Introduction to Correlational Research
Welcome to one of the most useful tools in a psychologist's toolkit! In our previous chapters, we looked at experiments where researchers change one thing to see what happens to another. However, sometimes we can't (or shouldn't) change things. For example, we can't ethically change a person’s genes to see if they become aggressive. This is where correlational research comes in.
In this chapter, we will explore how psychologists look for relationships between variables. Think of correlation as "detective work"—we are looking for patterns that already exist in the world to see how much two things relate to one another.
1. What is a Correlation?
A correlation is a non-experimental method used to measure the relationship between two variables. Unlike experiments, there is no Independent Variable (IV) being manipulated and no Dependent Variable (DV) being measured. Instead, we look at co-variables.
Co-variables
In correlational research, we refer to the two things being measured as co-variables. We call them this because they "co-vary"—as one changes, the other might change too. For example, if we are studying biological psychology, we might look at the relationship between testosterone levels (Co-variable 1) and aggressive behavior (Co-variable 2).
Quick Tip: If an exam question asks you to identify the variables in a correlation, never use the terms "IV" or "DV." Always use the term co-variables!
2. Types of Correlation
There are three ways two variables can relate to each other. Understanding these is key to interpreting psychological data.
Positive Correlation
A positive correlation means that as one co-variable increases, the other co-variable also increases. They move in the same direction.
Example: The more hours you spend revising, the higher your exam score is likely to be.
Negative Correlation
A negative correlation means that as one co-variable increases, the other decreases. They move in opposite directions.
Example: In biological psychology, researchers might find that as the amount of sleep a person gets increases, their stress levels decrease.
Zero Correlation
A zero correlation occurs when there is no relationship at all between the two variables.
Example: There is likely a zero correlation between your shoe size and your ability to remember a list of words.
Key Takeaway:
Positive = Same direction (Up/Up or Down/Down)
Negative = Opposite directions (Up/Down)
3. Visualising Data: Scatter Diagrams
Psychologists use scatter diagrams (also called scatter plots) to see the relationship between co-variables visually. One co-variable is placed on the \(x\)-axis and the other on the \(y\)-axis. Each participant is represented by a single dot on the graph.
How to read a scatter diagram:
- If the dots form a line pointing from the bottom-left to the top-right, it is a positive correlation.
- If the dots form a line pointing from the top-left to the bottom-right, it is a negative correlation.
- If the dots are scattered randomly all over the page like a cloud, it is a zero correlation.
Did you know? The closer the dots are to forming a perfect straight line, the stronger the correlation is!
4. The Golden Rule: Cause and Effect Issues
This is the most important point in this chapter: Correlation does NOT mean causation.
Just because two things are related doesn't mean one caused the other to happen. This is a common mistake students make in exams. Because we aren't controlling the environment (like in a lab experiment), we cannot be sure of the cause-and-effect.
The "Third Variable" Problem
Sometimes, a relationship exists because of a third, hidden variable that we haven't measured (an intervening variable).
Analogy: Imagine a study finds a positive correlation between ice cream sales and drowning incidents. Does eating ice cream cause drowning? No! The third variable is hot weather. Hot weather makes people buy ice cream AND makes people go swimming, which leads to more drownings.
Don't worry if this seems tricky! Just remember: Correlations tell us that a relationship exists, but they don't tell us why it exists.
5. Measuring Strength: Spearman’s Rank
In Topic C (Biological Psychology), you are required to know about Spearman’s Rank Correlation Coefficient. This is a statistical test used to turn a scatter diagram into a single number that tells us exactly how strong the relationship is.
The Correlation Coefficient (\( \rho \))
The result of a Spearman’s test is a number between \( -1 \) and \( +1 \).
- A score of \( +1 \) is a perfect positive correlation.
- A score of \( -1 \) is a perfect negative correlation.
- A score of \( 0 \) is no correlation.
The Formula
You do not need to memorize this formula, as it is provided in your exam, but you should recognize it:
\( \rho = 1 - \frac{6 \sum d^{2}}{n(n^{2} - 1)} \)
Where:
\( \rho \) = Spearman's rank correlation coefficient
\( d \) = The difference between the ranks
\( n \) = The number of participants
Determining Significance
To see if your result is "real" or just down to luck, you compare your calculated value (the number you got from the formula) to a critical value found in a statistical table. For Spearman’s rank, your calculated value must be equal to or exceed the critical value to be significant.
Psychologists usually look for a significance level of \( p \leq .05 \). This means there is a 5% or lower possibility that the results happened by chance.
Note: For more detail on how to calculate this, see the "Inferential Statistics (List B)" chapter.
6. Evaluating Correlational Research
When you are asked to evaluate correlations in an exam, think about these strengths and weaknesses:
Strengths
- Ethics: It allows us to study things that would be unethical to manipulate experimentally (e.g., the link between smoking and lung capacity).
- Starting Point: Correlations can suggest ideas for future experimental research if a strong relationship is found.
- Reliability: Since correlations often use quantitative data (numbers), they are easy to repeat to see if the same relationship is found again.
Weaknesses
- No Causation: As mentioned, we cannot say for sure that one variable causes the other.
- Internal Validity: Because we don't control extraneous variables, we can't be certain that something else isn't influencing the results.
- Non-linear relationships: Correlations only look for straight-line relationships. If a relationship is curved (e.g., stress helps performance up to a point, then hinders it), a correlation might incorrectly show a zero relationship.
Summary Checklist:
1. Can you define "co-variables"?
2. Can you draw a positive vs. negative scatter diagram?
3. Do you understand why correlation isn't the same as cause-and-effect?
4. Do you know that for Spearman's Rank, the calculated value must be greater than or equal to the critical value?