University of Sydney · S1 2026 · FACULTY OF STATISTICS

QBUS5001 Foundation in Data Analytics for Business

- one subject, every graph, every model, every mark

11 Chapters113-page Bible

Our own words - no uploaded lecturer files

Built to mirror S1 2026 · updated this semester

The Complete Exam Bible · S1 2026

QBUS5001 Overview

Foundation in Data Analytics for Business

— one subject, every concept, every calculation, every mark

University of Sydney
S1 2026
11 chapters
Statistics

QBUS5001 Foundation in Data Analytics for Business is the postgraduate statistics engine of the University of Sydney's Business School Masters programs — it carries you from sourcing and visualising data, through probability, distributions, sampling and the Central Limit Theorem, to confidence intervals, hypothesis testing and regression, all executed in Microsoft Excel.

Assessment is dominated by a closed-book final exam worth 50% that covers all of Modules 1 to 12 with a supplied 8-page formula sheet; the final is a hurdle — you must score at least 40 out of 100 on it and reach an overall mark of at least 50 to pass.

The skill the exam rewards is method discipline: pick the right test or distribution, run the formula sheet correctly to 4 decimal places, and interpret the number in business context.

Assessed by Weekly Homework (Mindtap/Aplia) 10% · In-Semester Test 25% · Group Assignment 15% · Final Exam 50% — Final Exam is a hurdle
Key terms Central Limit Theorem (CLT), Confidence interval, p-value, Type I / Type II error
How to prepare Treat the formula sheet as the course map: print it early and, for every formula…
Most asked Is the QBUS5001 final exam open or closed book?

QBUS5001 · University of Sydney

An independent, AskSia-authored study guide. AskSia is not affiliated with, endorsed by, or sponsored by University of Sydney; the course code and name are used for identification only.

Assessment

How QBUS5001 is assessed

Component	Weight	Format
Weekly Homework (Mindtap/Aplia)	10%	10 online homeworks, best 5 count; due Sundays from Week 2
In-Semester Test	25%	Closed-book, 1h30m; covers Modules 1–5; held Week 7; formula sheet supplied
Group Assignment	15%	Due Week 12
Final Exam · hurdle	50%	Closed-book; covers Modules 1–12; supplied 8-page formula sheet; ≥40/100 hurdle

Contents · every chapter, one map

What QBUS5001 covers

The whole subject → one exam-ready map. Each topic links to its free chapter guide.

Data, Visualisation & Ethics

Sourcing, classifying and communicating business data responsibly (Module 0)

Module 0 sets the foundation before any calculation: how to source, classify, visualise and communicate business data responsibly.

Data types: structured vs unstructured, primary vs secondary
Variable classification: categorical vs numerical, cross-section vs time-series
Univariate charts: bar, column, histogram, pie, box-and-whisker
Bivariate charts: the scatter plot
Tufte principles: clarity, data-ink ratio, avoiding distortion
Data communication: matching the chart to the variable type

Descriptive Statistics & Association

Centre, spread and the strength of a linear relationship (Module 1)

Module 1 builds the descriptive toolkit: measures of centre (mean, median, mode), measures of spread (range, variance, standard deviation, IQR, coefficient of variation) and measures of association (covariance and correlation).

Mean: population μ vs sample x̄
Median, mode and when each is preferred
Variance and standard deviation (note the n−1 divisor for samples)
Range and interquartile range (IQR = Q₃ − Q₁)
Coefficient of variation: CV = s/x̄ for unit-free comparison
Sample covariance and its direction-only meaning

Probability & Bayes' Theorem

Quantifying uncertainty and revising beliefs (Module 2)

Module 2 is the grammar of uncertainty: the addition, complement and multiplication rules, conditional probability and independence, and the headline result, Bayes' theorem, which revises a prior probability after new evidence arrives.

Probability interpretations: classical, relative-frequency, subjective
Complement rule: P(not A) = 1 − P(A)
Addition rule and mutual exclusivity
Conditional probability: P(A|B) = P(A∩B)/P(B)
Multiplication rule and statistical independence
Total probability rule for partitions

Random Variables & Discrete Distributions

Expected value, variance, portfolios, Binomial & Poisson (Module 3)

Module 3 moves from events to random variables: expected value and variance of a discrete distribution, the covariance of two random variables, and the rules for linear transforms and sums that power the portfolio application (where negative correlation reduces risk below the weighted-average SD).

Expected value: μ = E[X] = Σ xᵢ p(xᵢ)
Variance: σ² = E(X²) − [E(X)]²
Covariance and correlation of random variables
Linear transforms: E(aX) = aE(X), V(aX) = a²V(X)
Sum of two RVs: V(aX+bY) = a²V(X) + b²V(Y) + 2ab·Cov(X,Y)
Portfolio expected return and risk; diversification

Continuous Distributions & the Normal

Uniform, Exponential, Normal and Z standardisation (Module 4)

Module 4 covers the continuous distributions used to model business quantities: the Uniform (equal density over an interval), the Exponential (waiting times, rate λ) and, above all, the Normal distribution.

Uniform distribution: f(x) = 1/(b−a), mean (a+b)/2, variance (b−a)²/12
Probabilities as areas (proportions of the interval)
Exponential distribution: rate λ, cdf 1 − e⁻ᴻˣ, mean 1/λ
Normal distribution N(μ, σ²) and its shape
Standard Normal Z ~ N(0,1)
Z-standardisation: Z = (X − μ)/σ

Sampling Distributions, CLT & Estimation

From sample to population: CLT, LLN and confidence intervals (Modules 5-6)

Modules 5 and 6 connect samples to populations.

Sampling distribution of x̄: mean μ, standard error σ/√n
Central Limit Theorem: approximately Normal for n ≥ 30
Law of Large Numbers: x̄ → μ as n → ∞
Sampling distribution of a proportion (np ≥ 5, n(1−p) ≥ 5)
Unbiased and consistent point estimators
CI for a mean: σ known → Z, σ unknown → t(n−1)

Hypothesis Testing (One Population)

The 6-step framework for means and proportions (Module 8)

Module 8 formalises decision-making under uncertainty with the six-step framework: state H₀ and H₁, choose α, compute the test statistic, find the critical value or p-value, decide, and conclude in context.

The six-step hypothesis-testing framework
Null and alternative hypotheses; one- vs two-tailed
Significance level α and the rejection region
Test for a mean, σ known: Z = (x̄ − μ₀)/(σ/√n)
Test for a mean, σ unknown: T = (x̄ − μ₀)/(s/√n) ~ t(n−1)
Test for a proportion: Z = (p − π₀)/√(π₀(1−π₀)/n)

Two-Population Inference

A/B testing: comparing two means, proportions and variances (Module 9)

Module 9 extends testing to two populations — the statistical engine behind A/B testing.

A/B testing as a two-population comparison
Two means, variances known → Z
Two means, variances unknown and equal → pooled t (df = n₁+n₂−2)
Two means, variances unknown and unequal → Welch t
Pooled variance formula s_p²
Two proportions with a pooled proportion estimate

Simple Linear Regression

Estimating, interpreting and assessing a two-variable model (Module 10)

Module 10 fits a straight line to two variables: the least-squares estimates b₁ = Σ(xᵢ−x̄)(yᵢ−ȳ)/Σ(xᵢ−x̄)² and b₀ = ȳ − b₁x̄, giving the fitted line Ŷ = b₀ + b₁X.

Population model E(Y|X) = β₀ + β₁X and the error term
Least-squares slope b₁ and intercept b₀
Fitted values Ŷ and residuals e = Y − Ŷ
Interpreting the slope and intercept in business terms
Sums of squares: SST = SSR + SSE
R² = SSR/SST as the proportion of variation explained

Regression Diagnostics & Inference

L.I.N.E. assumptions, slope tests, intervals and autocorrelation (Module 11)

Module 11 asks whether a fitted line can be trusted. The four L.I.N.E. assumptions — Linearity, Independence of errors, Normality of errors, Equal variance — are checked through residual plots.

The four L.I.N.E. assumptions and residual-plot checks
t-test on the slope: T = (b₁ − β₁)/s(b₁) ~ t(n−2)
Confidence interval for the slope
t-test on the correlation coefficient
Confidence interval for mean Y at a given X
Prediction interval for an individual Y

Multiple Linear Regression & Dummy Variables

Many predictors, model comparison, F-tests and categorical data (Module 12)

Module 12 generalises regression to several predictors: Ŷ = b₀ + b₁X₁ + … + b_kX_k.

The MLR model and estimated equation Ŷ = b₀ + b₁X₁ + … + b_kX_k
Interpreting a coefficient “holding other variables fixed”
Coefficient of multiple determination R²
Adjusted R² and comparing competing models
Overall F-test: F = MSR/MSE ~ F(k, n−k−1)
Individual coefficient t-tests on n−k−1 df

Worked example · free

One-sample t-test on a mean (with confidence interval)

Q [8 marks]. A coffee chain claims the mean fill of its cups is 250 ml. A random sample of n = 36 cups gives a sample mean of x̄ = 246 ml with sample standard deviation s = 9 ml. Test at the 5% significance level whether the true mean fill differs from 250 ml, and confirm the conclusion with a 95% confidence interval. Use t(0.025, 35) ≈ 2.030.

1 markState the hypotheses. H₀: μ = 250 versus H₁: μ ≠ 250 (two-tailed, because “differs” gives no direction).
1 markChoose the test. σ is unknown and estimated by s, so use the t-statistic with df = n − 1 = 35.
1 markCompute the standard error: s/√n = 9/√36 = 9/6 = 1.5 ml.
2 marksCompute the test statistic: T = (x̄ − μ₀)/(s/√n) = (246 − 250)/1.5 = −4/1.5 = −2.6667.
1 markDecision rule. Reject H₀ if |T| > t(0.025, 35) ≈ 2.030. Here |−2.6667| = 2.6667 > 2.030, so reject H₀.
1 markConfirm with the 95% CI: 246 ± 2.030 × 1.5 = 246 ± 3.045 = [242.96, 249.05]. The claimed value 250 lies outside this interval — the same conclusion.
1 markConclude in context: at the 5% level there is sufficient evidence that the true mean fill differs from (is below) 250 ml.

T = −2.6667, which exceeds the critical value 2.030 in absolute terms, so reject H₀. The 95% CI [242.96, 249.05] excludes 250, confirming the result. There is significant evidence the mean fill differs from 250 ml.

Sia tip — The CI and the two-tailed test always agree: if the hypothesised value sits outside the (1−α) interval, you reject H₀ at α. Quoting both earns the interpretation marks cleanly.

Glossary

Key terms

Central Limit Theorem (CLT): For a sample of size n ≥ 30, the sampling distribution of the sample mean x̄ is approximately Normal regardless of the population shape, with mean μ and standard error σ/√n. It is what licenses Normal-based inference on non-Normal data.
Confidence interval: An interval of the form point estimate ± (critical value)(standard error) that, over repeated sampling, contains the true parameter (1−α) of the time — e.g. a 95% CI. It is a statement about the procedure, not a probability about the fixed parameter.
p-value: The probability, if H₀ were true, of observing a test statistic at least as extreme as the one obtained. Reject H₀ when the p-value ≤ α.
Type I / Type II error: A Type I error rejects a true H₀ (probability α); a Type II error fails to reject a false H₀ (probability β). Power = 1 − β.
R² (coefficient of determination): R² = SSR/SST, the proportion of variation in Y explained by the regression model. Adjusted R² penalises extra predictors so models with different numbers of variables can be compared fairly.
Standard error: The standard deviation of a sample statistic's sampling distribution — for the mean it is σ/√n. It shrinks as n grows, which is why larger samples give tighter estimates.

FAQ

QBUS5001 FAQ

Is the QBUS5001 final exam open or closed book?

Closed book. You are given a supplied 8-page formula sheet (printed separately and not removable from the venue) and may use a non-programmable calculator, but you cannot bring your own notes. The final covers Modules 1 to 12.

What is the pass hurdle for QBUS5001?

Two conditions must both hold: you must score at least 40 out of 100 on the final exam, and your overall mark across all assessment tasks must be at least 50. The final exam is therefore a hurdle as well as the largest single weight (50%).

What does the in-semester test cover and when is it?

The in-semester test is worth 25%, runs for 1 hour 30 minutes closed book, and covers Modules 1 to 5 (descriptive statistics through sampling distributions and the CLT). It is held in Week 7, with a formula sheet supplied.

Do I need to memorise the formulas?

No — a formula sheet is provided in both the in-semester test and the final. The marks come from choosing the correct formula for the scenario, substituting carefully (keep 4 decimal places for non-integer results), and interpreting the answer in business terms.

Which software does the course use?

Microsoft Excel 365 with the Analysis ToolPak add-in. You will use functions such as NORM.DIST, NORM.S.DIST, BINOM.DIST, POISSON.DIST, T.INV.2T, CORREL and the Data Analysis → Regression tool. Weekly homework is done on Cengage Mindtap/Aplia.

Chapter questions

Questions students ask in each QBUS5001 chapter

Is there much maths in Module 0?: No — it is conceptual. The value is in correctly classifying variables (which drives every later technique) and in the vocabulary of visualisation and ethics, which the group assignment and short-answer exam questions draw on. Chapter 01 · Data, Visualisation & Ethics
Why divide by n−1 instead of n for the sample variance?: Because the sample mean is itself estimated from the data, dividing by n−1 (the degrees of freedom) corrects the downward bias and makes s² an unbiased estimator of σ². Population variance, where μ is known, divides by N. Chapter 02 · Descriptive Statistics & Association
How do I know when to use Bayes' theorem?: Use it when you are given the reverse conditional and asked for the other direction — e.g. you know P(alert | fraud) but want P(fraud | alert). The given direction plus a prior and a false-positive rate is the signature of a Bayes question. Chapter 03 · Probability & Bayes' Theorem
How do I decide between the Binomial and the Poisson?: Use the Binomial when there is a fixed number of trials n and you count successes (e.g. 5 conversions out of 12 emails). Use the Poisson when you count events in a continuous interval of time or space with no fixed upper limit (e.g. chat requests per 30 minutes). Chapter 04 · Random Variables & Discrete Distributions
Should I use NORM.DIST or NORM.S.DIST?: NORM.DIST(x, μ, σ, 1) works directly on the raw value and gives the cumulative probability. NORM.S.DIST(z, 1) works on a standardised Z-score. Both give the same answer; standardise first only if the question asks for the Z-score or if you are reading a printed Z-table. Chapter 05 · Continuous Distributions & the Normal
Should I divide by σ or by σ/√n when standardising a sample mean?: By the standard error σ/√n. A single observation X uses σ; a sample mean x̄ uses σ/√n because averaging reduces variability. Mixing these up is the classic CLT error. Chapter 06 · Sampling Distributions, CLT & Estimation
How do I decide one-tailed versus two-tailed?: The alternative hypothesis decides. “Differs from” gives a two-tailed test (≠); “greater than” or “less than” gives a one-tailed test (> or <). Reading the question wording carefully here is worth a mark and changes the critical value. Chapter 07 · Hypothesis Testing (One Population)
How do I know whether to pool the variances?: Read the question for the variance assumption. If it says variances are equal (or an F-test fails to reject equality), use the pooled t. If variances are unknown and unequal, use the Welch t. If both σ are given, use the Z version. Chapter 08 · Two-Population Inference
How do I interpret the intercept b₀?: It is the expected value of Y when X = 0. Often this is outside the range of the data (e.g. zero advertising), so treat it as a mathematical anchor for the line rather than a meaningful business prediction unless X = 0 is realistic. Chapter 09 · Simple Linear Regression
What is the difference between a confidence interval and a prediction interval here?: The confidence interval estimates the mean of Y at a given X; the prediction interval estimates a single new Y at that X. The prediction interval is always wider because it includes the variability of an individual observation (the “1 +” term under the root). Chapter 10 · Regression Diagnostics & Inference
How many dummy variables do I need for a categorical predictor?: For a category with c levels you need c−1 dummies. For example, a 4-season variable needs 3 dummies; the omitted level becomes the reference category against which the others are compared. Using all c dummies causes the dummy-variable trap. Chapter 11 · Multiple Linear Regression & Dummy Variables

Study strategy

How to study for the exam

Treat the formula sheet as the course map: print it early and, for every formula, write beside it the one trigger phrase that selects it (“σ unknown → t”, “comparing two layouts → two-sample test”, “counts in a window → Poisson”). Because the exam is closed-book on method, build a one-page decision tree: is the question about a distribution, an interval, a test, or a regression? Then which sub-case?

Drill the in-semester material (Modules 1–5) to fluency by Week 7, since it reappears in the final. Practise every calculation twice — once by hand to 4 decimals and once in Excel — so you trust your numbers under time pressure. Finally, never stop at the number: each answer needs a one-sentence business interpretation, which is where easy marks are won and lost.

Study QBUS5001 with AI

Your AI Statistics tutor for QBUS5001

Stuck on a hard QBUS5001 question? Sia is AskSia’s AI Statistics tutor — ask any QBUS5001 Foundation in Data Analytics for Business 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.

A+Everything unlocked

Unlocks this Bible + all 165 of your University of Sydney subjects - and 1,000+ Bibles across every Australian university.

Sia - your QBUS5001 tutor, unlimited, worked the way the exam marks it

The full 113-page Bible + practice bank with worked solutions

Chrome extension - sync your LMS so Sia knows your deadlines

Bilingual EN / Chinese on every Bible and every Sia answer

$0.99 Trial

30-day money-back · cancel in one tap · how it works