Welcome to Correlation Vocabulary and Interpretation
In the world of statistics, we often want to know if two things are related. For example, does the amount of time you spend revising affect your exam score? Does the temperature outside affect how many ice creams are sold? Correlation is the word we use to describe the relationship between two variables. In this chapter, you will learn the specific "statistical language" needed to describe these relationships and how to explain what they mean in real-life situations.
1. Setting the Stage: Explanatory and Response Variables
Before we can talk about correlation, we need to know which variable goes where on a scatter diagram. When we look at two variables, we usually think one might "explain" the other.
The Explanatory Variable (Independent Variable): This is the variable that we think might be causing a change. It is always plotted on the horizontal \(x\)-axis.
Memory Trick: The word eXplanatory has an X in it!
The Response Variable (Dependent Variable): This is the variable that "responds" to the change. It is always plotted on the vertical \(y\)-axis.
Example: If you are investigating if "Study Time" affects "Test Scores," Study Time is the explanatory variable (\(x\)-axis) and Test Score is the response variable (\(y\)-axis).
2. The Three Directions of Correlation
When you look at a scatter diagram, the first thing you should check is the general direction the points are heading. There are three main types:
Positive Correlation
As the value of the \(x\) variable increases, the value of the \(y\) variable also increases. On a graph, the points trend upwards from left to right.
Real-world example: The taller a person is, the larger their shoe size tends to be.
Negative Correlation
As the value of the \(x\) variable increases, the value of the \(y\) variable decreases. On a graph, the points trend downwards from left to right.
Real-world example: The more miles a car has driven, the lower its resale value tends to be.
Zero Correlation
There is no visible pattern or relationship between the two variables. The points are scattered randomly.
Real-world example: There is likely zero correlation between your house number and your score on a maths test.
Key Takeaway: Direction tells you whether the variables are moving together (Positive) or in opposite directions (Negative).
3. Describing the Strength (By Inspection)
In your exam, you might be asked to describe the strength of the correlation "by inspection." This just means "by looking at it."
Strong Correlation: The points lie very close to a straight line. It is very easy to see the pattern.
Weak Correlation: The points follow a general direction, but they are more spread out and further away from where a line of best fit would be.
Don't worry if this seems subjective! In the exam, the patterns are usually clearly "strong" or "weak." If the points are tightly packed, call it strong. If they are loosely grouped but still showing a direction, call it weak.
4. How to Interpret Correlation in Context
This is a favorite exam question (AO2). When asked to interpret correlation, you must do more than just write "positive" or "negative." You must explain what that means using the words from the question.
The "Golden Rule" for Interpretation:
Use this sentence structure:
"As [Variable X] increases, [Variable Y] tends to [increase/decrease]."
Example Question: A scatter diagram shows a strong negative correlation between "Daily Temperature" and "Gas Used for Heating." Interpret this correlation.
Good Answer: "There is a strong negative correlation. This means that as the daily temperature increases, the amount of gas used for heating tends to decrease."
Quick Review: Correlation Vocabulary
- Positive: Both go up together.
- Negative: One goes up, the other goes down.
- Zero: No relationship.
- Strong: Points are close to a line.
- Weak: Points are widely spread.
5. Important Limits of Correlation
While correlation is useful, there are two things you must remember for your Pearson Edexcel exam:
1. Correlation does not imply causation: Just because two things are correlated doesn't mean one causes the other. For example, ice cream sales and shark attacks are positively correlated (both go up in summer), but eating ice cream does not cause shark attacks! This is often called spurious correlation. (We will cover this more in the next chapter).
2. Linear models only: At this level, we only focus on linear correlation (relationships that look like a straight line). If the points form a curve, we do not use the vocabulary of "strong/weak linear correlation" in the same way.
Common Mistake to Avoid:
Don't confuse "No Correlation" with "Negative Correlation." Negative correlation is still a strong relationship, it just goes downwards! Zero correlation looks like a "cloud" of dots with no direction at all.
Summary Checklist
✓ Can you identify the explanatory (\(x\)) and response (\(y\)) variables?
✓ Can you identify positive, negative, and zero correlation from a diagram?
✓ Can you describe a correlation as strong or weak by looking at it?
✓ Can you write a sentence interpreting the correlation in the context of a real-life problem?
Note: In other chapters, you will learn how to use numbers like Spearman's Rank or PMCC to measure this strength more accurately, and how to draw a Line of Best Fit to make predictions.