Estimates and Sample Sizes Notes
Chapter 6: Estimates and Sample Sizes
6-1 Review and Preview
- Inferential statistics involves two main activities:
- Using sample data to estimate population parameters
- Testing hypotheses about population parameters
- This chapter covers methods for estimating important population parameters:
- Proportions
- Means
- Variances
- Methods for determining necessary sample sizes for these estimations are also introduced.
6-2 Estimating a Population Proportion
- Key Concept: Use sample proportion to estimate the value of population proportion.
- The sample proportion is considered the best point estimate.
- Confidence intervals can be constructed to estimate the true value of the population proportion.
- Knowing how to calculate required sample sizes for estimating population proportions is essential.
- Definitions:
- Point Estimate: A single value used to approximate a population parameter (e.g., sample proportion ( \hat{p} ) ).
- Confidence Interval (CI): A range of values estimating the true value of a population parameter.
- Confidence Level: Probability that a CI contains the population parameter (e.g., 90%, 95%, 99%).
Example
- Finding Best Estimate: In a Pew Research Center poll, 70% of 1,501 adults believe in global warming. The best point estimate of ( p ) is 0.70.
Interpreting Confidence Intervals
- CI example interpretation: “We are 95% confident that the interval from 0.677 to 0.723 contains the true population proportion.”
- Critical Value ( z_{\alpha/2} ): Value separating likely and unlikely sample statistics. Available using z-tables for specified confidence levels (e.g., 1.96 for 95%).
- Margin of Error (E): The difference between the sample proportion and the true population proportion, calculated using the formula:
[ E = z_{\alpha/2} \sqrt{\frac{\hat{p}(1 - \hat{p})}{n}} ]
where n is the sample size.
Procedure for Constructing CI for Proportion
- Verify that the required assumptions are satisfied.
- Find the critical value for the desired confidence level.
- Evaluate the margin of error ( E ).
- Compute the confidence interval: ( \hat{p} - E < p < \hat{p} + E ).
6-3 Estimating a Population Mean (σ Known)
- Methods for estimating a population mean when the population standard deviation ( \sigma ) is known.
- Key Concepts:
- Sample mean ( \bar{x} ) is the best point estimate of population mean ( \mu ).
- Confidence intervals can be constructed for the population mean using sample data.
- Necessary sample size determination is crucial for accurate estimation.
- CI for Mean with Known ( \sigma ): ( \mu ) is estimated with
[ \bar{x} - E < \mu < \bar{x} + E ]
where ( E = z_{\alpha/2} \frac{\sigma}{\sqrt{n}} ).
6-4 Estimating a Population Mean (σ Not Known)
- Use the Student t Distribution when the population standard deviation is unknown.
- Key Concept: The t distribution accounts for additional variability expected with small samples.
- Margin of Error with t Distribution:
[ E = t{\alpha/2} \frac{s}{\sqrt{n}} ]
where ( t{\alpha/2} ) corresponds to the degrees of freedom ( n - 1 ). - Construct CI for mean with unknown ( \sigma ):
[ \bar{x} - E < \mu < \bar{x} + E ]
6-5 Estimating a Population Variance
- Chi-Square Distribution is introduced to construct confidence intervals for population standard deviation and variance.
- To estimate the population variance, the chi-square statistic is used:
[ \chi^2 = \frac{(n - 1)s^2}{\sigma^2} ]
where ( s^2 ) is the sample variance.
Confidence Interval for Variance
- Verify that a simple random sample has been taken, and the population is normally distributed.
- Use the critical values from the chi-square table for necessary bounds.
- Evaluate the confidence limits and take square root for standard deviation if needed.