Scatter graphs and correlation

Scatter graphs appear on every Edexcel Paper 2 or 3 — typically Foundation 4–5 marks, Higher with an "is X correlation evidence Y caused Z" critique.

Drawing a scatter graph

• Plot each (x, y) bivariate data point as a cross (×). Edexcel mark schemes typically award B1 for "all points plotted within ±½ small square".
• Label both axes including units.

Types of correlation

• Positive correlation — as x increases, y tends to increase. Line of best fit slopes upward.
• Negative correlation — as x increases, y tends to decrease. Line slopes downward.
• No correlation — no clear pattern. Do not draw a line of best fit.
• Strong vs weak — strong = points close to a clear line; weak = points scattered around it.

Line of best fit

Draw a single straight line that:

• Passes roughly through the middle of the data (about half the points each side).
• Goes through the mean point (x̄, ȳ) when both means are easy to compute.
• Does not need to pass through (0, 0).

Use a ruler. Free-hand wavy lines lose the C1 communication mark.

Using the line of best fit

• Interpolation — reading a y-value for an x within the data range. Reliable.
• Extrapolation — reading a y-value outside the data range. Unreliable; mention this when asked about reliability.

Outliers

A point that does not fit the trend. Edexcel asks "circle the outlier" B1 and may ask whether you would include it when drawing the line of best fit (typically, exclude it).

Correlation vs causation

This is the Higher exam question. Just because two variables correlate does not mean one causes the other.

Example: scatter graph of "ice cream sales" vs "drowning incidents" shows positive correlation. Cause is a third variable: hot weather. Both rise together but neither causes the other.

A correct exam answer is in three parts:

1. Describe the correlation seen.
2. State the relationship is not causation.
3. Suggest a confounding variable or alternative explanation.

Common Edexcel exam tip

For "is there a relationship?" questions, always quote both the strength and direction: "strong negative correlation" earns 2 marks where "negative" alone earns only 1.