Research Methods - Correlations Flashcards
Define correlation.
A mathematical technique in which as researcher investigates an association between two variables called co-variables.
What are co-variables?
The variables investigated within a correlation, for example height and weight. They aren’t referred to as the independant and dependant variables because a correlation investigates the association between the variables, rather than trying to show a cause-and-effect relationship.
What is a positive correlation?
As one co-variable increases, the other increases too. e.g. no. of people in a room and noise level tend to positively correlate!
What is a negative correlation?
A one co-variable increases, the other decreases. e.g. the no.of people in a room and the amount of personal space.
What does zero correlation mean?
When there is no relationship between the co-variables. e.g. the association between the no. of people in a room in Manchester and the total daily rainfall in Peru is likely to be 0.
What is the difference between correlations and experiments?
Experiments - researcher controls/manipulates IV to measure effect on DV = any change infers IV causes change in DV.
Correlation - no manipulation of one variable and not possible to establish cause-and-effect between one co-variable and another…only associations between variables.
Describe the strengths of correlations.
Useful preliminary tool for research - establish strength and direction of relationship…possible ideas for future research.
Quick and economical to carry out…no need for high control or manipulation of variables and can use secondary data!
Describe the limitations of correlations.
Only tell us how variables are related but not why…don’t know which variable is causing the other to change.
There may be a third intervening variable that is causing the relationship observed between the co-variables e.g. people who experience high consistent anxiety drink more caffeine…but so do people who have high-pressure jobs. Therefore, the key unaccounted for variable here is job type which is causing the relationship between the other two variables.
Correlations can also be easily misinterpreted and misused especially by the media especially when third intervening variables are not considered and addressed. e.g. people from broken homes are more likely to become criminals…however this relationship may be influenced by poverty.
