What is a line of best fit?
A scatter plot shows pairs of numbers as dots. When the dots go up (or down) in a rough straight band, we say there is a linear association.
A line of best fit (also called a trend line) is one straight line that shows this trend. It does not have to pass through any dot. It just goes through the middle of the cloud.
- Positive trend: line goes up to the right.
- Negative trend: line goes down to the right.
- No trend: dots are spread everywhere, so no useful line.
How to draw a line of best fit by eye
- Look at the shape of the dots. Is it a straight band?
- Place a clear ruler along the band.
- Turn and slide it until about half the dots are above and half below.
- Check the gaps: they should be small, and not all big gaps on one side.
- Ignore an outlier (a dot far away from the rest) when you judge the line.
The gap from a dot to the line, measured straight up or down, shows how well the line fits. Small gaps = good fit.
The equation: slope and intercept meaning
Pick two points on your line (not data dots), for example (2, 52) and (7, 82).
Slope m = rise Ãˇ run = (82 â 52) Ãˇ (7 â 2) = 30 Ãˇ 5 = 6.
Intercept c: put one point in y = mx + c â 52 = 6 Ã 2 + c â c = 40.
So score â 6 Ã hours + 40.
- Slope 6: each extra hour of study goes with about 6 more marks.
- Intercept 40: a student who studies 0 hours is predicted to score about 40.
Always say the meaning with units: "marks per hour", "rupees per kg", "cm per year".
Making predictions (and their limits)
To predict, put the x value into the equation, or read up to the line and across.
- Inside the data range (interpolation): fairly safe.
- Far outside the range (extrapolation): risky. 30 hours of study does not give 220 marks!
A trend line shows things change together. It does not prove that one causes the other.
Key formulas and definitions
- y = mx + c (line of best fit)
- slope m = (yâ â yâ) Ãˇ (xâ â xâ), using two points on the line
- intercept c = y â m x (value of y when x = 0)
- prediction: Ŏ = m à (given x) + c
Worked examples
1. A scatter plot of age (years) and height (cm) of children goes up to the right. What kind of association is it?
Positive linear association: as age goes up, height tends to go up.
2. A trend line passes through (0, 20) and (5, 45). Find its equation.
m = (45 â 20) Ãˇ (5 â 0) = 25 Ãˇ 5 = 5. c = 20 (value at x = 0). Equation: y = 5x + 20.
3. For score = 6h + 40, predict the score for 3.5 hours.
score = 6 Ã 3.5 + 40 = 21 + 40 = 61 marks.
4. A line for phone battery is: charge % = â8t + 100, where t is hours of use. Explain slope and intercept.
Slope â8: battery drops about 8 % each hour. Intercept 100: at t = 0 the phone is fully charged (100 %).
5. Two students draw lines. Line A has 9 dots above and 1 below. Line B has 5 above and 5 below with small gaps. Which is better?
Line B. A good line of best fit splits the dots roughly evenly and keeps gaps small; line A is too low.
6. Using taxi fare = 15d + 50 (rupees, d in km), how far can you go for âš230?
230 = 15d + 50 â 15d = 180 â d = 12 km.
Common mistakes
- Joining the first and last dot instead of going through the middle of all the dots.
- Forcing the line through the origin (0, 0) when the data does not start there.
- Using data dots that are off the line to find the slope; use two points that are on your line.
- Predicting far beyond the data, or saying the line proves one thing causes the other.