Study guide
15.4 Normal Distribution, Statistical Sampling, Correlation, and Regression
Gaokao Mathematics Exam
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Sets, Logic, and Algebraic Foundations
1.1 Sets and Set Operations 1.2 Propositional Logic and Quantifiers 1.3 Inequalities and Absolute Value Equations
2 Chapter 2: Functions and Fundamental Properties
2.1 Concept, Domain, Range, and Mapping of Functions 2.2 Monotonicity, Parity, and Periodicity of Functions 2.3 Function Graphs, Transformations, and Zeros
3 Chapter 3: Exponential, Logarithmic, and Power Functions
3.1 Exponents and Exponential Functions 3.2 Logarithms and Logarithmic Functions 3.3 Power Functions and Comparative Growth Rates
4 Chapter 4: Complex Numbers and Quadratic Function Views
4.1 Algebraic and Geometric Representations of Complex Numbers 4.2 Complex Number Operations and Equations 4.3 Quadratic Equations and Inequalities from a Function Viewpoint
5 Chapter 5: Trigonometric Functions and Identities
5.1 Angles, Radians, and Unit Circle Definitions 5.2 Trigonometric Identities and Reduction Formulas 5.3 Addition, Subtraction, and Double-Angle Formulas 5.4 Graphs and Properties of Sine, Cosine, and Tangent Functions
6 Chapter 6: Triangles and Trigonometric Applications
6.1 Sine Theorem, Cosine Theorem, and Triangle Area Formulas 6.2 Solving Triangles and Comprehensive Geometric Applications 6.3 Practical Measurement, Optimization, and Trigonometric Modeling
7 Chapter 7: Sequences and Series
7.1 Arithmetic Sequences and Sums 7.2 Geometric Sequences and Sums 7.3 Recurrence Relations and General Term Determination 7.4 Advanced Summation Techniques and Sequence Applications
8 Chapter 8: Plane Vectors
8.1 Vector Concepts, Addition, Subtraction, and Scalar Multiplication 8.2 Coordinate Operations of Vectors 8.3 Dot Products, Angles, Vector Projections, and Geometric Applications
9 Chapter 9: Lines and Circles in Cartesian Coordinates
9.1 Equations of Straight Lines and Positional Relationships 9.2 Standard and General Equations of Circles 9.3 Positional Relationships Between Lines and Circles
10 Chapter 10: Conic Sections: Ellipses
10.1 Standard Equations and Definitions of Ellipses 10.2 Geometric Properties of Ellipses (Vertices, Foci, Eccentricity) 10.3 Intersection of Lines and Ellipses and Chord Problems
11 Chapter 11: Conic Sections: Hyperbolas and Parabolas
11.1 Hyperbola Definitions, Standard Equations, and Asymptotes 11.2 Parabola Definitions, Standard Equations, and Focal Properties 11.3 Unified Properties of Conic Sections and Trajectory Problems
12 Chapter 12: Solid Geometry: Three-Dimensional Figures
12.1 Prisms, Pyramids, Cylinders, Cones, and Spheres 12.2 Three-Dimensional Views and Cross-Sections 12.3 Surface Area and Volume Calculations
13 Chapter 13: Spatial Lines, Planes, and Spatial Vectors
13.1 Positional Relationships of Lines and Planes in Space 13.2 Parallelism and Perpendicularity Proofs in Solid Geometry 13.3 Spatial Coordinates and Spatial Vector Operations 13.4 Using Spatial Vectors to Compute Angles and Distances
14 Chapter 14: Permutations, Combinations, and Binomial Theorem
14.1 Fundamental Counting Principles (Addition and Multiplication Rules) 14.2 Permutations and Combinations 14.3 Binomial Theorem, Expansion Terms, and Coefficients
15 Chapter 15: Probability and Statistics
15.1 Classical Probability and Conditional Probability 15.2 Independent Events and Repeated Bernoulli Trials 15.3 Discrete Random Variables, Probability Distributions, Expectation, and Variance 15.4 Normal Distribution, Statistical Sampling, Correlation, and Regression
16 Chapter 16: Derivatives and Mathematical Modeling
16.1 Limit Concepts and Derivatives of Elementary Functions 16.2 Tangent Lines, Derivatives, Monotonicity, and Extremum Analysis 16.3 Optimization Problems and Curve Sketching with Derivatives 16.4 Real-World Mathematical Modeling and Problem Solving
15.4 Normal Distribution, Statistical Sampling, Correlation, and Regression
Key Takeaways Explain This Topic AI
Organize raw samples with frequency tables, histograms, or stem-and-leaf plots before estimating a population.
Compute sample mean, variance/standard deviation, median, and quartiles from ordered data; the IQR summarizes spread resistant to extremes.
Simple random, stratified, and systematic sampling each control bias differently when estimating population parameters.
The Normal model N ( μ , σ 2 ) N(\mu,\sigma^2) N ( μ , σ 2 ) approximates many measurement distributions; standardize with Z = ( X − μ ) / σ Z=(X-\mu)/\sigma Z = ( X − μ ) / σ for probability lookups.
Scatter plots, correlation, and least-squares regression quantify linear association between paired variables.
Last updated: July 2026
- 15.4 Normal Distribution, Statistical Sampling, Correlation, and Regression
- 0. Descriptive Statistics from Raw Sample Data
Before Normal models and regression, Gaokao statistics questions often ask you to organize a raw sample and compute its basic features. Treat this as the mandatory descriptive layer of 统计.
Frequency tables, histograms, and stem-and-leaf plots
Given a sample of size n n n , group values into class intervals and record frequencies f i f_i f i (counts) and relative frequencies f i / n f_i/n f i / n .
Display What it shows Gaokao use
Frequency distribution table Class intervals with counts f i f_i f i Organize raw scores before estimating a population
Frequency histogram Adjacent bars whose heights are frequencies (or densities) Compare shape: unimodal, skewed, roughly Normal
Stem-and-leaf plot Stems (leading digits) with leaves (trailing digits) Preserve original data while showing distribution shape
Worked Example (raw sample). Scores: 72 , 85 , 78 , 90 , 66 , 85 , 73 , 88 , 79 , 84 72, 85, 78, 90, 66, 85, 73, 88, 79, 84 72 , 85 , 78 , 90 , 66 , 85 , 73 , 88 , 79 , 84 .
- Stem-and-leaf (tens | ones): 6 | 6
- 7 | 2 3 8 9
- 8 | 4 5 5 8
- 9 | 0
The sample is unimodal and slightly left-concentrated in the 70s–80s; a histogram with width-10 classes would show the same pattern.
Sample mean, variance, and quartiles
For raw data x 1 , … , x n x_1,\dots,x_n x 1 , … , x n :
Sample mean: x ˉ = 1 n ∑ i = 1 n x i \bar{x} = \dfrac{1}{n}\sum_{i=1}^n x_i x ˉ = n 1 ∑ i = 1 n x i
Sample variance (population-style form often used in high-school contexts): s 2 = 1 n ∑ i = 1 n ( x i − x ˉ ) 2 s^2 = \dfrac{1}{n}\sum_{i=1}^n (x_i-\bar{x})^2 s 2 = n 1 ∑ i = 1 n ( x i − x ˉ ) 2 (some problems use n − 1 n-1 n − 1 ; follow the formula stated in the question)
Sample standard deviation: s = s 2 s = \sqrt{s^2} s = s 2
Ordered sample x ( 1 ) ≤ ⋯ ≤ x ( n ) x_{(1)} \le \cdots \le x_{(n)} x ( 1 ) ≤ ⋯ ≤ x ( n ) ; median is the middle value (or average of two middle values). Quartiles Q 1 , Q 3 Q_1,Q_3 Q 1 , Q 3 split the lower/upper halves; the interquartile range is I Q R = Q 3 − Q 1 IQR = Q_3 - Q_1 I QR = Q 3 − Q 1 . A box plot marks min, Q 1 Q_1 Q 1 , median, Q 3 Q_3 Q 3 , max.
Worked Example (continued). For the ten scores above, ∑ x i = 800 \sum x_i = 800 ∑ x i = 800 , so x ˉ = 80 \bar{x} = 80 x ˉ = 80 . Using the 1 / n 1/n 1/ n variance form: s 2 = 1 10 [ ( 72 − 80 ) 2 + ( 85 − 80 ) 2 + ⋯ + ( 84 − 80 ) 2 ] = 544 10 = 54.4 , s = 54.4 ≈ 7.38. s^2 = \frac{1}{10}\left[(72-80)^2+(85-80)^2+\cdots+(84-80)^2\right]=\frac{544}{10}=54.4,\quad s=\sqrt{54.4}\approx 7.38. s 2 = 10 1 [ ( 72 − 80 ) 2 + ( 85 − 80 ) 2 + ⋯ + ( 84 − 80 ) 2 ] = 10 544 = 54.4 , s = 54.4 ≈ 7.38.
Ordered list: 66 , 72 , 73 , 78 , 79 , 84 , 85 , 85 , 88 , 90 66,72,73,78,79,84,85,85,88,90 66 , 72 , 73 , 78 , 79 , 84 , 85 , 85 , 88 , 90 . Median = 79 + 84 2 = 81.5 = \frac{79+84}{2} = 81.5 = 2 79 + 84 = 81.5 . Lower half median Q 1 = 73 Q_1 = 73 Q 1 = 73 , upper half median Q 3 = 85 Q_3 = 85 Q 3 = 85 , so I Q R = 12 IQR = 12 I QR = 12 .
These descriptive summaries are what sampling later estimates for a population, and they feed the correlation/regression displays that follow.
1. Normal Distribution
The Normal Distribution (or Gaussian distribution) is the most fundamental continuous probability distribution in statistics, describing phenomena affected by numerous small, independent random forces.
1.1 Probability Density Function
A continuous random variable X X X follows a Normal Distribution with mean μ \mu μ and standard deviation σ > 0 \sigma > 0 σ > 0 , denoted X ∼ N ( μ , σ 2 ) X \sim N(\mu, \sigma^2) X ∼ N ( μ , σ 2 ) , if its probability density function (PDF) is: f ( x ) = 1 2 π σ e − ( x − μ ) 2 2 σ 2 , − ∞ < x < ∞ f(x) = \frac{1}{\sqrt{2\pi}\sigma} e^{-\frac{(x-\mu)^2}{2\sigma^2}}, \quad -\infty < x < \infty f ( x ) = 2 π σ 1 e − 2 σ 2 ( x − μ ) 2 , − ∞ < x < ∞
1.2 Key Geometric Properties of the Normal Curve
Symmetry : The curve is bell-shaped and symmetric about the vertical line x = μ x = \mu x = μ .
Maximum Value : The maximum value occurs at x = μ x = \mu x = μ , where f ( μ ) = 1 2 π σ f(\mu) = \frac{1}{\sqrt{2\pi}\sigma} f ( μ ) = 2 π σ 1 . As σ \sigma σ decreases, the peak becomes taller and narrower; as σ \sigma σ increases, the curve flattens.
Asymptotes : The x-axis ( y = 0 y = 0 y = 0 ) is a horizontal asymptote as x → ± ∞ x \to \pm\infty x → ± ∞ .
Inflection Points : The curve has inflection points at x = μ − σ x = \mu - \sigma x = μ − σ and x = μ + σ x = \mu + \sigma x = μ + σ .
Standard Normal Distribution : When μ = 0 \mu = 0 μ = 0 and σ = 1 \sigma = 1 σ = 1 , Z ∼ N ( 0 , 1 ) Z \sim N(0, 1) Z ∼ N ( 0 , 1 ) is the Standard Normal Distribution. Any X ∼ N ( μ , σ 2 ) X \sim N(\mu, \sigma^2) X ∼ N ( μ , σ 2 ) can be standardized via Z = X − μ σ Z = \frac{X - \mu}{\sigma} Z = σ X − μ .
1.3 The 3 σ 3\sigma 3 σ Empirical Rule
For X ∼ N ( μ , σ 2 ) X \sim N(\mu, \sigma^2) X ∼ N ( μ , σ 2 ) , the area under the PDF curve yields standard probabilities:
P ( μ − σ ≤ X ≤ μ + σ ) ≈ 0.6827 P(\mu - \sigma \le X \le \mu + \sigma) \approx 0.6827 P ( μ − σ ≤ X ≤ μ + σ ) ≈ 0.6827
P ( μ − 2 σ ≤ X ≤ μ + 2 σ ) ≈ 0.9545 P(\mu - 2\sigma \le X \le \mu + 2\sigma) \approx 0.9545 P ( μ − 2 σ ≤ X ≤ μ + 2 σ ) ≈ 0.9545
P ( μ − 3 σ ≤ X ≤ μ + 3 σ ) ≈ 0.9973 P(\mu - 3\sigma \le X \le \mu + 3\sigma) \approx 0.9973 P ( μ − 3 σ ≤ X ≤ μ + 3 σ ) ≈ 0.9973
Due to symmetry about x = μ x = \mu x = μ : P ( X ≤ μ ) = P ( X ≥ μ ) = 0.5 P(X \le \mu) = P(X \ge \mu) = 0.5 P ( X ≤ μ ) = P ( X ≥ μ ) = 0.5 P ( X ≤ μ − a ) = P ( X ≥ μ + a ) = 1 − P ( X ≤ μ + a ) P(X \le \mu - a) = P(X \ge \mu + a) = 1 - P(X \le \mu + a) P ( X ≤ μ − a ) = P ( X ≥ μ + a ) = 1 − P ( X ≤ μ + a )
2. Statistical Sampling Methods
Statistical inference relies on selecting representative samples from a target population of size N N N .
2.1 Simple Random Sampling
Every combination of n n n units has an equal probability of selection.
Lottery Method : Draw numbered tokens from an urn (suitable for small N N N ).
Random Number Table : Select units matching pseudo-random digits.
2.2 Stratified Random Sampling
When the population consists of distinct sub-groups (strata) with high between-stratum variability and low within-stratum variability:
Divide population N N N into k k k mutually exclusive strata of sizes N 1 , N 2 , … , N k N_1, N_2, \dots, N_k N 1 , N 2 , … , N k ( ∑ N i = N \sum N_i = N ∑ N i = N ).
Sample n i n_i n i items from stratum i i i proportionally to its size: n i N i = n N ⟹ n i = n × N i N \frac{n_i}{N_i} = \frac{n}{N} \implies n_i = n \times \frac{N_i}{N} N i n i = N n ⟹ n i = n × N N i
Sample Mean Formula : x ˉ = ∑ i = 1 k N i N x ˉ i \bar{x} = \sum_{i=1}^k \frac{N_i}{N} \bar{x}_i x ˉ = i = 1 ∑ k N N i x ˉ i where x ˉ i \bar{x}_i x ˉ i is the sample mean of stratum i i i .
- 3. Bivariate Correlation and Linear Regression Analysis
- 3.1 Scatter Plots and Pearson Correlation Coefficient r r r
For paired sample data ( x 1 , y 1 ) , ( x 2 , y 2 ) , … , ( x n , y n ) (x_1, y_1), (x_2, y_2), \dots, (x_n, y_n) ( x 1 , y 1 ) , ( x 2 , y 2 ) , … , ( x n , y n ) , the sample correlation coefficient r r r measures the strength and direction of linear association: r = ∑ i = 1 n ( x i − x ˉ ) ( y i − y ˉ ) ∑ i = 1 n ( x i − x ˉ ) 2 ∑ i = 1 n ( y i − y ˉ ) 2 = ∑ i = 1 n x i y i − n x ˉ y ˉ ( ∑ x i 2 − n x ˉ 2 ) ( ∑ y i 2 − n y ˉ 2 ) r = \frac{\sum_{i=1}^n (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum_{i=1}^n (x_i - \bar{x})^2} \sqrt{\sum_{i=1}^n (y_i - \bar{y})^2}} = \frac{\sum_{i=1}^n x_i y_i - n \bar{x} \bar{y}}{\sqrt{(\sum x_i^2 - n \bar{x}^2)(\sum y_i^2 - n \bar{y}^2)}} r = ∑ i = 1 n ( x i − x ˉ ) 2 ∑ i = 1 n ( y i − y ˉ ) 2 ∑ i = 1 n ( x i − x ˉ ) ( y i − y ˉ ) = ( ∑ x i 2 − n x ˉ 2 ) ( ∑ y i 2 − n y ˉ 2 ) ∑ i = 1 n x i y i − n x ˉ y ˉ
− 1 ≤ r ≤ 1 -1 \le r \le 1 − 1 ≤ r ≤ 1 .
r > 0 r > 0 r > 0 : Positive linear correlation; r < 0 r < 0 r < 0 : Negative linear correlation.
∣ r ∣ → 1 |r| \to 1 ∣ r ∣ → 1 : Strong linear relationship; ∣ r ∣ → 0 |r| \to 0 ∣ r ∣ → 0 : Weak linear relationship.
3.2 Least Squares Linear Regression
When a strong linear relationship exists, we fit the linear model y ^ = b ^ x + a ^ \hat{y} = \hat{b}x + \hat{a} y ^ = b ^ x + a ^ by minimizing the sum of squared residual errors ∑ i = 1 n ( y i − y ^ i ) 2 \sum_{i=1}^n (y_i - \hat{y}_i)^2 ∑ i = 1 n ( y i − y ^ i ) 2 .
Estimation Formulas: b ^ = ∑ i = 1 n ( x i − x ˉ ) ( y i − y ˉ ) ∑ i = 1 n ( x i − x ˉ ) 2 = ∑ i = 1 n x i y i − n x ˉ y ˉ ∑ i = 1 n x i 2 − n x ˉ 2 \hat{b} = \frac{\sum_{i=1}^n (x_i - \bar{x})(y_i - \bar{y})}{\sum_{i=1}^n (x_i - \bar{x})^2} = \frac{\sum_{i=1}^n x_i y_i - n \bar{x} \bar{y}}{\sum_{i=1}^n x_i^2 - n \bar{x}^2} b ^ = ∑ i = 1 n ( x i − x ˉ ) 2 ∑ i = 1 n ( x i − x ˉ ) ( y i − y ˉ ) = ∑ i = 1 n x i 2 − n x ˉ 2 ∑ i = 1 n x i y i − n x ˉ y ˉ a ^ = y ˉ − b ^ x ˉ \hat{a} = \bar{y} - \hat{b} \bar{x} a ^ = y ˉ − b ^ x ˉ
Critical Property of Linear Regression:
The regression line always passes through the sample centroid ( x ˉ , y ˉ ) (\bar{x}, \bar{y}) ( x ˉ , y ˉ ) , satisfying y ˉ = b ^ x ˉ + a ^ \bar{y} = \hat{b} \bar{x} + \hat{a} y ˉ = b ^ x ˉ + a ^ .
4. 2 × 2 2 \times 2 2 × 2 Contingency Table and χ 2 \chi^2 χ 2 Independence Test
To test whether two categorical variables A A A and B B B are statistically independent, we construct a 2 × 2 2 \times 2 2 × 2 contingency table:
- Variable B B B B ‾ \overline{B} B Total
- A A A a a a b b b a + b a + b a + b
- A ‾ \overline{A} A c c c d d d c + d c + d c + d
- Total a + c a + c a + c b + d b + d b + d n = a + b + c + d n = a + b + c + d n = a + b + c + d
4.1 Test Statistic K 2 K^2 K 2 ( χ 2 \chi^2 χ 2 ) Formula K 2 = n ( a d − b c ) 2 ( a + b ) ( c + d ) ( a + c ) ( b + d ) K^2 = \frac{n(ad - bc)^2}{(a+b)(c+d)(a+c)(b+d)} K 2 = ( a + b ) ( c + d ) ( a + c ) ( b + d ) n ( a d − b c ) 2
4.2 Decision Rule Table (Critical Values)
Compare calculated K 2 K^2 K 2 against standard critical thresholds x α x_\alpha x α :
P ( K 2 ≥ x α ) P(K^2 \ge x_\alpha) P ( K 2 ≥ x α ) 0.10 0.10 0.10 0.05 0.05 0.05 0.010 0.010 0.010 0.001 0.001 0.001
- Critical Value x α x_\alpha x α 2.706 2.706 2.706 3.841 3.841 3.841 6.635 6.635 6.635 10.828 10.828 10.828
- Confidence Level 90 % 90\% 90% 95 % 95\% 95% 99 % 99\% 99% 99.9 % 99.9\% 99.9%
If K 2 ≥ 3.841 K^2 \ge 3.841 K 2 ≥ 3.841 , we have at least 95 % 95\% 95% confidence ( p < 0.05 p < 0.05 p < 0.05 ) to reject independence and conclude A A A and B B B are associated.
If K 2 < 3.841 K^2 < 3.841 K 2 < 3.841 , there is insufficient evidence at the 5 % 5\% 5% significance level to claim association.
- 5. Worked Gaokao Exam Examples
- Example 1: Normal Distribution Probability Calculation
Problem : The weight X X X of apples produced by an orchard follows a normal distribution N ( 200 , 100 ) N(200, 100) N ( 200 , 100 ) (in grams, so μ = 200 \mu = 200 μ = 200 , σ 2 = 100 ⟹ σ = 10 \sigma^2 = 100 \implies \sigma = 10 σ 2 = 100 ⟹ σ = 10 ).
Find P ( 190 ≤ X ≤ 220 ) P(190 \le X \le 220) P ( 190 ≤ X ≤ 220 ) .
If 10,000 apples are harvested, estimate the number of apples weighing more than 220 g 220\text{ g} 220 g .
Detailed Solution :
Express intervals in terms of μ \mu μ and σ \sigma σ :
- μ − σ = 200 − 10 = 190 \mu - \sigma = 200 - 10 = 190 μ − σ = 200 − 10 = 190
- μ + σ = 200 + 10 = 210 \mu + \sigma = 200 + 10 = 210 μ + σ = 200 + 10 = 210
- μ + 2 σ = 200 + 20 = 220 \mu + 2\sigma = 200 + 20 = 220 μ + 2 σ = 200 + 20 = 220
We know from the empirical rule:
P ( 190 ≤ X ≤ 210 ) = P ( μ − σ ≤ X ≤ μ + σ ) ≈ 0.6827 P(190 \le X \le 210) = P(\mu-\sigma \le X \le \mu+\sigma) \approx 0.6827 P ( 190 ≤ X ≤ 210 ) = P ( μ − σ ≤ X ≤ μ + σ ) ≈ 0.6827
P ( 180 ≤ X ≤ 220 ) = P ( μ − 2 σ ≤ X ≤ μ + 2 σ ) ≈ 0.9545 P(180 \le X \le 220) = P(\mu-2\sigma \le X \le \mu+2\sigma) \approx 0.9545 P ( 180 ≤ X ≤ 220 ) = P ( μ − 2 σ ≤ X ≤ μ + 2 σ ) ≈ 0.9545
By symmetry:
P ( X ≥ 220 ) = 1 − P ( 180 ≤ X ≤ 220 ) 2 ≈ 1 − 0.9545 2 = 0.0455 2 = 0.02275 P(X \ge 220) = \frac{1 - P(180 \le X \le 220)}{2} \approx \frac{1 - 0.9545}{2} = \frac{0.0455}{2} = 0.02275 P ( X ≥ 220 ) = 2 1 − P ( 180 ≤ X ≤ 220 ) ≈ 2 1 − 0.9545 = 2 0.0455 = 0.02275
P ( X ≤ 190 ) = 1 − P ( 190 ≤ X ≤ 210 ) 2 ≈ 1 − 0.6827 2 = 0.3173 2 = 0.15865 P(X \le 190) = \frac{1 - P(190 \le X \le 210)}{2} \approx \frac{1 - 0.6827}{2} = \frac{0.3173}{2} = 0.15865 P ( X ≤ 190 ) = 2 1 − P ( 190 ≤ X ≤ 210 ) ≈ 2 1 − 0.6827 = 2 0.3173 = 0.15865
Therefore: P ( 190 ≤ X ≤ 220 ) = 1 − P ( X < 190 ) − P ( X > 220 ) = 1 − 0.15865 − 0.02275 = 0.8186 P(190 \le X \le 220) = 1 - P(X < 190) - P(X > 220) = 1 - 0.15865 - 0.02275 = 0.8186 P ( 190 ≤ X ≤ 220 ) = 1 − P ( X < 190 ) − P ( X > 220 ) = 1 − 0.15865 − 0.02275 = 0.8186
Expected count weighing > 220 g > 220\text{ g} > 220 g : N > 220 = 10000 × P ( X > 220 ) = 10000 × 0.02275 ≈ 227.5 ≈ 228 apples N_{>220} = 10000 \times P(X > 220) = 10000 \times 0.02275 \approx 227.5 \approx 228 \text{ apples} N > 220 = 10000 × P ( X > 220 ) = 10000 × 0.02275 ≈ 227.5 ≈ 228 apples
Example 2: Comprehensive Statistics (Regression + χ 2 \chi^2 χ 2 Test)
Problem : A study investigated the relationship between weekly study hours ( x x x ) and exam score ( y y y ) for 5 students:
- Student 1 2 3 4 5
- Study Hours x i x_i x i 2 4 5 6 8
- Exam Score y i y_i y i 60 70 75 80 90
Calculate the linear regression equation y ^ = b ^ x + a ^ \hat{y} = \hat{b}x + \hat{a} y ^ = b ^ x + a ^ .
Predict the score of a student who studies 7 hours per week.
In a survey of 100 students on whether studying > 5 > 5 > 5 hours improves passing rates, a 2 × 2 2 \times 2 2 × 2 table gave a = 40 , b = 10 , c = 20 , d = 30 a=40, b=10, c=20, d=30 a = 40 , b = 10 , c = 20 , d = 30 . Test if studying > 5 >5 > 5 hours is associated with passing at the 99 % 99\% 99% confidence level ( x 0.01 = 6.635 x_{0.01} = 6.635 x 0.01 = 6.635 ).
Detailed Solution :
Regression Calculations :
x ˉ = 2 + 4 + 5 + 6 + 8 5 = 25 5 = 5 \bar{x} = \frac{2 + 4 + 5 + 6 + 8}{5} = \frac{25}{5} = 5 x ˉ = 5 2 + 4 + 5 + 6 + 8 = 5 25 = 5
y ˉ = 60 + 70 + 75 + 80 + 90 5 = 375 5 = 75 \bar{y} = \frac{60 + 70 + 75 + 80 + 90}{5} = \frac{375}{5} = 75 y ˉ = 5 60 + 70 + 75 + 80 + 90 = 5 375 = 75
∑ x i 2 = 2 2 + 4 2 + 5 2 + 6 2 + 8 2 = 4 + 16 + 25 + 36 + 64 = 145 \sum x_i^2 = 2^2 + 4^2 + 5^2 + 6^2 + 8^2 = 4 + 16 + 25 + 36 + 64 = 145 ∑ x i 2 = 2 2 + 4 2 + 5 2 + 6 2 + 8 2 = 4 + 16 + 25 + 36 + 64 = 145
∑ x i y i = ( 2 × 60 ) + ( 4 × 70 ) + ( 5 × 75 ) + ( 6 × 80 ) + ( 8 × 90 ) = 120 + 280 + 375 + 480 + 720 = 1975 \sum x_i y_i = (2 \times 60) + (4 \times 70) + (5 \times 75) + (6 \times 80) + (8 \times 90) = 120 + 280 + 375 + 480 + 720 = 1975 ∑ x i y i = ( 2 × 60 ) + ( 4 × 70 ) + ( 5 × 75 ) + ( 6 × 80 ) + ( 8 × 90 ) = 120 + 280 + 375 + 480 + 720 = 1975
Now compute slope b ^ \hat{b} b ^ : b ^ = ∑ x i y i − 5 x ˉ y ˉ ∑ x i 2 − 5 x ˉ 2 = 1975 − 5 ( 5 ) ( 75 ) 145 − 5 ( 25 ) = 1975 − 1875 145 − 125 = 100 20 = 5 \hat{b} = \frac{\sum x_i y_i - 5 \bar{x} \bar{y}}{\sum x_i^2 - 5 \bar{x}^2} = \frac{1975 - 5(5)(75)}{145 - 5(25)} = \frac{1975 - 1875}{145 - 125} = \frac{100}{20} = 5 b ^ = ∑ x i 2 − 5 x ˉ 2 ∑ x i y i − 5 x ˉ y ˉ = 145 − 5 ( 25 ) 1975 − 5 ( 5 ) ( 75 ) = 145 − 125 1975 − 1875 = 20 100 = 5
Compute intercept a ^ \hat{a} a ^ : a ^ = y ˉ − b ^ x ˉ = 75 − 5 ( 5 ) = 50 \hat{a} = \bar{y} - \hat{b} \bar{x} = 75 - 5(5) = 50 a ^ = y ˉ − b ^ x ˉ = 75 − 5 ( 5 ) = 50
Thus, the regression equation is: y ^ = 5 x + 50 \hat{y} = 5x + 50 y ^ = 5 x + 50
Prediction for x = 7 x = 7 x = 7 : y ^ = 5 ( 7 ) + 50 = 35 + 50 = 85 points \hat{y} = 5(7) + 50 = 35 + 50 = 85 \text{ points} y ^ = 5 ( 7 ) + 50 = 35 + 50 = 85 points
χ 2 \chi^2 χ 2 Independence Test : n = 40 + 10 + 20 + 30 = 100 n = 40 + 10 + 20 + 30 = 100 n = 40 + 10 + 20 + 30 = 100 . a d − b c = ( 40 × 30 ) − ( 10 × 20 ) = 1200 − 200 = 1000 ad - bc = (40 \times 30) - (10 \times 20) = 1200 - 200 = 1000 a d − b c = ( 40 × 30 ) − ( 10 × 20 ) = 1200 − 200 = 1000 . ( a d − b c ) 2 = 1000 2 = 1 , 000 , 000 (ad - bc)^2 = 1000^2 = 1,000,000 ( a d − b c ) 2 = 100 0 2 = 1 , 000 , 000 .
Row totals: a + b = 50 a+b = 50 a + b = 50 , c + d = 50 c+d = 50 c + d = 50 . Column totals: a + c = 60 a+c = 60 a + c = 60 , b + d = 40 b+d = 40 b + d = 40 . K 2 = 100 × 1 , 000 , 000 50 × 50 × 60 × 40 = 100 , 000 , 000 6 , 000 , 000 = 100 6 ≈ 16.667 K^2 = \frac{100 \times 1,000,000}{50 \times 50 \times 60 \times 40} = \frac{100,000,000}{6,000,000} = \frac{100}{6} \approx 16.667 K 2 = 50 × 50 × 60 × 40 100 × 1 , 000 , 000 = 6 , 000 , 000 100 , 000 , 000 = 6 100 ≈ 16.667
Since K 2 ≈ 16.667 > 6.635 = x 0.01 K^2 \approx 16.667 > 6.635 = x_{0.01} K 2 ≈ 16.667 > 6.635 = x 0.01 (and even > 10.828 = x 0.001 > 10.828 = x_{0.001} > 10.828 = x 0.001 ), we have over 99.9 % 99.9\% 99.9% confidence to reject independence and conclude that studying > 5 >5 > 5 hours per week is significantly associated with passing.