Bivariate Flashcards
(6 cards)
What coefficeint of determination show us
“variation in the response variable that can be explained by the variation in the explanatory variable”
Correlation between 2 variables
A correlation tells you about the strength of an association between two variables, however it does not give any indication as to the source or cause of the association.
For Example:
strong positve correlation between ppl fainitng and putting on suncsrren but this doesnt mean sunscreen cause ppl to fain, 3rd common variable is obvs the sun and temperature rising
There are three main reasons an association would exist but would not be considered a cause.
Common response- associated due to a common third variable
Confounding variable- two possible explanatory variables affecting the one response variables, no way of knowing which is the actual cause of the association
Coincidence- association just happens by chance with no plausible explanation
Does association equal causation
By itself, an observed association between two variables is never enough to conclude that the two variables are causally related.
Association ≠ Causation
Extrapolition
process of predicting data values outside the range of data is a process called extrapolation, and we should not treat these as valid predictions.
intrapolition
, predicting data values within the range of data is a process called interpolation, and we can treat these as valid predictions and we would expect a reasonably reliable result.