Correlations

Question 1

Define correlation

Answer 1

This is a systematic assosiation between two continuous variables, if they are not linked at all there is a zero correlation, they might both increase together as a positive correlation, or as one variable increases the other decreases which is a negative correlation

Question 2

What do we do when we conduct a study using a correlational analysis?

Answer 2

We need to produce a correlational hypothesis which states the expected association between the 2 variables

Question 3

How can we illustrate a correlation?

Answer 3

Using a scattergram, for each individual we obtain two scores which are used to plot one dot for that individual- the co variables determine the x and y position of the dot. The scatter of the dots indicated the degree of correlation between the co variables

Question 4

How should we use a scattergram to measure the strength of a systematic assosiation?

Answer 4

A statistical test to calculate the correlation co efficient, a measure of the extent of correlation that exists between the co variables. This is a number between -1 and +1, - means strongest negative correlation and + means a strong positive correlation
To find if the correlation co efficient is significant is to use a table of significance to see how big the co efficient needs to be in order for the correlation to count as significant

Question 4

Describe the strengths of correlations

Answer 4

• They are used to investigate trends in data, if the correlation is significant further experiment is warranted
• Procedures in a correlation can usually be easily repeated meaning findings can be confirmed

Question 5

Describe the limitations of correlations

Answer 5

• Variables are simply measured, no deliberate change is made therefore no conclusion can be made about one variable causing the other.
• People may assume casual conclusions, and this is an issue because it means such misinterpretion of correlation might mean people will design programmes for improvement based from false premises
• Furthermore, the supposed casual connection may be due to intervening variables which are unknown that can explain why the two variables are linked which may be more important than the identified variable
• Correlations may lack internal/ external validity