HL only This tool is for Higher Level students.
Using the relationship between two variables to make a forecast — and knowing when the forecast deserves to be trusted.
Guiding question
If sales have risen with advertising in the past, how confidently can a business predict sales at a new level of spending?
Success criteria
- I can construct and interpret a scatter diagram and a line of best fit
- I can describe the strength and direction of correlation
- I can use a line of best fit to make predictions and evaluate the reliability of interpolation and extrapolation
HL only
This tool is HL only. The guide lists it as including scatter diagrams, line of best fit and correlation/extrapolation, suggested for units 1, 4 and 5 — for example, sales forecasting (4.3) and data analytics (5.9).
1. What it is and when to use it
Definition Simple linear regression models the relationship between one independent variable (x, such as advertising spending) and one dependent variable (y, such as sales) with a straight line, y = a + bx, where a is the intercept and b is the gradient (the change in y for each one-unit change in x). A scatter diagram plots pairs of data values as points. A line of best fit is the straight line that most closely follows the pattern of the points.
Correlation describes the strength and direction of the relationship. The correlation coefficient r ranges from −1 (perfect negative) through 0 (no linear relationship) to +1 (perfect positive). Interpolation predicts within the range of the data; extrapolation predicts beyond it and is less reliable.