lost.
- Build a one-page "which test?" flowchart (sample size, one or two groups, mean or proportion, paired or independent, slope) and rehearse classifying past-paper questions before computing anything.
- Practise the exact mark-earning ritual on every short-answer question: state H0 and Ha in population parameters, define notation, substitute, read the right table row, then write the one-sentence conclusion ("weak or strong evidence that...").
- Drill StatKey for the bootstrap CI and the randomization p-value until you can set one up fast, and always plot the bootstrap distribution to check it is smooth before trusting the percentile method.
- Work the past finals timed with only the provided formula sheet and a non-programmable calculator, then check method rather than just the answer key, since the real paper combines two skills per question.
Syllabus
The 11 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 · Sampling, bias and study design
Lock5 Ch 1.1 to 1.3Population versus sample and inference, sampling bias and random sampling, association versus causation, confounding variables, observational versus experimental studies, randomization, blinding and the placebo effect, and natural experiments.
T2 · Describing data: centre, spread and shape
Lock5 Ch 2.1 to 2.5Categorical versus quantitative data, histograms and skew, mean versus median and resistance, standard deviation, range, IQR and the five-number summary, boxplots and the 1.5 times IQR outlier rule, z-scores, the 95% rule, and correlation r.
T3 · Confidence intervals and the bootstrap
Lock5 Ch 3.1 to 3.4The sampling distribution and standard error, interval estimates and confidence intervals, the 95% rule (statistic plus or minus 2 SE), interpreting a CI correctly, and the bootstrap (resampling with replacement, the percentile method, when it fails).
T4 · Hypothesis testing and randomization
Lock5 Ch 4.1 to 4.5Null and alternative hypotheses in population parameters, the randomization distribution, the p-value, the significance level, one- versus two-sided tests, the CI to HT equivalence, and Type I and Type II errors with power.
T5 · The normal distribution and the CLT
Lock5 Ch 4.3 to 4.5, 5.1 to 5.2The normal density and the standard normal N(0,1), standardising with z-scores, reading areas, the Central Limit Theorem and its conditions, and normal-approximation confidence intervals and hypothesis tests.
T6 · Inference for proportions
Lock5 Ch 6.1The SE formula for a proportion, CI and HT for one proportion under the CLT conditions, the difference in two proportions, and the pooled-proportion two-sample z-test under the null of equal proportions.
T7 · Inference for means: one-, two-sample and paired
Lock5 Ch 6.2 to 6.5The t-distribution and degrees of freedom, CI and HT for one mean, the difference in two independent means, and paired data (why pairing removes between-subject variability, shrinks the SE and raises power).
T8 · Simple linear regression
Lock5 Ch 2.6, 9.1 to 9.2The least-squares line, slope and intercept, residuals, interpreting slope and intercept and the extrapolation warning, inference for the slope (t with df = n − 2), the correlation test, and goodness of fit (R squared = r squared, the ANOVA decomposition).
T9 · Probability
Lock5 Ch P.1 to P.2Sample spaces and events, the probability axioms, the complement and addition rules, conditional probability and the multiplication rule, independence versus mutual exclusivity, the law of total probability, and Bayes' rule (the false-positive paradox).
T10 · Random variables and distributions
Lock5 Ch P.3 to P.5Discrete random variables and the pmf, expected value and the expected-value rules, variance and SD of a random variable, the binomial distribution (mean np, SD root np(1 − p)), and continuous random variables with the pdf and the uniform distribution.
T11 · Covariance and statistical theory
Lock5 Ch P (capstone)Population covariance and correlation, the covariance and variance rules (with proofs), independence implies zero covariance, and the estimator-versus-estimate double structure (E of X-bar = mu, Var of X-bar = sigma squared over n) that is the algebraic root of the CLT and bridges to ECMT1020.
How it's assessed
Assessment structure
| Component | Weight | Format & timing |
|---|---|---|
| Weekly workshops and muddy cards | 10% | Weekly workshops (Workshops 1 to 12) with a muddy-card feedback loop. Weekly across the semester. Weight is an ESTIMATE subject to confirmation against the official unit outline (no assessment page was captured in the source). |
| Individual assignment | 20% | Individual data-analysis assignment using StatKey and Excel. Set around Week 9, due late in semester (the captured S1 run: 11:59pm Sydney time, Sunday of late May); subject to confirmation. Weight is an ESTIMATE subject to confirmation against the official unit outline. |
| In-semester test | 20% | Closed-book, in-person, 1 hour writing plus 5 minutes reading: 35 multiple-choice questions on a generalised answer sheet; non-programmable calculator permitted; the formula sheet and distribution tables are supplied in the paper. Covers Weeks 1 to 6. Held in Week 7 (workshops run that week but no lecture); date and time subject to confirmation. FORMAT is hard-grounded from the captured MST past papers; the WEIGHT is an estimate subject to confirmation. |
| Final exam | 50% | In-person, supervised. The captured S2 2024 paper has a Section B of short-answer problems worth 50 marks (suggested time about 100 minutes) and almost certainly a Section A of MCQ; the formula sheet plus N(0,1) and t critical-value tables are supplied in the paper; non-programmable calculator permitted, closed-book otherwise. Covers the whole unit. Formal examination period at the end of semester. FORMAT is grounded from the captured S2 2024 final paper; the WEIGHT is an estimate subject to confirmation. |
- Pass on a weighted average of at least 50%. No single-component hurdle is stated in the materials reviewed (subject to confirmation against the official unit outline).
- The in-semester test is 35 MCQ over Weeks 1 to 6. The final adds a short-answer problem section (about 50 marks in the captured S2 2024 paper) on top of a likely MCQ section, covering the whole unit. The displayed component weights are estimates subject to confirmation. Because the paper supplies the formula sheet and the N(0,1) and t tables, marks are in choosing the right test, defining notation and stating the hypotheses, reading the table for the right df and tail, and writing the one-sentence conclusion in context, not in formula recall.
- Calculator policy: Non-programmable calculator permitted in both the in-semester test and the final exam. Both are closed-book otherwise; the formula sheet and distribution tables are bound into each paper.
This is an exam-cram unit. With the exams at 70% of the grade and the final exam alone at 50%, your result is overwhelmingly decided by how well you perform under time pressure. FORMAT is grounded from the captured S2 2024 final paper; the WEIGHT is an estimate subject to confirmation.
Final exam timing: approx Nov 2026 (S2 2026 offering, confirm against the official exam timetable). Confirm the exact date and venue on the official exam timetable.
How to actually pass it
A weekly rhythm, two checklists, and the traps to avoid
The unit 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
- Drill Weeks 1 to 6 under closed-book, 1-hour, 35-MCQ conditions with only a non-programmable calculator, since that is exactly the in-semester test format.
- Be fast on descriptive statistics (mean, SD with the n − 1 divisor, z-scores, the five-number summary and the 1.5 times IQR rule) and on correlation r.
- Practise reading a bootstrap distribution and constructing statistic plus or minus 2 SE confidence intervals, and interpreting a CI correctly (the procedure, not a probability about one interval).
- Rehearse the randomization-test logic and the p-value definition, and the CI-to-HT equivalence, until they are instant.
Before the final heaviest topics
- Cover the whole unit: the final is comprehensive, so the regression, probability, random-variable and Week-13 theory blocks all matter, not just the inference toolkit.
- Work all available past finals timed with the provided formula sheet and tables, then check method; the real paper combines two skills per question (for example an independent and a paired test on the same data, then "why do they differ?").
- Drill the proportion and mean inference families side by side so you never confuse the SE for a proportion, the pooled two-proportion SE, the one-mean t, and the two-mean t.
- Rehearse Bayes' rule and the false-positive paradox, the expected-value and variance rules, the binomial, and the estimator results (E of X-bar = mu, Var of X-bar = sigma squared over n), since the Weeks 11 to 13 block is examinable and easy to under-prepare.
The mistakes that cost marks
Choosing the wrong test. Most lost marks come from running a one-mean t-test on two-sample data, forgetting to pair, or treating a proportion question as a mean question. Before computing anything, classify: how many groups, mean or proportion, paired or independent, or a regression slope. The right test drives everything downstream.
Forgetting the /n inside the standard error. The SE for a proportion is root[ p-hat(1 − p-hat)/n ], not root[ p-hat(1 − p-hat) ]. Dropping the /n inflates the SE wildly and can produce a confidence interval that runs below 0 or above 1, which is impossible for a probability and is the quick sanity check that flags the slip.
Misinterpreting the confidence interval. A 95% CI means the procedure captures the parameter in 95% of all samples (19 in 20), not that there is a 95% chance the parameter is in this one fixed interval. Short-answer questions reward the correct wording, and the wrong wording loses the interpretation mark.
Skipping the hypotheses and the conclusion. Jumping straight to the test statistic skips the marks for stating H0 and Ha in population parameters and for the one-sentence conclusion in context. The exam supplies the formula sheet precisely so it can reward setup and interpretation, so write both out every time.
Teaching team
Who teaches ECMT1010
The bios below are factual. The star ratings are not ours: they are impressions from students who have taken the unit, so you can hear from people who sat in the lectures.
Tim Fisher
Named lecturer for ECMT1010 Introduction to Economic Statistics in the School of Economics, University of Sydney.
Teaching team as listed in the unit materials reviewed. AskSia does not rate lecturers; star ratings are submitted by students who have taken ECMT1010.
Where it fits
Prerequisites, related units & why it matters
No formal economics prerequisite; ECMT1010 is a year-1 gateway unit assuming HSC-level mathematics. It is the assumed-knowledge foundation for the econometrics sequence (ECMT1020 and beyond) and for quantitative economics and finance majors. BUSS1020 and DATA1001 are parallel first-year statistics units for the commerce and science audiences; check the handbook for any overlap or prohibited-combination rules.
Your ECMT1010 study toolkit
Study the unit with Sia, not just read about it
Each tool already knows ECMT1010: 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 ECMT1010 hard?
It has a reputation as one of the harder first-year units at the University of Sydney; its own Week 2 page says exactly that and recommends 6 to 8 hours of private study a week on top of the lecture and workshop. The difficulty is the two-pronged structure (a simulation track of bootstrap and randomization in StatKey running alongside a formula and CLT track) plus a probability-theory turn in Weeks 11 to 13. It is very manageable with consistent weekly practice, because the exams hand you the formula sheet and tables and reward setup and interpretation, not memorisation.
How is ECMT1010 assessed?
Through an in-semester test in Week 7 (35 multiple-choice questions over Weeks 1 to 6, closed-book, 1 hour), an individual data-analysis assignment using StatKey and Excel, and a final exam covering the whole unit (with a short-answer problem section worth about 50 marks in the captured past paper). The exact component weights were not in the materials reviewed, so they are subject to confirmation against the official unit outline; you pass on a weighted average of at least 50%.
What is the final exam format?
In-person and supervised. The captured S2 2024 paper has a Section B of short-answer problems worth 50 marks (suggested time about 100 minutes) and almost certainly a Section A of multiple-choice. The formula sheet and the N(0,1) and t critical-value tables are bound into the paper, and a non-programmable calculator is permitted. It covers the whole unit, so every chapter is examinable.
What software does ECMT1010 use?
StatKey (from lock5stat.com) for the bootstrap, randomization tests and distribution areas, and Excel (with the Data Analysis ToolPak) for exploratory data analysis and regression. You do not need to code; StatKey is point-and-click, and the assignment is a guided data-analysis task.
Do I need to buy the textbook?
No. The set text is Statistics: Unlocking the Power of Data by Lock, Lock, Lock, Lock and Lock (the Lock5 book), and it is provided free to ECMT1010 students through Wiley Course Resources. The weekly pages map each topic to specific Lock5 chapters.
How do I not lose marks on the exam?
The marks are in four places: choosing the right test (one mean, a difference in means, paired, one proportion, a difference in proportions, or a regression slope), stating the hypotheses in population parameters with notation defined, reading the supplied t or N(0,1) table for the correct degrees of freedom and tail, and writing the one-sentence conclusion in context ("weak or strong evidence that..."). Because the formula sheet is provided, the setup and interpretation, not the formula, are where students gain or lose marks.
Study ECMT1010 with Sia
Work through inference for means: one- and the rest of the unit with a tutor that knows it and quizzes you on the topics the assessments weight most heavily.
Start studying with Sia