Normal Distributions and Empirical Rule (68-95-99.7)

Master normal distributions: 68-95-99.7 rule, standard deviation, z-scores, and probability calculations with 5 detailed exercises.

Exercise Solutions 1-3
1 Empirical Rule Application
Exercise 1
Heights of adult women in a city are normally distributed with mean 65 inches and standard deviation 2.5 inches. What percentage of women have heights between 60 and 70 inches?
Key Definitions:

Normal Distribution: Symmetric, bell-shaped distribution characterized by mean (μ) and standard deviation (σ)

Empirical Rule (68-95-99.7): In normal distributions, 68% within 1σ, 95% within 2σ, 99.7% within 3σ of mean

Standard Deviation (σ): Measure of spread in a normal distribution

Empirical Rule Method:
  1. Identify mean (μ) and standard deviation (σ)
  2. Calculate how many standard deviations each boundary is from the mean
  3. Apply the empirical rule to find the percentage
Step 1: Identify parameters

Mean (μ) = 65 inches, Standard deviation (σ) = 2.5 inches

Step 2: Calculate standard deviations from mean

60 inches: (60 - 65) ÷ 2.5 = -2 standard deviations

70 inches: (70 - 65) ÷ 2.5 = +2 standard deviations

Step 3: Apply empirical rule

Between μ - 2σ and μ + 2σ: 95% of the data

60 inches = μ - 2σ
70 inches = μ + 2σ
Percentage = 95%
Final Answer:

Approximately 95% of women have heights between 60 and 70 inches. This range covers 2 standard deviations below and above the mean, which corresponds to the 95% portion of the empirical rule.

Applied Rules:

• Empirical Rule: 68% within 1σ, 95% within 2σ, 99.7% within 3σ

• Standardization: (Value - Mean) ÷ Standard Deviation

• Normal distribution: Symmetric around the mean

2 Z-Score Calculation
Exercise 2
IQ scores are normally distributed with mean 100 and standard deviation 15. What percentage of people have IQ scores above 130? What is the z-score for an IQ of 130?
Key Definitions:

Z-Score: Number of standard deviations a value is from the mean (z = (x - μ) ÷ σ)

Standard Normal Distribution: Normal distribution with μ = 0 and σ = 1

Z-Score and Probability Method:
  1. Calculate the z-score using the formula
  2. Use the empirical rule to find the percentage
  3. Account for the portion above the threshold
Step 1: Calculate the z-score

z = (130 - 100) ÷ 15 = 30 ÷ 15 = 2

Step 2: Apply empirical rule

Within 2 standard deviations: 95% of data

Outside 2 standard deviations: 100% - 95% = 5%

Step 3: Find percentage above 130

Since normal distribution is symmetric, 2.5% is above +2σ and 2.5% is below -2σ

Percentage above 130 = 2.5%

Z-score for 130 = 2.0
Percentage above 130 = 2.5%
Final Answer:

The z-score for an IQ of 130 is 2.0, and approximately 2.5% of people have IQ scores above 130. This is because 130 is 2 standard deviations above the mean, and 2.5% of the data lies beyond 2 standard deviations in the upper tail.

Applied Rules:

• Z-score formula: z = (x - μ) ÷ σ

• Empirical rule: 95% within 2σ means 2.5% in each tail

• Symmetry: Normal distribution is symmetric around the mean

3 Percentile Calculation
Exercise 3
SAT scores are normally distributed with mean 1000 and standard deviation 200. What score corresponds to the 16th percentile? What about the 84th percentile?
Key Definitions:

Percentile: Value below which a given percentage of observations fall

16th Percentile: Value below which 16% of the data falls

84th Percentile: Value below which 84% of the data falls

Percentile Conversion Method:
  1. Identify the corresponding z-score for the percentile
  2. Use the empirical rule to find the z-score
  3. Convert z-score back to original units using: x = μ + zσ
Step 1: Find z-score for 16th percentile

From empirical rule: 68% within 1σ means 32% outside, so 16% in each tail

16th percentile corresponds to z = -1 (1σ below mean)

Step 2: Calculate score for 16th percentile

x = μ + zσ = 1000 + (-1)(200) = 1000 - 200 = 800

Step 3: Find score for 84th percentile

84th percentile = 100% - 16% = 84% below, which corresponds to z = +1

x = μ + zσ = 1000 + (1)(200) = 1000 + 200 = 1200

16th percentile = 800
84th percentile = 1200
Difference = 400 points
Final Answer:

The 16th percentile corresponds to a score of 800, and the 84th percentile corresponds to a score of 1200. These values are 1 standard deviation below and above the mean, respectively, encompassing the middle 68% of the distribution.

Applied Rules:

• Empirical rule: 68% within 1σ means 16% in each tail

• Percentile conversion: x = μ + zσ

• Symmetry: 16th and 84th percentiles are equidistant from the mean

Exercise Solutions 4-5
4 Finding Values Using Empirical Rule
Exercise 4
The weights of bags of flour are normally distributed with mean 5 pounds and standard deviation 0.2 pounds. Between what two weights would you expect to find the middle 99.7% of the bags?
Key Definitions:

99.7% Rule: Approximately 99.7% of data falls within 3 standard deviations of the mean

Range for 99.7%: From μ - 3σ to μ + 3σ

Range Calculation Method:
  1. Identify mean (μ) and standard deviation (σ)
  2. Calculate lower bound: μ - 3σ
  3. Calculate upper bound: μ + 3σ
  4. State the range
Step 1: Identify parameters

Mean (μ) = 5 pounds, Standard deviation (σ) = 0.2 pounds

Step 2: Calculate lower bound

Lower bound = μ - 3σ = 5 - 3(0.2) = 5 - 0.6 = 4.4 pounds

Step 3: Calculate upper bound

Upper bound = μ + 3σ = 5 + 3(0.2) = 5 + 0.6 = 5.6 pounds

Lower bound = 4.4 lbs
Upper bound = 5.6 lbs
Range: 4.4 to 5.6 pounds
Final Answer:

The middle 99.7% of bags would weigh between 4.4 and 5.6 pounds. This range represents 3 standard deviations below and above the mean, capturing nearly all of the data in a normal distribution.

Applied Rules:

• 99.7% rule: 99.7% within 3σ of the mean

• Range calculation: μ ± 3σ

• Almost all data: 99.7% captures virtually all observations

5 Complex Probability Problem
Exercise 5
Cholesterol levels in adults are normally distributed with mean 200 mg/dL and standard deviation 30 mg/dL. What percentage of adults have cholesterol levels between 170 and 260 mg/dL?
Key Definitions:

Interval Probability: Probability that a value falls within a specific range

Combined Empirical Rules: Using multiple parts of the 68-95-99.7 rule

Interval Probability Method:
  1. Convert boundaries to standard deviations from the mean
  2. Identify which empirical rule ranges apply
  3. Combine appropriate percentages
Step 1: Convert boundaries to standard deviations

170 mg/dL: (170 - 200) ÷ 30 = -1 standard deviation

260 mg/dL: (260 - 200) ÷ 30 = +2 standard deviations

Step 2: Break down the interval

From μ - 1σ to μ + 2σ

This includes: from μ - 1σ to μ (half of 68%) + from μ to μ + 2σ (half of 95%)

Step 3: Calculate the probability

From μ - 1σ to μ: 68% ÷ 2 = 34%

From μ to μ + 2σ: 95% ÷ 2 = 47.5%

Total: 34% + 47.5% = 81.5%

170 mg/dL = μ - 1σ
260 mg/dL = μ + 2σ
Percentage = 81.5%
Final Answer:

Approximately 81.5% of adults have cholesterol levels between 170 and 260 mg/dL. This interval spans from 1 standard deviation below the mean to 2 standard deviations above the mean.

Applied Rules:

• 68-95-99.7 rule: Use appropriate portions for partial ranges

• Symmetry: Half of each percentage applies to each side

• Interval combination: Add probabilities for overlapping ranges

Comprehensive Summary: Normal Distributions & Empirical Rule
z = \(\frac{x - \mu}{\sigma}\)
Z-Score Formula
68%
Within 1σ
95%
Within 2σ
99.7%
Within 3σ
-3σ
-2σ
-1σ
μ
+1σ
+2σ
+3σ
Key Concepts:

Normal Distribution: A continuous probability distribution that is symmetric and bell-shaped, completely determined by its mean (μ) and standard deviation (σ).

Empirical Rule (68-95-99.7 Rule): In any normal distribution, approximately 68% of values lie within 1 standard deviation of the mean, 95% within 2 standard deviations, and 99.7% within 3 standard deviations.

Z-Score: A standardized value that indicates how many standard deviations an element is from the mean. Formula: z = (x - μ) / σ.

Core Methods and Procedures:
  1. Standardization: Convert raw scores to z-scores using z = (x - μ) / σ
  2. Empirical Rule Application: Use 68-95-99.7 percentages for quick estimates
  3. Percentile Conversion: x = μ + zσ to convert z-scores back to original units
  4. Probability Calculation: Determine the area under the curve for specific intervals
Tip 1: Remember: 68-95-99.7. The percentages increase by approximately 27% each time you add another standard deviation.
Tip 2: In a normal distribution, the mean, median, and mode are all equal and located at the center of the distribution.
Tip 3: The empirical rule only applies to normal distributions. Check for symmetry before applying.
Tip 4: Use symmetry: 16% lies below μ - 1σ and 16% lies above μ + 1σ (since 100% - 68% = 32%, divided equally).
Common errors: Applying empirical rule to non-normal distributions, confusing percentages in tails, miscalculating z-scores.
Memorization aids: "68-95-99.7" or "68-95-99.7, that's the rule that's neat!" Think of bell shape and symmetry.
Essential Rules:

• 68-95-99.7 Rule: 68% within 1σ, 95% within 2σ, 99.7% within 3σ

• Z-score: z = (x - μ) / σ, where μ is mean and σ is standard deviation

• Symmetry: 50% of data lies above and below the mean

• Tail probabilities: 2.5% in each tail beyond 2σ, 0.15% beyond 3σ

Detailed Course Content: Normal Distributions
f(x) = \(\frac{1}{\sigma\sqrt{2\pi}}e^{-\frac{(x-\mu)^2}{2\sigma^2}}\)
Normal Distribution Function
Z-Score
z = \(\frac{x - \mu}{\sigma}\)
Standardization
Empirical Rule
68-95-99.7%
Standard deviations
Percentile Conversion
x = μ + zσ
Back to original scale
Detailed Definitions:

Parameters of Normal Distribution: The normal distribution is completely described by two parameters: the mean (μ) which determines the center of the distribution, and the standard deviation (σ) which determines the spread.

Standard Normal Distribution: A special case of the normal distribution with mean μ = 0 and standard deviation σ = 1. Any normal distribution can be converted to the standard normal using z-scores.

Inflection Points: Points on the normal curve where the curvature changes from concave to convex (at μ ± σ).

Advanced Problem-Solving Techniques:
  1. Pattern Recognition: Identify if the problem involves normal distribution by looking for mean and standard deviation
  2. Boundary Analysis: Convert raw scores to standard deviations from the mean
  3. Empirical Rule Matching: Match the problem to appropriate part of the 68-95-99.7 rule
  4. Interval Combination: For complex intervals, break them into combinations of empirical rule segments
  5. Verification: Check that probabilities make sense (between 0 and 100%)
Advanced Tip 1: For problems involving "more than" or "less than," consider using the complement rule.
Advanced Tip 2: When dealing with percentiles, remember that percentiles divide the distribution into 100 equal parts.
Advanced Tip 3: The 50th percentile is always equal to the mean in a normal distribution due to symmetry.
Advanced Tip 4: Very extreme values (beyond 3σ) are rare in normal distributions (only 0.3% combined).
Real-world applications: Heights, weights, test scores, measurement errors, IQ scores, and many natural phenomena follow normal distributions.
Quality control: Used in manufacturing to determine acceptable ranges and identify defective products.

Questions & Answers

Question: How can I tell if a distribution is normal enough to use the empirical rule? What if it's only approximately normal?

Answer: Good question! The empirical rule works best with perfectly normal distributions, but it provides reasonable approximations for distributions that are approximately normal. Look for these characteristics:

  • Symmetry: The distribution should be roughly symmetric around the mean
  • Unimodal: Should have one clear peak in the center
  • Bell-shaped: Should have the characteristic bell curve shape
  • Tail behavior: Should approach zero gradually without heavy tails

If the distribution is only approximately normal, the empirical rule still provides useful estimates, though they may be off by a few percentage points. For critical applications, consider using more precise methods or statistical software.

Question: I'm confused about why we use z-scores. Can't we just work with the original values?

Answer: Z-scores are incredibly useful because they standardize different normal distributions to a common scale. Here's why:

  • Comparison: They allow you to compare values from different normal distributions with different means and standard deviations
  • Reference: They provide a universal reference point (standard deviations from the mean)
  • Simplicity: Once converted to z-scores, you can use the same empirical rule regardless of the original units
  • Probability: They connect to standardized probability tables and calculators

For example, a height of 70 inches might be exceptional in one group but normal in another. The z-score tells you how exceptional it is relative to its own group.

Question: How do I handle problems where the values don't correspond exactly to 1, 2, or 3 standard deviations? What if I need the percentage between 1.5 and 2.5 standard deviations?

Answer: The empirical rule gives us key reference points, but for more precise values, you'll need to interpolate or use statistical tables/software. However, for estimation purposes:

  • Interpolation: Estimate between known values (e.g., 1.5σ is halfway between 1σ and 2σ)
  • Combination: Break complex intervals into combinations of known segments
  • Approximation: Use the closest standard deviation value for rough estimates

For example, between 1.5σ and 2.5σ: you could estimate by finding the percentage from 1σ to 2σ (about 13.5%) and adjusting based on the additional 0.5σ on each end. For precise work, statistical software or tables are needed.