14. correlations Flashcards
correlation
a research method which looks for a relationship between 2 measured co-variables
a change in 1 variable is related to another
eg. height is positively correlated to shoe size. the taller someone is, the larger their shoe size tends to be
correlation pt2
no manipulation of data, conditions or groups
no IV or DV, just 2 co-variables
when can correlations be used?
used when it is unethical/impossible to manipulate variables
eg. if you want to study how much
Nutella people eat in a day
to conduct a correlation, both co-variables must:
- exist over a range
- be able to be measured numerically
positive correlation
2 co-variables increase together, as one increases, so does the other
eg. number of aggressive models in a household & violent behavour
negative correlation
the co-variables are opposite, as one increases, the other decreases
eg. hours spent watching TV and relationship satisfaction
no correlation
the co-variables are not related, there is no clear trend
eg. how much you eat for lunch and the pope’s happiness
correlation co-efficient
a number between -1 and +1 that indicates the direction and strength of a correlation
+1.0 perfect positive
+0.8 strong
+0.5 moderate
+0.3 weak
0 no correlation
-0.3 weak
-0.5 moderate
-0.8 strong
-1.0 perfect negative
directionality
is X causing Y
is Y causing X
pros and cons of correlation
PROS:
- can understand the relationship between 2 variables (pos/neg, weak/strong)
- makes a good pilot study to generate a hypothesis for new areas of research
- can research variables that would be impossible or unethical to manipulate
CONS:
- tells us nothing about other variables that may be the real cause
- often correlations can be misleading (non-linear relationship)
- they have the same weakness as whatever method was used to gather the data (observation/self report)
- correlations DOES NOT EQUAL causation