business stat unit 7 notes

Chapter 7: Sampling and Sampling Distributions

7.1 Why Sample?

  • Population vs. Sample

    • Population: All subjects of interest.

    • Sample: A subset of the population.

  • Reasons for Sampling:

    • Cost Efficiency: Examining the entire population is expensive and time-consuming.

    • Destructive Testing: Some tests destroy the subject, making total examination impractical.

  • Importance of Proper Sampling:

    • Properly selected samples can yield accurate assessments for the entire population.

7.2 Types of Sampling

  • Probability Sampling

    • Definition: Each population member has a known, non-zero chance of selection.

    • Advantage: Supports the performance of inferential statistical tests.

  • Types of Probability Sampling:

    • Simple Random Sampling: Every member has an equal chance.

    • Systematic Sampling: Every kth member is chosen.

    • Stratified Sampling: Population divided into strata; random samples from each stratum.

    • Cluster Sampling: Population divided into clusters; entire clusters are randomly selected.

    • Resampling: Involves drawing multiple samples to improve estimation accuracy.

  • Nonprobability Sampling

    • Definition: Selection probability of members is unknown.

    • Convenience Sampling: Selecting samples for accessibility.

    • Advantages vs. Disadvantages:

      • Quick and easy to obtain.

      • May not represent the population accurately.

7.3 Sampling and Nonsampling Errors

  • Definitions:

    • Parameter: Value that describes a population characteristic.

    • Statistic: Value calculated from sample data.

  • Sampling Error: Difference between the sample statistic and the population parameter.

  • Nonsampling Errors: Caused by issues like ambiguous questions or data collection errors.

7.4 The Central Limit Theorem (CLT)

  • Definition: Sample means of sufficiently large samples will be normally distributed regardless of the population's distribution shape.

  • Conditions for CLT:

    • Sample sizes of 30 or larger typically yield normal distributions.

    • If the population is normally distributed, so are the sample means.

7.5 The Sampling Distribution of the Mean with Finite Population

  • Finite Population Correction: If n/N > 5% and sampling is without replacement, adjustments to the standard error are needed.

  • Example Calculation: When sampling leads to a mean of 7.5 from an average population mean of 7.2, implications about population ratings can be drawn based on probabilities of observed mean.

7.6 The Sampling Distribution of the Proportion

  • Definition: Explores the distribution of sample proportions.

  • Requirements for Normal Approximation: Ensure that np ≥ 5 and n(1-p) ≥ 5.

  • Example Scenario: Evaluating claim of political party support through proportions.

    • Sample of 250 with 130 supporting yields insights into claim validity regarding 60% support.

Key Takeaways

  • Understanding sampling techniques and their implications are crucial in statistics.

  • The validity of claims about populations relies heavily on sampling methods and the application of the Central Limit Theorem.