Pearson International A Level Further Mathematics: S3 Statistics 3

S3 is an externally assessed modular unit used within Pearson Edexcel International A Level Further Mathematics. The current Mathematics, Further Mathematics and Pure Mathematics specification is Issue 3. This note follows the official S3 order and keeps qualification cash-in choices separate from the mathematical content: your centre must still confirm that the unit combination is eligible for the award you intend to claim.

Official unit scope

1. Combinations of random variables.
2. Sampling distributions and the central limit idea.
3. Estimation and confidence intervals.
4. Tests for means, proportions and differences.
5. Chi-squared goodness-of-fit and contingency-table tests.

The specification assumes prerequisite knowledge stated for the unit, so later-unit questions may combine earlier methods without re-teaching them. Treat the list above as an integrated toolkit. A question can begin in one topic and finish in another, such as using algebra to form a model, calculus to optimise it and a graph to interpret the result.

Core reasoning and methods

• Distinguish a population parameter, a sample statistic and the sampling distribution of that statistic.
• Match the test procedure to the design, distributional assumptions and parameter being tested.
• For chi-squared work, compute expected frequencies from the null model and combine categories when conditions require it.

Write mathematical arguments so another reader can reproduce every transition. Define symbols that are introduced, preserve exact values until a decimal is requested, and place restrictions beside the step that creates them. Calculator use can support arithmetic, graph exploration and checking, but it does not replace a proof, derivation or required chain of working.

When a model is used, state the simplifying assumptions and interpret the answer in the original setting. A mathematically valid root may be inadmissible because it lies outside a time interval, represents a negative length, violates a probability range or conflicts with a geometric domain. The final check is therefore both algebraic and contextual.

Worked example

A sample mean of 52 is obtained from n = 64 observations with known population standard deviation 8. A 95% confidence interval for the population mean is 52 plus or minus 1.96 times 8 divided by square root 64, giving 52 plus or minus 1.96, or approximately (50.04, 53.96). The interval concerns the unknown population mean, not 95% of individual observations. Its construction is based on the sampling distribution of the mean and the chosen confidence procedure.