Lesson 4 Correlation Flashcards
What is correlation
A correlation is a technique for analysing the strength of the relationship between two quantitative variables, known as co-variables.
A correlational analysis will show one of three things:
1) A positive correlation - this means that as one variable increases the other variable increases as well, e.g. the number of ice cream sold increases as the temperature increases
2) A negative correlation - this means that as one variable increases the other variable decreases, for instance the more time a student spends playing video games, the less grade A’s they will achieve in the exams.
3) No correlation – there is no relationship between the two variables. For example there is no relationship between head size and IQ levels, so if you have a large head, it does not mean that you are more intelligent!
What is correlation coefficient
• The strength of a correlation can be between -1 and +1.
• A strong positive correlation coefficient will be a number closer to +1
• 0 means that there is no correlation. If a correlation coefficient is a number which is closer to 0 then this is normally shown as no correlation.
• -1 means that there is a strong negative correlation.
• The strength of the correlation is known as the correlation coefficient.
What are correlations plotted on
Correlations are plotted on a scattergram or a scattergraph.
One co-variable forms the X-axis and the other the Y-axis.
Each point or dot on the graph is the x and y position of each co-variable
Advantages of correlation
1) Correlations allows psychologists to establish the strength of the relationship between two variables and measure it precisely for example there might be a very strong positive relationship +1 between the variables of being very wealthy and being very happy
2) Correlations allows researchers to investigate things that could not be manipulated experimentally for ethical or practical reasons, for instance a natural relationship might occur between the variables of being highly stressed and having cardiovascular disease. It would be unethical to intentionally inflict stress or cardiovascular disease upon participants, but if it occurred naturally then it would be fine to study this issue
3) Once a correlation has been conducted predictions can be made about one of the variables based on what is known about the other variable. For instance if we know that high stress causes high cholesterol, then we might state that as stress increases, cholesterol might increase. This can help inform conclude that lowering stress might lower cholesterol.
Disadvantages of correlation
1) Correlational analysis cannot demonstrate cause and effect; we cannot tell which variable influences the other. For instance can we sure that high stress causes an increase in cholesterol, or does having high cholesterol cause an increase in stress?
2) Even if there is a correlation between two variables it may be the case that the variables are not actually related but that there is a third unknown variable which influences both. For instance maybe an unhealthy diet might cause an increase in cholesterol (rather than stress causing the increase)?
3) Correlations can only measure linear relationships and does not detect curvilinear relationships. This is when there is a positive relationship up to a certain point but after that the relationship becomes negative or (vice versa).
