Skip to main content
Indietro

Confidence Intervals in Introductory Statistics: Mean, Proportion, Variance, and Standard Deviation

Guida di studio - Note intelligenti

Appunti personalizzati basati sui tuoi materiali, ampliati con definizioni chiave, esempi e contesto.

Confidence Intervals

Introduction to Confidence Intervals

Confidence intervals are fundamental tools in statistics for estimating population parameters based on sample data. They provide a range of plausible values for the parameter, along with a specified level of confidence.

  • Point Estimate: A single value estimate for a population parameter, such as the sample mean (x̅) for the population mean (μ).

  • Interval Estimate: A range of values used to estimate a population parameter, reflecting uncertainty due to sampling variability.

  • Level of Confidence (c): The probability that the interval estimate contains the population parameter, assuming repeated sampling.

Critical values and confidence level on a normal curve

Critical Values and Confidence Levels

Critical values separate probable from improbable sample statistics. For a given confidence level, the critical value (zc or tc) determines the endpoints of the confidence interval.

  • Standard Normal Distribution: Used when the population standard deviation (σ) is known.

  • Common Confidence Levels: 90% (zc = 1.645), 95% (zc = 1.96), 99% (zc = 2.575).

90% confidence interval on a normal curve95% confidence interval on a normal curve

Confidence Intervals for the Mean (μ): σ Known

Constructing Confidence Intervals

When the population standard deviation is known, the confidence interval for the mean is constructed using the standard normal distribution.

  • Margin of Error (E):

  • Confidence Interval:

  • Conditions: Random sample, population is normally distributed or .

Example: Athletic Activities

A sample of 40 student-athletes yields a sample mean of 19.6 hours. With σ = 1.2 and a 95% confidence level:

  • Margin of error:

  • Confidence interval: or

Confidence interval for mean number of hours

Using Technology

Statistical software (e.g., Minitab, StatCrunch, TI-84 Plus) can compute confidence intervals efficiently.

Minitab and StatCrunch output for confidence interval

Interpretation

The confidence interval does not state the probability that μ is in the interval, but rather that, in repeated sampling, the interval will contain μ in approximately c% of cases.

Multiple confidence intervals and population mean

Minimum Sample Size

To achieve a desired margin of error E at confidence level c:

  • Round up to the next whole number.

Confidence Intervals for the Mean (μ): σ Unknown

The t-Distribution

When σ is unknown, use the t-distribution, which accounts for additional uncertainty from estimating σ with the sample standard deviation (s).

  • Degrees of Freedom (d.f.):

  • Critical Value (tc): Found using t-tables or technology.

  • Margin of Error:

  • Confidence Interval:

t-distribution curves with different degrees of freedomt-table for critical valuest-distribution with critical valuesTI-84 Plus output for t critical value

Example: Coffee Shop Temperatures

Sample of 16 coffee shops, sample mean 162.0°F, s = 10.0, 95% confidence:

  • Margin of error:

  • Confidence interval: or

Confidence interval for coffee temperature

Example: Car Dealership Days

Sample of 36 cars, sample mean 9.75 days, s = 2.39, 99% confidence:

  • Margin of error:

  • Confidence interval:

Confidence interval for days on lotStatCrunch output for confidence interval

Choosing Between Normal and t-Distribution

Use the standard normal distribution if σ is known and the population is normal or . Use the t-distribution if σ is unknown.

Flowchart for choosing normal or t-distribution

Confidence Intervals for Population Proportions (p)

Point Estimate for Proportion

The sample proportion (p̂) is the best estimate for the population proportion (p).

  • , where x = number of successes, n = sample size

  • , proportion of failures

Constructing Confidence Intervals for Proportion

  • Conditions: , (normal approximation)

  • Margin of Error:

  • Confidence Interval:

Example: Summer Travel

902 out of 1100 adults plan to travel. , , 95% confidence:

  • Margin of error:

  • Confidence interval:

Confidence interval for proportion of travelersStatCrunch output for confidence interval

Example: News Preferences

48% of 800 adults prefer news websites/social media. , , 99% confidence:

  • Margin of error:

  • Confidence interval:

Bar chart of news preferencesConfidence interval for news preference proportionTI-84 Plus output for proportion confidence interval

Minimum Sample Size for Proportion

  • If no preliminary estimate, use

Political campaign illustration

Confidence Intervals for Variance (σ2) and Standard Deviation (σ)

The Chi-Square Distribution

Used to construct confidence intervals for variance and standard deviation when the population is normally distributed.

  • Degrees of Freedom:

  • Critical Values: (left-tail), (right-tail)

Chi-square distribution curvesChi-square critical values illustrationChi-square distribution with confidence area

Constructing Confidence Intervals

  • Variance:

  • Standard Deviation:

Chi-square table for critical valuesChi-square distribution with critical values

Example: Allergy Medicine Weights

Sample of 30, s = 1.20 mg, 99% confidence:

  • Variance interval:

  • Standard deviation interval:

Minitab output for variance and standard deviation confidence intervals

Summary Table: Confidence Interval Formulas

Parameter

Distribution

Formula

Mean (μ), σ known

Normal (z)

Mean (μ), σ unknown

t-distribution

Proportion (p)

Normal (z)

Variance (σ2)

Chi-square

Standard Deviation (σ)

Chi-square

Pearson Logo

Study Prep