Correlation Flashcards
What is a positive correlation
As one co-variable goes up so does the other
What is a negative correlation
One co-variable increases as the other co variable decreases
What is meant by no correlation
No link between the two co-variable
What is a research question
A broad question about the concept being investigated
What is a research aim
Having identified the broad research question. The research aim is then identifying a more specific concept within this question which the researcher wants to investigate
What level of data does a correlation collect
Only ordinal or interval
What is meant by operationalised variable
A variable that is made to be measurable
What is a two tailed correlational hypothesis
There will be a significant correlation/relationship between co variables x and y
What is a one tailed correlational hypothesis
There will be a positive correlation between co variables x and y
What is a null hypothesis
there will not be a significant correlation between view variables x and y
What is primary data
Data gathered directly from the participants researcher
What is secondary data
Data that has been gathered by someone other than the researcher
What are the advantages of correlation studies
Although correlations may not involve the collection of new data, they can still tell us something new: thus, a researcher carrying out a
correlation study will have a hunch that two things might be related (it is this hunch that is expressed in their hypothesis) and, if it turns out that they are related, then something new has been learned about this pre-existing data. Furthermore, it can tell us not just whether two co-variables are related but also the direction of their relationship and the strength of the relationship, too.
2. As there is no manipulation in a correlation study - it involves no more than the statistical analysis of two measured co-variables - it is a useful technique when either practical or ethical reasons mean that variables cannot be manipulated. (For example, if a researcher wanted to find out whether there was a relationship between accidents and alcohol consumption, it might be feasible to see if the number of people attending A&E goes up at times of the year or even of the week when alcohol consumption is at its greatest; however, an experiment in which participants were either left sober or made drunk before being placed in situations where they might have accidents would be unlikely to meet with BPS approval!).
3. Correlation studies can also act as a good starting point for research because, once relationships have been established between co-variables, more research (e.g. experiments) can then be conducted to investigate them further. (Equally, if no significant relationship is found between two co-variables then a causal relationship between the two co-variables can be ruled out).
What are the disadvantages of correlation studies
It could be that watching violent television was causing people to be aggressive
• It could also be that being aggressive was leading people to watch violent
TV
• It could even be that there was some hidden third factor (e.g. the type of childhood a person had had, a genetic abnormality, or lead poisoning) that caused people to both (a) watch violent TV and (b) engage in aggressive behaviour.
