📖 Short lecture
A scatter plot shows paired data $(x,y)$ as points. If $y$ tends to increase as $x$ increases there is a positive association; if $y$ tends to decrease there is a negative association; otherwise there is no association. Points far from the pattern are outliers.
When the association is linear, a line of best fit follows the trend with about as many points above it as below. Its equation $y=mx+b$ is used to make predictions, and its slope is the average change in $y$ per unit of $x$. Two-way tables organise categorical data in rows and columns, so we can compare relative frequencies.
⚠️ Common mistakes
- Association is not causation: two variables can move together because of a third factor.
- Do not use a line of fit far outside the data range (extrapolation) — predictions can become impossible.
- For a relative frequency, divide by the correct total: the row total, the column total, or the grand total.
🔑 Key formulas
✏️ Worked example
The line of fit for hours studied $x$ and test score $y$ is $y=6x+52$. Predict the score for $5$ hours and interpret the slope.
- $\displaystyle y=6(5)+52=82$
- $\displaystyle m=6:\;\text{about }+6\text{ points per extra hour}$
✔ About $82$ points; each extra hour adds about $6$ points.
🧮 More worked examples
Each step says what we do and why.
Example 2A line of fit for TV hours per day $x$ and test score $y$ is $y=-3x+90$. Predict the score for $4$ hours and interpret the slope and intercept.
- Substitute $x=4$.$$y=-3(4)+90=78$$
- Slope $-3$: each extra hour goes with about $3$ fewer points.$$m=-3$$
- Intercept: predicted score with no TV.$$b=90$$
Example 3Of $60$ students, $40$ play a sport, and $28$ of them also like music. Of the $20$ who do not play, $8$ like music. Is there an association?
- Relative frequency in the “sport” row.$$\frac{28}{40}=0.70$$
- Relative frequency in the “no sport” row.$$\frac{8}{20}=0.40$$
- Compare the row percentages.$$70\%\neq40\%$$
Example 4A line of fit passes through $(2,50)$ and $(10,90)$. Write its equation and predict $y$ at $x=6$.
- Slope.$$m=\frac{90-50}{10-2}=5$$
- Intercept from $(2,50)$.$$50=5(2)+b\;\Rightarrow\;b=40$$
- Predict.$$y=5(6)+40=70$$
🖐️ Interactive demo & illustration
Move the sliders or press the buttons: the figure and the numbers update immediately. Try to predict the result before you look.
🎬 Fit a line by eye, then check
Move the line with the sliders. The vertical segments are the residuals (actual − predicted); a good line keeps them small, so the sum of squared residuals is small. Press Best fit to slide to the least-squares line, or make new data with a different association.
🌍 Modelling problems
Real-life situations: translate the words into mathematics, solve, then interpret the answer.
🍦 Ice cream and temperature
A shop's daily cone sales fit $y=12x-200$ where $x$ is the temperature in °C (data from $20$°C to $40$°C). Predict the sales at $35$°C and at $10$°C.
Show the solution
- At $35$°C (inside the data range):$$12(35)-200=220$$
- At $10$°C (far outside the range):$$12(10)-200=-80$$
- Negative sales are impossible: extrapolation fails.
✔ About $220$ cones at $35$°C; the model cannot be used at $10$°C.
🚙 Car value
A car is worth $80\,000$ AED new and about $50\,000$ AED after $5$ years. Using a line of fit, estimate its value after $8$ years.
Show the solution
- Slope: change in value per year.$$m=\frac{50\,000-80\,000}{5-0}=-6000$$
- Equation and prediction.$$y=-6000x+80\,000,\qquad y(8)=32\,000$$
✔ About $32\,000$ AED.
☕ Tea or coffee?
A survey of $200$ people: of $120$ adults, $72$ prefer tea; of $80$ teenagers, $24$ prefer tea. Is drink preference associated with age group?
Show the solution
- Adults:$$\frac{72}{120}=0.6$$
- Teenagers:$$\frac{24}{80}=0.3$$
✔ Yes: $60\%$ of adults but only $30\%$ of teenagers prefer tea.
✅ Multiple-choice quiz
📝 Short essay
Does a strong association between two variables prove that one causes the other? Give an example.
No. Ice-cream sales and swimming accidents both rise in summer, so they show a positive association, but ice cream does not cause accidents: hot weather affects both. Association shows a pattern, not a cause.