Clever Grades

🎧 Read Aloud

Understanding Correlation

Core Definition

What is it?

Correlation is a statistical technique that measures the strength and direction of a relationship between two variables, known as co-variables.

Crucially, unlike experiments, correlations do not involve manipulation and do not infer causality.

Types of Relationships

➕

Positive Correlation

Both variables increase or decrease together (e.g., height and weight).
➖

Negative Correlation

One variable increases as the other decreases (e.g., stress and health).
0️⃣

Zero Correlation

No relationship exists between the variables.

The Correlation Coefficient (r)

r ∈ [-1, +1]
Correlation is expressed by a coefficient (r), which ranges from -1 (perfect negative) to +1 (perfect positive). 0 means no correlation.

Strengths of Correlational Analysis

Correlations are valuable tools for preliminary research and unethical scenarios.

✓

Unmanipulatable Variables

Allows study of variables that cannot be manipulated.
✓

Hypothesis Generation

Provides direction for future causal research.
✓

Retrospective Analysis

Can be used to analyze data retrospectively.

Primary Uses

Three contexts where correlation is the appropriate technique:

1

Ethical Investigation

To investigate relationships where manipulation is impossible or unethical (e.g., impact of smoking on health).
2

Research Hypotheses

To generate hypotheses for further experimental research.
3

Natural Data

To analyze naturally occurring data.

Correlation vs. Experiment

✅
Correlation Fact Correlation does not involve manipulation of variables; experiments do.
❌
Causation Limit Correlation cannot establish causation, only association.
✅
Experimental Advantage Experiments have greater control over confounding variables and allow causal inference.

Limitations of Analysis

Key weaknesses inherent in correlational research design:

1

No Cause-and-Effect

Cannot establish cause-and-effect relationships.
2

Confounding Variables

Prone to confounding variables that may influence both co-variables.
3

Coincidental Findings

Correlation may be coincidental, particularly in small samples.
Correlation Flashcards
Term
Correlation

What does correlation measure?

Answer
Definition

The strength and direction of a relationship between two variables.

Term
Positive Correlation

What is a positive correlation?

Answer
Definition

When both variables increase or decrease together.

Term
Negative Correlation

What is a negative correlation?

Answer
Definition

When one variable increases as the other decreases.

Term
Correlation Coefficient +1

What does a correlation coefficient of +1 indicate?

Answer
Definition

A perfect positive correlation.

Term
Correlation Coefficient 0

What does a correlation coefficient of 0 indicate?

Answer
Definition

No correlation between the variables.

Term
Correlation vs Experiment

What is the difference between correlation and an experiment?

Answer
Difference

Correlation does not manipulate variables and cannot prove causation, experiments manipulate variables and can establish cause-and-effect.

Term
Using Correlation vs Experiment

Why might correlation be used instead of an experiment?

Answer
Reason

When manipulation is impossible or unethical.

Term
Limitation of Correlation

Name one limitation of correlational analysis.

Answer
Limitation

It cannot establish cause and effect.

Term
Confounding Variable

What is a confounding variable?

Answer
Definition

A variable that influences both co-variables and may distort the correlation.

Tip: Tap a card to flip. Shuffle or Reset anytime.

📊 Correlational Analysis Quiz

1. What does a correlation coefficient of -0.8 indicate?

The coefficient is close to -1, showing a strong negative relationship.

2. Which of the following is TRUE about correlation?

Correlation shows association but does not prove causation.

3. When is correlational analysis particularly useful?

Correlation is used when experimentation is unethical.

4. Which statement about zero correlation is correct?

Zero correlation means no association between variables.

5. What does correlational analysis NOT control for?

Correlation cannot control for variables influencing both co-variables.

📊 Results

  • Notes and Quiz
  • Psychology
  • AS & A Level
  • Research Methods
  • AQA