correlations

Question 1

what is a correlation?

Answer 1

● a statistical technique for analysing data

● uses numerical (quantitive data)

● researcher investigates an association between 2 variables (co-variables)

● results shown on scatter graph

Question 2

what is positive correlation?

Answer 2

as one co-variable increases, so does the other

Question 3

what is negative correlation?

Answer 3

as one co-variable increases, the other decreases

Question 4

what is no correlation?

Answer 4

where is there no correlation between the co-variables

Question 5

what is a correlation coefficient?

Answer 5

● a numerical measure of strength and direction of relationship between co-variables

● between +1 and -1

● closer to +1 or -1 the stronger the correlation is

● at 0 there is no relationship at all

Question 6

how do you write a correlation hypothesis?

Answer 6

● must include both variables that the correlation is testing and be fully operationalised

● directional - must state if correlation / relationship is positive or negative

● non-directional - must state correlation / relationship but no direction is stated

● null hypothesis - must state NO correlation / relationship

● NEVER use the word difference, effect or link

Question 7

what are the strengths of correlation?

Answer 7

● usually used as initial research to see if there is relationship. then full-scale experiment can take place. useful starting point for research and often more ethical than an experiment (no manipulation of variables

● relatively quick and economical to carry out - can use secondary data

Question 8

what are the limitations of correlation?

Answer 8

● cannot establish cause and effect - can only tell us how variables are related and not why

● another untested variable could be causing the relationship between the 2 co-variables that we are interested in (the third variable problem)

Question 9

what are the differences between correlations and experiments?

Answer 9

● iv / dv:
experiments - researcher controls / manipulates iv to measure affect on the dv
correlations - correlations have co-variables, there is no manipulation of an iv

● cause and effect:
experiments - can establish how iv effects dv
correlations - as there is no manipulation of variables, we do not know why relationship exists

● control over extraneous variables:
experiments - high control allows cause and effect to be established (in some cases)
correlations - does not consider extraneous variables (third variable problem)