The four AO3 skill areas (cartographic, graphical, numerical and statistical, and fieldwork and geospatial); descriptive statistics; and correlation and significance tests such as Spearman's rank.

What this dot point is asking

Eduqas wants you to command the four AO3 skill areas (cartographic, graphical, numerical and statistical, and fieldwork and geospatial), use descriptive statistics, and apply correlation and significance tests such as Spearman's rank, interpreting the result.

The answer

The four AO3 skill areas

The skills are not a separate topic but run through the whole course. Cartographic skills cover reading and constructing maps of many kinds, OS maps, choropleth, isoline, flow-line and proportional-symbol maps, and GIS outputs. Graphical skills cover constructing and interpreting line, bar, scatter, logarithmic, compound and triangular graphs, pie charts, population pyramids and Lorenz curves, and identifying trends and anomalies. Numerical and statistical skills cover percentages, rates, indices, descriptive statistics and tests. Fieldwork and geospatial skills cover designing an enquiry, sampling, ethics and risk, and using GIS to handle located data. Choosing the right technique for the data is itself a skill examiners reward.

Descriptive statistics

Before testing a relationship you describe the data. The mean, median and mode capture the typical value (the median is more robust where there are anomalies), and the range and especially the standard deviation capture the spread. A small standard deviation means the data are consistent (clustered near the mean); a large one means they are variable. Using the right measure, and reading what it tells you about the pattern, is a core numerical skill and the foundation for choosing and interpreting a statistical test.

Correlation and significance: Spearman's rank

A statistical test turns a subjective graph reading into an objective measure. Spearman's rank correlation ($r_s$) measures the strength and direction of the relationship between two ranked variables, giving a coefficient between $+1$ (perfect positive), $0$ (none) and $-1$ (perfect negative). The formula is $r_s = 1 - \frac{6\sum d^2}{n(n^2-1)}$, where $d$ is the difference in ranks at each point and $n$ is the number of pairs. You then test significance by comparing $r_s$ to critical values for that $n$: if it exceeds the critical value at the $0.05$ level, you reject the null hypothesis and accept that the relationship is unlikely to be due to chance. Other tests include the Mann-Whitney U test (comparing two samples), chi-squared (comparing observed and expected frequencies) and nearest-neighbour analysis (measuring how clustered or dispersed points are).

Examples in context

Example 1. Spearman's rank in a coastal investigation. A coastal enquiry testing whether pebble size decreases along a beach is the classic use of Spearman's rank. The student ranks distance and mean pebble size at each site, computes $r_s$ with $r_s = 1 - \frac{6\sum d^2}{n(n^2-1)}$, and interprets the coefficient's strength and direction, then checks it against critical values to judge significance. A significant negative $r_s$ supports the hypothesis that attrition and selective transport reduce pebble size downdrift. This is the standard Eduqas statistical application and the reason the skill belongs with the independent investigation.

Example 2. Choosing graphical and cartographic techniques. A human-geography enquiry into deprivation across a town deploys several AO3 techniques: a choropleth map (cartographic) to show the spatial pattern of a deprivation index, a scatter graph (graphical) to explore the relationship between two variables, descriptive statistics (numerical) such as the mean and standard deviation of an environmental quality score, and a Mann-Whitney U test to compare two districts. Selecting the technique that suits each data type, a choropleth for spatial data, a scatter for correlation, is itself a marked skill, showing that AO3 is about judgement as well as calculation.

Try this

Q1. Write the formula for Spearman's rank correlation and define its terms. [2 marks]

• Cue. $r_s = 1 - \frac{6\sum d^2}{n(n^2-1)}$, where $d$ is the difference in ranks of the two variables at each point and $n$ is the number of pairs.

Q2. Explain why a significance test is needed after calculating a correlation coefficient. [3 marks]

• Cue. A coefficient shows the strength and direction of a relationship, but a significance test (comparing it to critical values) judges whether the relationship is likely to be real rather than due to chance, so the conclusion (rejecting or accepting the null hypothesis) is defensible.