correlations Flashcards
what are correlations
mathematical technique used to investigate an association between 2 variables (co variables)
how do correlations differ to experiments (2)
-variables are measured, not manipulated
-associations are found, not cause and effect relationships
what are co-variables
variables investigated within a correlation eg. height and weight
what is a positive correlation (3)
-as one variable increases, so does the other
-positive gradient
-correlation coefficient is more than 0
what is a negative correlation (3)
-as one variable increases, the other decreases
-negative gradient
-correlation coefficient is less than 0
what is a zero correlation (3)
-no relationship
-no line of best fit
-correlation coefficient is 0
what is a curvilinear relationship (3)
-as one variable increases so does the other, up until a certain point- after that it decreases as the other variable continues to increase
-forms an inverted ‘U’ shape on graph
evaluation:
strength: by assessing the strength+direction of a relationship they provide a precise and quantifiable measure of how 2 variables are related (2)
-ideas for possible future research if variables are strongly related/show interesting pattern
-correlations used as starting points to assess patterns between variables before doing an actual experiment
evaluation:
strength: correlations are quick and economical to do+secondary data can be used (2)
-no need for controlled environment and no manipulation of variables required
-secondary data can be used so its less time consuming than experiments
evaluation:
criticism: correlations can’t manipulate and control variables like experiments (2)
-studies can only tell us how variables are related, not why
-cant establish cause and effect relationships so we don’t know which co-variable is causing the other to change
evaluation:
criticism: third variable problem
-third variable causes the relationship between the 2 co-variables- researcher is unaware of third variable
evaluation:
criticism: correlations can be misused/misinterpreted (2)
-can be misused/misinterpreted especially by the media
-sometimes relationships between co-variables are presented as casual factors, when in reality they may not be
