Statistics
We study the normal distribution, z-scores, and apply regression and correlation to real data in final-preparation contexts.
9.1 Normal Distribution & Regression
- Understand the shape and properties of the normal distribution
- Calculate and interpret z-scores
- Use the 68-95-99.7 rule for normal distributions
- Apply regression and correlation for data analysis and prediction
Real-World Connection
Test scores, heights, and manufacturing tolerances all follow an approximate normal distribution. Quality control engineers use the 68-95-99.7 rule: 95% of products within 2σ of the mean are acceptable; those outside are defects.
Property / Rule
Normal Distribution Properties
Bell-shaped, symmetric about the mean . Mean = Median = Mode. Area under the curve = 1 (total probability). Characterised by and .
z-score
$x$ = data value; $\mu$ = mean; $\sigma$ = standard deviation; $z$ = number of standard deviations from mean
Property / Rule
68-95-99.7 Rule (Empirical Rule)
For a normal distribution: approximately 68% of data within 1σ of mean; 95% within 2σ; 99.7% within 3σ.
ℹ️ Note
A z-score of means the value is 2 standard deviations above the mean. A negative z-score means the value is below the mean. Comparing z-scores from different distributions allows fair comparison.
Worked Example
z-score and normal distribution
Problem
CAPS Cognitive Level Distribution