Correlations

Question 1

What is a correlation?

Give an example

Answer 1

A method of demonstrating a relationship between two variables

Two variables are measured and analysed to see if there is a relationship

E.g. people with higher scores on an IQ test will have a larger head size.

Question 2

Characteristics of correlations

Answer 2

A statistical technique

One participants provides data for both variables (repeated measures)

No IV and DV (no cause and effect) – variables that exist alongside each other

Allows us to see if two variables are related and how strongly

Question 3

When would we use a correlation?

Give examples

Answer 3

To test a hypothesis between the relationship of two variables e.g. the more cake people eat the happier they will be.

When an experiment would be unethical – we can test a naturally occurring relationship

E.g. Alcohol consumption and violence, cocaine use and depression

Question 4

How can you find a relationship?

Answer 4

Take a group of participants and measure and record the variables.

The variables are CO-VARIABLES, not an IV and DV (experimental method).

Scatter graphs

Question 5

What is a positive correlation?

Give an example

Answer 5

Both variables increase or decrease together e.g…

The taller a person is the heavier they are likely to be

The more revision you do the higher the exam grade will be

The more sexual partners you have the higher the chances of catching an STD

Question 6

What is a negative correlation?

Give an example

Answer 6

As one variable increases the other will decrease e.g…

The more you practice an instrument the less mistakes you will make

The younger you are the more text messages you are likely to send

The more alcohol you drink, the less you can remember

Question 7

What does no correlation look like?

Answer 7

Question 8

Why else are correlations useful?

Answer 8

Testing reliability – scores from one test can be correlated against the same test taken later

Comparing relatives (specifically twins) to see if behaviours/ mental illness are innate or not.

Question 9

What can’t correlations tell us?

Answer 9

WE CANNOT ESTABLISH CAUSE AND EFFECT

Just because two variables increase together does not mean an increase in one causes an increase in the other. So, although a correlation could suggest smoking causes cancer, we cannot establish this. Other variables may be responsible.

Question 10

Strengths of correlations

Answer 10

Useful starting point for research.

By assessing the strength and direction of a relationship, correlations provide a measure of how two variables are related. If variables are strongly related it may suggest hypothesis for future research.

Relatively economical

Unlike a lab study, there is no need for a controlled environment and can use secondary data (e.g., government data). So correlations are less time-consuming than experiments.

Question 11

Limitations of correlations

Answer 11

No cause and effect

Correlations are often presented as casual, e.g., by the media, when they only show how two variables are related. This leads to false conclusions about causes of behaviour.

Intervening variables

Another untested variable may explain relationship between co-variables. This may also lead to false conclusion.