Specialist Mathematics · Unit 4 · Topic 5

Statistical inference

Treat the sample mean as a random variable, use its mean, its spread σn\frac{\sigma}{\sqrt{n}} and the central limit theorem to find probabilities, then build, interpret and plan approximate confidence intervals for a population mean.

  • 8 lessons
  • 1 h 56 min of lessons
  • 10 worked examples
  • 13 hours in the syllabus
0 of 8

Start here

Sample means vary

Lesson 1 of 8 · 13 min

Start the topic

Lessons

8
  1. Sample means vary (up next)

    See the sample mean as a random variable whose value changes from sample to sample, tell statistics from parameters, and find the exact distribution of X̄ for a small population.

  2. The mean and spread of sample means

    Use E(Xˉ)=μE(\bar{X}) = \mu and SD(Xˉ)=σn\mathrm{SD}(\bar{X}) = \frac{\sigma}{\sqrt{n}}, see why larger samples give less variable means, and write Xˉ∼N(μ,σ2n)\bar{X} \sim N\left(\mu, \frac{\sigma^2}{n}\right) when the population is normal.

  3. The central limit theorem

    See sample means from skewed, uniform and two-peaked populations become approximately normal as nn grows, and standardise a sample mean with sn\frac{s}{\sqrt{n}} for n≥30n \geq 30.

  4. Probabilities for sample means

    Calculate probabilities and quantiles for a sample mean with technology or the 68–95–99.7% rule, compare them with probabilities for one value, and handle totals.

  5. Confidence intervals for a mean

    Turn one sample's mean and standard deviation into an approximate confidence interval xˉ±zsn\bar{x} \pm z\frac{s}{\sqrt{n}} for the population mean, choosing the right zz for the confidence level.

  6. Margin of error and sample size

    Calculate the margin of error E=zsnE = z\frac{s}{\sqrt{n}}, read xˉ\bar{x} and EE from an interval, see how nn and the confidence level change EE, and choose the sample size for a required precision.

  7. What 95% confidence means

    Explain the confidence level through repeated sampling, see why most but not all intervals contain μ, avoid the probability misreading, and use an interval to judge a claim.

  8. Intervals for values and for the mean

    Use x̄ and s to estimate an interval that covers a given proportion of individual values of a normal variable, and contrast it with the much narrower confidence interval for μ.

  9. Topic quiz

    Ten questions across the topic, from foundation to complex unfamiliar. Your results feed your mastery.

About this topic

A factory can't weigh every bag of rice it fills, and a researcher can't time every commute in Brisbane. They take a sample, calculate its mean xˉ\bar{x}, and use it to say something about the population mean μ\mu. The catch is that a different sample would give a different xˉ\bar{x}. This topic is about how much sample means vary, and what that variation lets you conclude.

In the first half you treat the sample mean Xˉ\bar{X} as a random variable. Through simulation and a little algebra you find that its mean is μ\mu, its standard deviation is σn\frac{\sigma}{\sqrt{n}}, and, by the central limit theorem, it is approximately normal for large samples whatever the shape of the population. That lets you calculate probabilities about sample means with the normal distribution from Methods.

In the second half you turn that around. From one sample's xˉ\bar{x} and ss you build an approximate confidence interval xˉ±zsn\bar{x} \pm z\frac{s}{\sqrt{n}} for μ\mu, control its margin of error with the sample size and confidence level, interpret it correctly, and contrast it with an interval that covers most of the individual values. It is the same reasoning you used for sample proportions in Methods, now applied to measurements.

Before you start

Builds on Continuous random variables and the normal distribution, Sampling and proportions and Interval estimates for proportions.

  • Normal distributions X∼N(μ,σ2)X \sim N(\mu, \sigma^2), zz-scores, the 68–95–99.7% rule, and normal probabilities and quantiles with a calculator (Methods Unit 4 Topic 3).
  • Populations, random samples, the sample proportion p^\hat{p} and its approximate normal distribution (Methods Unit 4 Topic 4).
  • Approximate confidence intervals and margins of error for a proportion, and the zz values 1.6451.645, 1.9601.960 and 2.5762.576 for 90%, 95% and 99% (Methods Unit 4 Topic 5).
  • Mean, variance and standard deviation of a discrete random variable, and the binomial distribution (Methods Unit 3 Topic 5).
  • Calculating the mean xˉ\bar{x} and standard deviation ss of a data set with a calculator.

What QCAA says you need to know12 dot points

Sample means

7 hours
  • Understand the concept of the sample mean Xˉ\bar{X} as a random variable whose value varies between samples where XX is a random variable with mean μ\mu and the standard deviation σ\sigma.LessonSample means vary
  • Use repeated random sampling data from a variety of distributions and a range of sample sizes to examine properties of the distribution of Xˉ\bar{X} across samples of a fixed size nn, including its mean μ\mu, its standard deviation σn\frac{\sigma}{\sqrt{n}} (where μ\mu and σ\sigma are the mean and standard deviation of XX) and its approximate normality if nn is large.LessonsSample means varyThe mean and spread of sample meansThe central limit theorem
  • Recognise and use the link between the normal distribution of the sample mean and the statistical notation Xˉ∼N(μ,σ2n)\bar{X} \sim N\left(\mu, \frac{\sigma^2}{n}\right).LessonsThe mean and spread of sample meansProbabilities for sample means
  • Use repeated random sampling data from a variety of distributions and a range of sample sizes to examine the approximate standard normality of Xˉ−μsn\frac{\bar{X} - \mu}{\frac{s}{\sqrt{n}}} for large samples (n≥30n \geq 30), where ss is the sample standard deviation (Central limit theorem).LessonThe central limit theorem
  • Model and solve problems that involve sample means, with and without technology.LessonProbabilities for sample means

Confidence intervals for means

6 hours
  • Understand the concept of an interval estimate for a parameter associated with a random variable.LessonsConfidence intervals for a meanWhat 95% confidence means
  • Understand and use the approximate confidence interval (xˉ−zsn,xˉ+zsn)\left(\bar{x} - z\frac{s}{\sqrt{n}}, \bar{x} + z\frac{s}{\sqrt{n}}\right), as an interval estimate for μ\mu, the population mean, where zz is the appropriate quantile for the standard normal distribution.LessonConfidence intervals for a mean
  • Understand and use the approximate margin of error, E=zsnE = z\frac{s}{\sqrt{n}}.LessonMargin of error and sample size
  • Understand and use the relationship between margin of error, level of confidence and sample size.LessonMargin of error and sample size
  • Understand and use the concept that there are variations in confidence intervals between samples and that most but not all confidence intervals contain μ\mu.LessonWhat 95% confidence means
  • Use xˉ\bar{x} and ss to estimate μ\mu and σ\sigma, to obtain approximate intervals covering desired proportions of values of a normal random variable and compare with an approximate confidence interval for μ\mu.LessonIntervals for values and for the mean
  • Model and solve problems that involve interval estimates for sample means, with and without technology.LessonsWhat 95% confidence meansIntervals for values and for the mean

Subject matter from the Specialist Mathematics General senior syllabus 2025 v1.4 (opens in a new tab), CC BY 4.0 © State of Queensland (QCAA).

The focus areas you can practise on their own. Each one maps to dot points above.

  • The sample mean as a random variable

    Tell a statistic from a parameter, explain why x̄ varies between samples, and find the exact distribution of X̄ for a small population.

    Practise
  • Mean and standard deviation of X̄

    Use E(X̄) = μ and SD(X̄) = σ/√n, write X̄ ~ N(μ, σ²/n), and predict how the spread changes with n.

    Practise
  • The central limit theorem

    Decide when X̄ is approximately normal from the population shape and n, and standardise with (x̄ − μ)/(s/√n) for n ≥ 30.

    Practise
  • Probabilities for sample means

    Calculate probabilities and quantiles for X̄ with technology, or with the 68–95–99.7% rule without it, and compare them with probabilities for one value.

    Practise
  • Confidence intervals for a mean

    Choose z for a confidence level and calculate the approximate interval x̄ ± z s/√n for μ, using σ when it is known.

    Practise
  • Margin of error for a mean

    Calculate E = z s/√n, read x̄ and E from an interval, and work back to s or the confidence level.

    Practise
  • Margin of error and sample size

    Find the smallest sample for a required margin of error and explain how n and the confidence level change E.

    Practise
  • Interpreting confidence intervals

    Explain what the confidence level means through repeated sampling, why most but not all intervals contain μ, and judge claims with an interval.

    Practise
  • Intervals for values versus for μ

    Estimate an interval covering a proportion of individual values with x̄ ± z s, and contrast it with the confidence interval for μ.

    Practise
  • Modelling with sample means

    Solve unfamiliar problems with sample means, totals and intervals, state assumptions and evaluate the reasonableness of conclusions.

    Practise

Worked examples10

All 10

Common mistakes

Dividing by nn instead of n\sqrt{n}

The standard deviation of Xˉ\bar{X} is σn\frac{\sigma}{\sqrt{n}}. With σ=12\sigma = 12 and n=36n = 36 it is 22, not 1236\frac{12}{36}. It is the variance that is divided by nn: Var(Xˉ)=σ2n\mathrm{Var}(\bar{X}) = \frac{\sigma^2}{n}.

Using σ\sigma instead of σn\frac{\sigma}{\sqrt{n}} for a sample mean

A question about the mean of a sample needs SD(Xˉ)=σn\mathrm{SD}(\bar{X}) = \frac{\sigma}{\sqrt{n}}. Using σ\sigma answers a question about one individual value instead.

Writing the standard deviation where N(μ,σ2)N(\mu, \sigma^2) needs a variance

Xˉ∼N(μ,σ2n)\bar{X} \sim N\left(\mu, \frac{\sigma^2}{n}\right) has the variance second. With σ=10\sigma = 10 and n=25n = 25, write N(μ,4)N(\mu, 4), and enter the standard deviation 22 in your calculator.

Thinking the population becomes normal

The central limit theorem is about the distribution of Xˉ\bar{X}. Individual values keep the population's shape, and the histogram of one large sample looks like the population, not like a bell.

"There is a 95% probability that μ is in this interval"

μ\mu is fixed and a calculated interval either contains it or not. The 95% describes the method: about 95% of intervals built this way contain μ\mu. Say "we are 95% confident", and explain it through repeated sampling.

Confusing the confidence interval with the range of values

A 95% confidence interval for μ\mu is about 1n\frac{1}{\sqrt{n}} times as wide as the interval holding 95% of individual values. It does not say where 95% of the data lie.

Rounding the sample size down

If n≥157.4n \geq 157.4 then n=157n = 157 is too small. A required sample size is always rounded up.

Wrong z for the confidence level

For 95% the central area is 0.950.95, so zz cuts off 0.0250.025 in each tail: z=1.960z = 1.960, not 1.6451.645 (that is 90%). Use invNorm(1+C2)\left(\frac{1 + C}{2}\right).

Halving n to halve the margin of error

EE is proportional to 1n\frac{1}{\sqrt{n}}, so halving EE needs four times the sample, not twice.

Exam tips

  1. State the distribution you use, with its parameters: "Xˉ∼N(480,92)\bar{X} \sim N(480, 9^2) approximately, by the central limit theorem since n=64n = 64 is large." It earns the method mark and shows you used σn\frac{\sigma}{\sqrt{n}}.

  2. Say why normality is justified: XX is normal (exact), or nn is large (approximate, central limit theorem). If nn is small and XX is not normal, say a normal model may not be reasonable.

  3. The confidence interval and margin of error for a mean are on the formula book, but the zz values are not. Learn 1.6451.645, 1.9601.960 and 2.5762.576, and know how to get others from invNorm.

  4. Paper 1 questions use whole numbers of standard errors with the 68–95–99.7% rule, or z=1.96z = 1.96 with tidy numbers like n=64n = 64 or 100100. Keep the working exact until the end.

  5. In Paper 2, write the calculator command and its inputs, for example "normCdf(lower =−∞= -\infty, upper =470= 470, μ=480\mu = 480, σ=9\sigma = 9) =0.1334= 0.1334".

  6. Interpret an interval in context, and use the word "confident", not "probability": "We are 95% confident that the mean commute time of all Brisbane workers is between 31.5 and 36.9 minutes."

  7. When you judge a claim, say whether the claimed value lies inside the interval, and conclude with care: inside means "plausible", not "proved".

  8. Give sample sizes as whole numbers rounded up, and intervals to the precision the question asks (often 2 decimal places or one more than the data).

Where this shows up

  • Manufacturers test samples of products (bag weights, battery life, bottle volumes) to check that the process mean meets its label.

  • Medical and nutrition researchers estimate mean blood pressure, cholesterol or sleep time with confidence intervals.

  • Environmental scientists estimate mean pollutant concentrations, fish lengths or rainfall from samples.

  • Engineers use the distribution of a sample mean to set safe loads, such as the total mass a lift can carry.

  • Survey designers choose sample sizes so that an estimate of a mean income or travel time has a useful margin of error.

  • Quality-control charts plot sample means against limits μ±3σn\mu \pm 3\frac{\sigma}{\sqrt{n}} to spot a process drifting.

Save your progress with a free account