Statistics & probability • 2026 edition
\( CI = \bar{x} \pm Z_{\alpha/2} \left( \frac{\sigma}{\sqrt{n}} \right) \)
Where:
A confidence interval estimates the range within which a population parameter lies with a certain level of confidence. For example, a 95% confidence interval means that if we took 100 samples, about 95 of them would contain the true population parameter.
Example: For a sample with \( \bar{x} = 50 \), \( \sigma = 10 \), \( n = 100 \), and 95% confidence level (\( Z = 1.96 \)):
Margin of Error: \( MOE = 1.96 \times \frac{10}{\sqrt{100}} = 1.96 \)
Confidence Interval: \( 50 \pm 1.96 = [48.04, 51.96] \)
Thus, we are 95% confident that the true population mean lies between 48.04 and 51.96.
A confidence interval is a range of values that is likely to contain an unknown population parameter. It provides an estimated range of values which is likely to include an unknown parameter, the estimated range being calculated from a given set of sample data. For example, a 95% confidence interval means that if we were to take 100 different samples and compute a confidence interval for each sample, then approximately 95 of the 100 confidence intervals will contain the true population parameter.
The standard confidence interval calculation uses the following formula:
Where:
Common confidence levels and their corresponding Z-scores:
A range of values likely to contain a population parameter with specified confidence.
\(CI = \bar{x} \pm Z \left( \frac{\sigma}{\sqrt{n}} \right)\)
Where CI=confidence interval, x̄=sample mean, Z=critical value, σ=std dev, n=sample size.
Used in research, quality control, polling, medical studies, and scientific experiments.
Which of the following best describes what a 95% confidence interval means?
The answer is B) If we took 100 samples and computed confidence intervals, about 95 would contain the true population parameter. This is the correct frequentist interpretation of confidence intervals. The confidence level represents the long-run proportion of confidence intervals that would contain the true parameter if we repeated the sampling process many times.
It's crucial to understand that a confidence interval is not a probability statement about a specific interval containing the true parameter. Instead, it's a statement about the procedure used to generate intervals. Once we've calculated a specific interval from our sample, the true parameter is either inside it or not - there's no probability involved. The 95% refers to the reliability of the method across many samples.
Confidence Interval: A range of values computed from sample data that likely contains the true population parameter
Confidence Level: The percentage of confidence intervals that would contain the true parameter if we repeated sampling many times
Population Parameter: The true value we're trying to estimate (like population mean)
• Confidence intervals are about the procedure, not individual intervals
• Higher confidence levels produce wider intervals
• Larger samples produce narrower intervals
• Remember: "We are 95% confident that..." not "There is a 95% probability that..."
• Think of confidence intervals as a net cast around the true parameter
• The confidence level reflects the reliability of the estimation procedure
• Interpreting the confidence level as the probability for a specific interval
• Confusing confidence intervals with prediction intervals
• Assuming the true parameter has a probability distribution
A sample of 64 students has a mean test score of 78 with a known population standard deviation of 12. Calculate the 90% confidence interval for the population mean. Show your work.
Using the confidence interval formula: \(CI = \bar{x} \pm Z \left( \frac{\sigma}{\sqrt{n}} \right)\)
Given:
Step 1: Calculate Standard Error: \(SE = \frac{\sigma}{\sqrt{n}} = \frac{12}{\sqrt{64}} = \frac{12}{8} = 1.5\)
Step 2: Calculate Margin of Error: \(MOE = Z \times SE = 1.645 \times 1.5 = 2.4675\)
Step 3: Calculate Confidence Interval: \(CI = 78 \pm 2.4675 = [75.5325, 80.4675]\)
Therefore, the 90% confidence interval is [75.53, 80.47].
This problem demonstrates how to apply the confidence interval formula step by step. Notice that since the population standard deviation is known and the sample size is large (n=64), we use the Z-distribution. The standard error quantifies the variability of the sample mean across different samples, and the critical value scales this to achieve the desired confidence level.
Standard Error: The standard deviation of the sampling distribution of a statistic
Margin of Error: The product of the critical value and the standard error
Critical Value: The number of standard deviations needed to achieve the desired confidence level
• Use Z-distribution when population std dev is known or n ≥ 30
• Use T-distribution when population std dev is unknown and n < 30
• Always check assumptions before calculating confidence intervals
• Remember: SE = σ/√n
• MOE = Critical Value × SE
• CI = Point Estimate ± MOE
• Forgetting to divide standard deviation by √n to get standard error
• Using the wrong critical value for the confidence level
• Mixing up standard deviation and standard error
A researcher calculates a 95% confidence interval for a population mean and finds it to be [45.2, 54.8]. If the researcher wants to reduce the margin of error by half while maintaining the same confidence level, what should happen to the sample size? Explain your reasoning.
The margin of error is given by: \(MOE = Z \left( \frac{\sigma}{\sqrt{n}} \right)\)
To reduce the margin of error by half, we need: \(\frac{MOE}{2} = Z \left( \frac{\sigma}{\sqrt{n'}} \right)\)
Since Z and σ remain constant: \(\frac{1}{2} \cdot \frac{\sigma}{\sqrt{n}} = \frac{\sigma}{\sqrt{n'}}\)
Solving for n': \(\frac{1}{2\sqrt{n}} = \frac{1}{\sqrt{n'}}\)
\(\sqrt{n'} = 2\sqrt{n}\)
\(n' = 4n\)
Therefore, the sample size must be quadrupled to halve the margin of error.
This demonstrates the inverse square root relationship between sample size and margin of error. To achieve greater precision (smaller margin of error), researchers need significantly larger samples. This is because the standard error decreases at the rate of the square root of the sample size. This relationship is fundamental in determining sample sizes for studies.
Margin of Error: The maximum expected difference between the observed sample statistic and the true population parameter
Sample Size: The number of observations in a sample
Precision: The degree to which a measurement is exact
• MOE ∝ 1/√n (margin of error is inversely proportional to square root of n)
• To halve the margin of error, quadruple the sample size
• Larger samples provide more precise estimates
• To double precision, multiply sample size by 4
• To triple precision, multiply sample size by 9
• The relationship follows: New MOE = Original MOE / √(multiplier for n)
• Thinking that doubling the sample size halves the margin of error
• Not accounting for the square root relationship
• Confusing the relationship between sample size and precision
A factory produces light bulbs with a target lifespan of 1000 hours. A sample of 25 bulbs has a mean lifespan of 985 hours with a standard deviation of 30 hours. Calculate a 95% confidence interval for the true mean lifespan. Based on this interval, should the factory adjust its production process? Explain your reasoning.
Since the population standard deviation is unknown and n < 30, we use the t-distribution.
Given:
Step 1: Calculate Standard Error: \(SE = \frac{s}{\sqrt{n}} = \frac{30}{\sqrt{25}} = \frac{30}{5} = 6\)
Step 2: Calculate Margin of Error: \(MOE = t \times SE = 2.064 \times 6 = 12.384\)
Step 3: Calculate Confidence Interval: \(CI = 985 \pm 12.384 = [972.616, 997.384]\)
The 95% confidence interval is [972.62, 997.38] hours.
Since the target of 1000 hours is not within this interval, the factory should consider adjusting its production process as the true mean appears to be below the target.
This demonstrates practical application of confidence intervals in quality control. When the population standard deviation is unknown (common in practice), we use the sample standard deviation and the t-distribution. The decision rule is whether the target value falls within the confidence interval. In this case, since 1000 is not within the interval, there's evidence that the process mean differs from the target.
T-Distribution: Used when population standard deviation is unknown and sample size is small
Degrees of Freedom: n-1 for t-distribution, affecting the shape of the distribution
Quality Control: Statistical methods to monitor and maintain product quality
• Use t-distribution when σ is unknown and n < 30
• For quality control, compare target values to confidence intervals
• If target is outside CI, process adjustment may be needed
• Remember to use t-values for small samples with unknown σ
• Look up t-tables or use software for critical values
• Quality control decisions often involve comparing targets to intervals
• Using Z-distribution instead of t-distribution for small samples with unknown σ
• Not adjusting for degrees of freedom in t-distribution
• Misinterpreting the confidence interval as definitive proof
Which of the following statements about confidence levels and intervals is TRUE?
The answer is C) A 99% confidence interval is wider than a 95% confidence interval. This is because higher confidence levels require larger critical values (Z or t scores), which in turn result in wider intervals. For example, the Z-score for 95% confidence is 1.96, while for 99% confidence it's 2.576. The larger critical value increases the margin of error, making the interval wider.
There's a direct trade-off between confidence level and precision (interval width). To be more confident that our interval captures the true parameter, we must accept a wider range of possible values. This is reflected mathematically in the confidence interval formula where the critical value increases with confidence level, directly widening the interval. Students often expect higher confidence to mean greater precision, which is counterintuitive.
Confidence Level: The probability that the confidence interval will contain the true parameter
Precision: How narrow or wide the confidence interval is
Trade-off: The balance between confidence and precision
• Higher confidence level = wider interval
• Lower confidence level = narrower interval
• There's always a trade-off between confidence and precision
• Think of confidence as certainty and precision as sharpness
• You can't have maximum certainty and maximum precision simultaneously
• Common choices are 90%, 95%, and 99% depending on context
• Believing that higher confidence levels produce narrower intervals
• Confusing confidence level with precision
• Not understanding the trade-off between confidence and precision
Q: When should I use the t-distribution versus the normal distribution for confidence intervals?
A: Use the t-distribution when:
Use the normal (Z) distribution when:
Mathematically, the t-statistic is defined as:
\( t = \frac{\bar{x} - \mu}{s/\sqrt{n}} \)
Where s is the sample standard deviation. As the sample size increases, the t-distribution approaches the normal distribution.
Q: How does sample size affect the width of confidence intervals?
A: Sample size has an inverse relationship with the width of confidence intervals. The margin of error formula shows this relationship:
\( MOE = Z \left( \frac{\sigma}{\sqrt{n}} \right) \)
As sample size (n) increases, the standard error \( \frac{\sigma}{\sqrt{n}} \) decreases, resulting in a narrower confidence interval. Specifically:
This is because the standard error is inversely proportional to the square root of n: \( SE \propto \frac{1}{\sqrt{n}} \).