UNIT 4: Scatter diagrams and correlation Flashcards
Key points for drawing a line of best fit
- Find the mean point
- Plot the mean point in a different colour with a +
- Line of best fit must go through the mean point
- Aim for an equal number of points either side of the line
- Doesn’t always need to go through the origin
Positive Correlation
When an increase in one variable means an increase in the second variable
Negative Correlation
When an increase in one variable means a decrease in the second variable
No Correlation
When the two variables are not related
Interpolation
Making a prediction INSIDE the data range
Extrapolation
Making a predication OUTSIDE the data range
Spearman’s Rank Correlation Coefficient
Numerical value between -1 and 1 which tells you how much in AGREEMENT two sets of RANKED data are
Pearson’s correlation coefficient (product moment correlation coefficient)
Numerical value between -1 and 1 which tells you whether data has a LINEAR correlation
Correlation and causation
Correlation does not imply causation.
Causation is when there is a direct link between the two variables
Explanatory (Independent) Variable
x- axis.
Often it’s a variable you can control - or goes up in equal amounts.
Response (Dependent) Variable
y-axis. Usually responds to the change in your explanatory variable