correlations Flashcards
twins and Brendgen et al.
what is a positive correlation?
when both variables move in the same direction.
what is a negative correlation?
an increase in one variable is associated with a decrease in another.
what is no correlation?
no relationship in one variable is associated with a decrease in the other.
what are correlations?
They measure the relationship between two variables, are also known as co-variables (have 2 sets of data)
what is a correlation co efficient?
Measures the strength of a corelation.
-1.0=perfect negative
+1.0=perfect positive
0=no relationship.
what are the 2 reasons why correlations don’t have sample groups?
-researchers collect data on two variables from a single sample of participants.
-it focuses on variable relationships, not group differences.
will the correlation hypothesis be about a difference or a relationship?
it will not be about a difference between the two conditions; it will be about a relationship between the two variables.
why would the correlation hypothesis be directional?
because it could predict a positive or negative correlation.
do correlations have independent or dependent variables?
neither, there are two variables of an equal importance.
what sort of data do correlations use?
interval/ratio data and ordinal data.
what controls do correlations involve?
what do you control when looking for a relationship between two variables?
-keeping all variables the same except the ones being studied.
-you might want to control something which is related to the variables.
what are 2 strength for correlation?
-correlation designs often have high levels of reliability.
-they often involve an easily replicable method (self-report questionnaires provide standardised procedures)
-this means that it is easier to replicate correlational research multiple times to find consistent results
-both sets of data are provided by the same person.
-as they are looking for a link between 2 or more variables, the same person will complete all questionnaires.
-individual differences will therefore not affect the result
what are 2 weaknesses for correlation?
-the measures might not produce valid data.
-time in therapy is a clear measure but the benefits of therapy are not easy to quantify.
-this is a weakness as it cant be accurately looking at what it aims to look at (lacks internal validity)
-the methods used to gather data are often not valid. the research relies heavily on methods like self-report methods.
-this may lead to social desirability bias if honest answers are not disclosed, then the researcher might not be measuring what they aim to measure
