When we draw scatter diagrams, we can say if two variables are connected. To be able to use this to make predictions, we must draw a line of best fit.

**What is a Line of Best Fit.**

A line of best fit is a straight line that we draw on the scatter graph that shows the trend (this could also be called the pattern). When we are drawing it, it must follow three rules.

- It must follow the pattern (for example, bottom left to top right for a positive correlation)
- It must go through the middle of the pattern with about the same amount of points above the line as below.
- It does not have to go through the zero.

What is a line of best fit for?

We already know that scatter graphs show the connection between two variables, a line of best fit allows us to make predictions based on this connection.

Example : A scatter graph showing height against weight is shown below.

- Draw a line of best fit.

Following the rules we discussed earlier. The line of best fit goes through the middle of the pattern, has a similar amount of points above and below the line (7 above, 6 below and 1 on the line) and doesn’t go through the origin.

- A man is 125 cm tall, predict how much he weighs.

We know the man is 125cm tall, so we draw a line up from 125 cm to the line of best fit, we then draw across the weight axis. We can predict that he weights about 75-76 kg

- A woman weighs 100 kg. How tall is she?

This is very similar to the previous question but this time we start from the weight axis, go across to the line of best fit and then read down to height. We can say she is about 147 cm tall.

It’s absolutely vital to remember that as you are reading off a graph, you will always get a bit of a margin for your answers.