Sampling Distribution of Sample Proportion: The probability distribution of all possible sample proportions p̂ from samples of size n drawn from a population with proportion p. The mean is μₚ̂ = p and the standard deviation is σₚ̂ = √[p(1-p)/n].
- Identify population proportion p and sample size n
- Calculate mean: μₚ̂ = p
- Calculate standard deviation: σₚ̂ = √[p(1-p)/n]
- Verify conditions for normal approximation if needed
Population proportion p = 0.4, sample size n = 100.
For the sampling distribution of p̂: μₚ̂ = p = 0.4.
Standard deviation of p̂: σₚ̂ = √[p(1-p)/n] = √[0.4(1-0.4)/100] = √[0.4×0.6/100] = √0.0024 ≈ 0.0490.
The sampling distribution of the sample proportion has mean μₚ̂ = 0.4 and standard deviation σₚ̂ ≈ 0.0490.
• Mean of p̂: μₚ̂ = p
• Standard deviation of p̂: σₚ̂ = √[p(1-p)/n]
• Standard error: The standard deviation of the sampling distribution
Normal Approximation Conditions: For the sampling distribution of p̂, use normal approximation if both np ≥ 5 and n(1-p) ≥ 5. This ensures the distribution is not too skewed.
np = 100 × 0.4 = 40 ≥ 5 ✓ and n(1-p) = 100 × 0.6 = 60 ≥ 5 ✓. Both conditions are satisfied.
Since conditions are met: p̂ ~ N(μₚ̂, σₚ̂²) = N(0.4, 0.0490²).
For large samples satisfying the conditions, we can use the normal distribution to approximate probabilities related to the sample proportion.
Yes, the normal approximation can be used. The sampling distribution of p̂ is approximately normal with mean 0.4 and standard deviation 0.0490.
• Normal approximation conditions: np ≥ 5 and n(1-p) ≥ 5
• Normal distribution: p̂ ~ N(p, p(1-p)/n)
• Continuity correction: May be needed for discrete approximations
Effect of Sample Size: As sample size increases, the standard deviation of the sampling distribution of p̂ decreases. The relationship is σₚ̂ = √[p(1-p)/n], so standard deviation decreases as 1/√n.
σₚ̂ = √[p(1-p)/n₁] = √[0.3(0.7)/50] = √[0.21/50] = √0.0042 ≈ 0.0648.
σₚ̂ = √[p(1-p)/n₂] = √[0.3(0.7)/200] = √[0.21/200] = √0.00105 ≈ 0.0324.
When sample size quadruples from 50 to 200, the standard deviation halves from 0.0648 to 0.0324. This demonstrates the 1/√n relationship.
For n₁ = 50, standard deviation ≈ 0.0648. For n₂ = 200, standard deviation ≈ 0.0324. Increasing sample size from 50 to 200 reduces the standard deviation by approximately half.
• Standard deviation decreases: As n increases, σₚ̂ = √[p(1-p)/n] decreases
• Sample size effect: Quadrupling n halves the standard deviation
• 1/√n relationship: Standard deviation is inversely proportional to √n
Sample Proportion: The proportion of successes in a sample of size n. Denoted as p̂ = X/n where X is the number of successes.
Sampling Distribution of p̂: The probability distribution of all possible sample proportions from samples of size n drawn from a population with proportion p.
Standard Error of Proportion: The standard deviation of the sampling distribution of p̂, calculated as √[p(1-p)/n].
Normal Approximation: For large samples, the sampling distribution of p̂ can be approximated by a normal distribution.
Continuity Correction: Adjustment made when approximating discrete distributions with continuous distributions.
- Identify population parameters: Find p (population proportion)
- Determine sample size: Identify n
- Check conditions: Verify np ≥ 5 and n(1-p) ≥ 5
- Calculate parameters: Find μₚ̂ = p and σₚ̂ = √[p(1-p)/n]
- Select approach: Use exact binomial or normal approximation
- Calculate probabilities: Apply appropriate method
• Mean of p̂: μₚ̂ = p
• Standard deviation of p̂: σₚ̂ = √[p(1-p)/n]
• Standard error: SE = √[p(1-p)/n]
• Normal approximation conditions: np ≥ 5 and n(1-p) ≥ 5
• Normal approximation: p̂ ~ N(p, p(1-p)/n)
Probability Calculation: Using the normal approximation to the sampling distribution of p̂, we can calculate probabilities related to sample proportions by standardizing with z-scores: z = (p̂ - p) / √[p(1-p)/n].
np = 400 × 0.25 = 100 ≥ 5 ✓ and n(1-p) = 400 × 0.75 = 300 ≥ 5 ✓. Normal approximation is appropriate.
SE = √[p(1-p)/n] = √[0.25(0.75)/400] = √[0.1875/400] = √0.00046875 ≈ 0.02165.
For p̂ = 0.20: z₁ = (0.20 - 0.25) / 0.02165 = -0.05 / 0.02165 ≈ -2.31
For p̂ = 0.30: z₂ = (0.30 - 0.25) / 0.02165 = 0.05 / 0.02165 ≈ 2.31
P(0.20 ≤ p̂ ≤ 0.30) = P(-2.31 ≤ Z ≤ 2.31) = P(Z ≤ 2.31) - P(Z ≤ -2.31) ≈ 0.9896 - 0.0104 = 0.9792.
The probability that the sample proportion is between 0.20 and 0.30 is approximately 0.9792 or 97.92%.
• Standardization: z = (p̂ - p) / √[p(1-p)/n]
• Normal probability: Use standard normal tables or calculator
• Continuity correction: May be needed depending on exact requirements
Margin of Error: The maximum expected difference between the sample proportion and the population proportion at a given confidence level. MOE = z* × SE where z* is the critical value and SE is the standard error.
SE = √[p̂(1-p̂)/n] = √[0.52(0.48)/1000] = √[0.2496/1000] = √0.0002496 ≈ 0.0158.
For 95% confidence level, z* = 1.96 (from standard normal distribution).
MOE = z* × SE = 1.96 × 0.0158 ≈ 0.0310.
We are 95% confident that the true population proportion is within ±0.0310 of the sample proportion 0.52, i.e., between 0.489 and 0.551.
The margin of error for a 95% confidence interval is approximately 0.0310 or 3.1%.
• Standard error: SE = √[p̂(1-p̂)/n]
• Margin of error: MOE = z* × SE
• Confidence interval: p̂ ± MOE
Sample Proportion: The proportion of successes in a sample of size n. Denoted as p̂ = X/n where X is the number of successes.
Sampling Distribution of p̂: The probability distribution of all possible sample proportions from samples of size n drawn from a population with proportion p.
Standard Error of Proportion: The standard deviation of the sampling distribution of p̂, calculated as √[p(1-p)/n].
Normal Approximation: For large samples, the sampling distribution of p̂ can be approximated by a normal distribution when np ≥ 5 and n(1-p) ≥ 5.
Margin of Error: The maximum expected difference between the sample proportion and the population proportion at a given confidence level.
- Identify population parameters: Find p (population proportion)
- Determine sample size: Identify n
- Check conditions: Verify np ≥ 5 and n(1-p) ≥ 5
- Calculate parameters: Find μₚ̂ = p and σₚ̂ = √[p(1-p)/n]
- Select approach: Use exact binomial or normal approximation
- Calculate probabilities: Apply appropriate method
- Interpret results: Understand practical implications
• Mean of p̂: μₚ̂ = p
• Standard deviation of p̂: σₚ̂ = √[p(1-p)/n]
• Standard error: SE = √[p(1-p)/n]
• Normal approximation conditions: np ≥ 5 and n(1-p) ≥ 5
• Normal approximation: p̂ ~ N(p, p(1-p)/n)
• Confidence interval: p̂ ± z* × √[p̂(1-p̂)/n]
• Simple: Calculating sampling distribution parameters
• Intermediate: Checking normal approximation conditions
• Advanced: Probability calculations using normal approximation
• Complex: Confidence intervals and hypothesis testing
• Sampling distribution of p̂ describes how sample proportions vary
• Mean of sampling distribution equals population proportion
• Standard error decreases with larger sample sizes
• Normal approximation requires np ≥ 5 and n(1-p) ≥ 5
• Margin of error = critical value × standard error
Questions & Answers
Question: Why do we need np ≥ 5 and n(1-p) ≥ 5 for normal approximation?
Answer: These conditions ensure the binomial distribution underlying p̂ is not too skewed:
- np ≥ 5: Ensures enough expected successes
- n(1-p) ≥ 5: Ensures enough expected failures
- Together: Prevents the distribution from being too extreme (too close to 0 or 1)
When these conditions are met, the binomial distribution is approximately symmetric and can be well-approximated by a normal distribution.
Question: What happens to the sampling distribution as sample size increases?
Answer: As sample size increases:
- Standard error decreases: σₚ̂ = √[p(1-p)/n] becomes smaller
- More precise estimates: Sample proportions cluster closer to the population proportion
- More normal: Distribution becomes closer to normal (CLT)
The sampling distribution becomes taller and narrower, indicating less variability in the sample proportions.
Question: When constructing confidence intervals, should I use p or p̂ in the standard error formula?
Answer: This depends on the context:
- For confidence intervals: Use p̂ (sample proportion) since p is unknown
- For probability calculations about p̂: Use p (population proportion) if known
- For planning studies: Use a prior estimate or 0.5 (conservative)
The confidence interval formula is: p̂ ± z* × √[p̂(1-p̂)/n].