Sampling distribution of the sample mean: The distribution of all possible sample means from samples of size n drawn from a population with mean μ and standard deviation σ
- Identify the population parameters: μ and σ
- Identify the sample size: n
- Calculate the mean of the sampling distribution: μₓ̄ = μ
- Calculate the standard deviation of the sampling distribution: σₓ̄ = σ/√n
Population mean μ = 1050, Population standard deviation σ = 200
The mean of the sampling distribution equals the population mean: μₓ̄ = μ = 1050
The standard deviation of the sampling distribution is: σₓ̄ = σ/√n = 200/√25 = 200/5 = 40
The sampling distribution of the sample mean has mean 1050 and standard deviation 40
The sampling distribution of the sample mean has mean μₓ̄ = 1050 and standard deviation σₓ̄ = 40
• Mean property: μₓ̄ = μ (sampling distribution mean equals population mean)
• Standard deviation property: σₓ̄ = σ/√n (standard error decreases as sample size increases)
• Central limit theorem: As n increases, sampling distribution approaches normality
• The standard deviation of the sampling distribution is called the standard error
• Larger sample sizes lead to smaller standard errors
• The sampling distribution becomes more concentrated around the population mean as n increases
Normal approximation: When n ≥ 30, the sampling distribution of the sample mean is approximately normal regardless of the population distribution shape
μₓ̄ = μ = 175, σₓ̄ = σ/√n = 8/√64 = 8/8 = 1
z = (X̄ - μₓ̄)/σₓ̄ = (177 - 175)/1 = 2/1 = 2
P(Z > 2) = 1 - P(Z ≤ 2) = 1 - 0.9772 = 0.0228
There is a 2.28% chance that the sample mean exceeds 177 cm
The probability that the sample mean is greater than 177 cm is 0.0228 or 2.28%
• Z-score formula: z = (X̄ - μₓ̄)/σₓ̄ for standardizing sample means
• Standard error: σₓ̄ = σ/√n reduces uncertainty with larger samples
• Normal distribution: Use standard normal table for probabilities
• Even though we started with a normal population, the sampling distribution is also normal
• The probability is small because 177 cm is 2 standard errors above the mean
• Z-scores tell us how many standard errors a sample mean is from the population mean
Central Limit Theorem: For large samples (n ≥ 30), the sampling distribution of the sample mean approaches a normal distribution regardless of the population distribution shape
μₓ̄ = μ = 50, σₓ̄ = σ/√n = 15/√100 = 15/10 = 1.5
For X̄ = 48: z₁ = (48 - 50)/1.5 = -2/1.5 = -1.33
For X̄ = 52: z₂ = (52 - 50)/1.5 = 2/1.5 = 1.33
P(48 < X̄ < 52) = P(-1.33 < Z < 1.33) = P(Z < 1.33) - P(Z < -1.33)
= 0.9082 - 0.0918 = 0.8164
There is an 81.64% chance that the sample mean falls between 48 and 52
The probability that the sample mean falls between 48 and 52 is 0.8164 or 81.64%
• CLT condition: n = 100 ≥ 30, so normal approximation is valid
• Standardization: Convert sample means to z-scores using σₓ̄ = σ/√n
• Probability calculation: P(a < X̄ < b) = P(z₁ < Z < z₂)
• The interval [48, 52] is symmetric around the population mean μ = 50
• With n = 100, the standard error is quite small (1.5), making the sampling distribution tight
• The CLT allows us to use normal distribution even if the original population isn't normal
Central Limit Theorem (CLT): For sufficiently large samples, the sampling distribution of the sample mean approaches a normal distribution
Sampling Distribution: Distribution of all possible sample statistics for a given sample size
Standard Error: Standard deviation of the sampling distribution
- Verify CLT conditions: Check if n ≥ 30 (or population is normal)
- Identify parameters: μ, σ, and n
- Calculate sampling distribution parameters: μₓ̄ = μ and σₓ̄ = σ/√n
- Standardize: Use z-score formula for probabilities
- Calculate probability: Use standard normal table
• Mean of sampling distribution: μₓ̄ = μ
• Standard deviation of sampling distribution: σₓ̄ = σ/√n
• Shape approaches normal as n increases
• Works for any population distribution when n ≥ 30
Standard Error: σₓ̄ = σ/√n measures the variability of sample means around the population mean
σₓ̄₁ = σ/√n₁ = 20/√25 = 20/5 = 4
σₓ̄₂ = σ/√n₂ = 20/√100 = 20/10 = 2
σₓ̄₂ = 2 is half of σₓ̄₁ = 4
Increasing sample size from 25 to 100 reduces standard error by 50%, improving precision
For n₁ = 25: σₓ̄₁ = 4; For n₂ = 100: σₓ̄₂ = 2. Larger samples provide more precise estimates.
• Standard error formula: σₓ̄ = σ/√n shows inverse relationship with √n
• Precision improvement: Doubling sample size reduces standard error by √2 ≈ 1.41
• Statistical efficiency: Larger samples yield more reliable estimates
• The relationship between sample size and precision follows the square root law
• To halve the standard error, you need to quadruple the sample size
• This demonstrates why larger samples are preferred in statistical studies
Real-world application: Using CLT to make inferences about population parameters from sample statistics
n = 50 ≥ 30, so normal approximation is valid
μₓ̄ = μ = 30, σₓ̄ = σ/√n = 10/√50 = 10/7.07 ≈ 1.41
z = (X̄ - μₓ̄)/σₓ̄ = (28 - 30)/1.41 = -2/1.41 ≈ -1.41
P(X̄ < 28) = P(Z < -1.41) = 0.0793
There is a 7.93% chance that the sample mean is less than 28 minutes
The probability that the sample mean commute time is less than 28 minutes is 0.0793 or 7.93%
• CLT applicability: n = 50 is large enough for normal approximation
• Standard error calculation: σₓ̄ = σ/√n accounts for sample size effect
• Probability interpretation: Small probability indicates unusual sample mean
• A sample mean of 28 minutes is about 1.41 standard errors below the population mean
• This relatively low probability suggests such a sample would be somewhat unusual
• The CLT allows us to quantify how unusual this sample mean would be
Central Limit Theorem (CLT): The sampling distribution of the sample mean approaches a normal distribution as the sample size increases, regardless of the population distribution shape
Sampling Distribution: The probability distribution of a statistic obtained from repeated sampling of a population
Standard Error: The standard deviation of the sampling distribution of a statistic
Sample Mean: The average of observations in a sample, denoted as X̄
- Verify conditions: Ensure n ≥ 30 (or population is normal) for CLT application
- Identify parameters: Determine μ (population mean), σ (population SD), and n (sample size)
- Calculate sampling distribution parameters: μₓ̄ = μ and σₓ̄ = σ/√n
- Standardize the sample mean: Use z-score formula z = (X̄ - μ)/(σ/√n)
- Compute probabilities: Use standard normal table or calculator
• Mean of sampling distribution: μₓ̄ = μ (unbiased estimator)
• Standard deviation of sampling distribution: σₓ̄ = σ/√n (standard error)
• Shape approaches normal as n increases (regardless of population shape)
• Rule of thumb: n ≥ 30 ensures adequate normal approximation
• Z-score for sample mean: z = (X̄ - μ)/(σ/√n)
Questions & Answers
Question: I don't understand why the standard deviation of the sampling distribution is σ/√n instead of just σ. Can you explain the intuition behind this formula?
Answer: Great question! The key insight is that sample means are less variable than individual observations. Here's the intuition:
- Individual observations can vary widely around the population mean
- Sample means, however, represent averages of multiple observations
- Averaging tends to cancel out extreme values, making sample means cluster more closely around the population mean
Think of it this way: if you flip a coin once, you might get heads or tails (high variability). But if you flip it 100 times and calculate the proportion of heads, you're much more likely to get close to 0.5 (lower variability).
The √n factor comes from mathematical derivation showing that when you average n independent random variables, the variance gets divided by n, so the standard deviation gets divided by √n.
This is why larger samples provide more precise estimates - the standard error decreases as √n increases.
Question: When exactly do I need to apply the Central Limit Theorem? Do I always use it when dealing with sample means?
Answer: The Central Limit Theorem applies under specific conditions:
- Large sample size: When n ≥ 30, the sampling distribution of the sample mean is approximately normal regardless of the population distribution
- Small sample size: If n < 30, you can only assume normality if the population itself is normally distributed
You don't always need to use the CLT. If the population is known to be normal, then the sampling distribution of the sample mean is normal for any sample size. The CLT is specifically needed when the population distribution is unknown or non-normal.
For example, if you're sampling from a normal population with n = 15, you can use normal probability calculations directly. But if you're sampling from an unknown population with n = 15, you cannot rely on normality without additional assumptions.
Always check the sample size condition: n ≥ 30 for the CLT to apply.
Question: I have trouble distinguishing between the population parameters (μ, σ) and the sampling distribution parameters (μₓ̄, σₓ̄). Can you clarify this?
Answer: Here's a clear distinction between these parameters:
- Population parameters (μ, σ): Describe the entire population
- μ = population mean (center of the population distribution)
- σ = population standard deviation (spread of individual values in the population)
- Sampling distribution parameters (μₓ̄, σₓ̄): Describe the distribution of sample means
- μₓ̄ = mean of sampling distribution = μ (same as population mean)
- σₓ̄ = standard deviation of sampling distribution = σ/√n (also called standard error)
Think of it this way: The population has a certain average and spread. When you take samples and calculate their means, those sample means also have an average and spread, but the spread of sample means (σₓ̄) is always smaller than the spread of individual values (σ).
For example, if individual heights vary by σ = 10 cm, then sample means based on n = 25 people would vary by σₓ̄ = 10/√25 = 2 cm.
Question: How does the sample size affect the accuracy of probability calculations using the CLT?
Answer: Sample size affects both the validity and accuracy of CLT applications:
Validity: The CLT requires n ≥ 30 for the normal approximation to be reasonable. Smaller samples may not provide accurate probability calculations.
Accuracy: Larger samples provide better approximations:
- For n = 30: Moderate approximation quality
- For n = 50: Better approximation quality
- For n = 100+: Very good approximation quality
Additionally, larger samples result in smaller standard errors (σₓ̄ = σ/√n), which means sample means cluster more tightly around the population mean. This makes probability statements more precise.
For example, with n = 30, there's more uncertainty in where the sample mean falls compared to n = 100, where the sample mean is much more predictable.
The relationship is: larger n → smaller σₓ̄ → more precise estimates → more accurate probability calculations.
Question: What happens to the sampling distribution when the population distribution is already normal? Does the CLT still apply?
Answer: This is an important special case! When the population is normally distributed:
- The sampling distribution of the sample mean is exactly normal for any sample size
- The CLT is not needed as a justification for normality
- The sampling distribution properties still hold: μₓ̄ = μ and σₓ̄ = σ/√n
So if X ~ N(μ, σ²), then X̄ ~ N(μ, σ²/n) exactly, regardless of the value of n.
The CLT is primarily useful when the population distribution is unknown or non-normal. In those cases, we need a large sample size (n ≥ 30) to guarantee approximate normality of the sampling distribution.
However, even when the population is normal, the CLT results (μₓ̄ = μ and σₓ̄ = σ/√n) still describe the sampling distribution perfectly.
This is why problems sometimes specify that the population is normal - it allows exact probability calculations even for small samples.