Estimating Population Proportions and Determining Sample Sizes (Chapter 7.1 Study Notes)

Estimating a Population Proportion

Introduction

Estimating a population proportion is a fundamental concept in inferential statistics. It involves using sample data to make inferences about the proportion of a population that possesses a certain characteristic. This process typically includes constructing a confidence interval to express the uncertainty associated with the estimate and determining the sample size required for a desired level of accuracy.

Learning Objectives

• Construct a confidence interval estimate of a population proportion and interpret such estimates.

• Identify the requirements necessary for valid confidence interval procedures.

• Determine the sample size necessary to estimate a population proportion with a specified margin of error.

Key Concepts in Estimating Population Proportion

Definitions

• Population Proportion (p): The fraction of the entire population that has a particular attribute.

• Sample Proportion (\( \hat{p} \)): The fraction of the sample that has the attribute, used as a point estimate for the population proportion.

• Confidence Interval: A range of values, derived from the sample, that is likely to contain the true population proportion.

• Margin of Error (E): The maximum expected difference between the true population parameter and a sample estimate.

Example: STLCC "Traditional Aged" Students

Suppose the traditional college age is defined as 18 to 24 years old. The following example demonstrates how to estimate the proportion of traditional-aged students at STLCC using sample data.

• What percent of students at STLCC would you expect to be traditional college age?

• How many people are traditional aged in your group?

• Is your percent in your group the same as your prediction?

• Does this indicate you are correct or incorrect?

Tabular Data: STLCC Student Age Distribution (Fall 2024)

The following table summarizes the age distribution of students at STLCC in Fall 2024. This data is used to estimate the proportion of students aged 18-24.

Age Group	Total	Men	Women
All Students	15,649	5,068	10,581
Under 18	2,613	931	1,682
18-19	2,613	947	1,666
20-21	1,895	561	1,334
22-24	2,273	867	1,406
25-29	1,869	450	1,419
30-39	754	176	578
40-49	216	61	155
50-64	49	27	22
65 and over	67	27	40
Age Unknown/unreported	0	0	0

Additional info: The total number of students aged 18-24 is the sum of the 18-19, 20-21, and 22-24 age groups: 2,613 + 1,895 + 2,273 = 6,781. However, the notes use 7,804, which may include some additional students or a different grouping. For calculation, use the provided total.

Calculating the Sample Proportion

• Number of students aged 18-24: 7,804

• Total number of students: 15,649

• Sample proportion: \( \hat{p} = \frac{7,804}{15,649} \approx 0.499 \)

Constructing a Confidence Interval for a Population Proportion

A confidence interval provides a range of plausible values for the population proportion based on sample data.

• Point Estimate: The sample proportion \( \hat{p} \) is the best point estimate for the population proportion \( p \).

• Margin of Error (E): The maximum likely error in the estimate.

Formulas:

• Point estimate of \( p \): $\hat{p} = \frac{\text{upper confidence interval limit} + \text{lower confidence interval limit}}{2}$

• Margin of error: $E = \frac{\text{upper confidence interval limit} - \text{lower confidence interval limit}}{2}$

• General confidence interval for \( p \): $\hat{p} - E < p < \hat{p} + E$

• Alternate formats: $\hat{p} \pm E$   or   $(\hat{p} - E, \hat{p} + E)$

Example: Confidence Interval Calculation

• Lower limit: 0.4908562

• Upper limit: 0.50652383

• Confidence level: 95%

• Interval: (0.491, 0.507) (rounded to three decimal places)

Interpretation: We are 95% confident that the true proportion of traditional-aged students at STLCC is between 0.491 and 0.507.

Correct and Incorrect Interpretations of Confidence Intervals

• Correct: "We are 95% confident that the interval from 0.491 to 0.507 actually does contain the true value of the population proportion \( p \)."

• Incorrect: "There is a 95% chance that the true value of \( p \) will fall between 0.491 and 0.507."

• Incorrect: "95% of sample proportions will fall between 0.491 and 0.507."

Additional info: The confidence interval refers to the process, not the probability for a specific interval.

The Process Success Rate

A 95% confidence level means that, in the long run, 95% of confidence intervals constructed from repeated samples will contain the true population proportion.

Requirements for Constructing a Confidence Interval for a Proportion

• The sample is a simple random sample.

• The conditions for the binomial distribution are satisfied:

  • Fixed number of trials

  • Independent trials

  • Two categories of outcomes

  • Constant probability for each trial

• There are at least 5 successes and 5 failures in the sample.

Determining Sample Size for Estimating a Population Proportion

Key Considerations

• Confidence Level: Commonly 90%, 95%, or 99%

• Margin of Error (E): Desired maximum error

• Target Proportion: Use a previous sample estimate or assume 0.5 if unknown

In StatCrunch, you must enter the confidence level, target proportion, and width (which is double the margin of error, or \( 2E \)).

Sample Size Calculation Example

• Confidence level: 95%

• Margin of error: 1% (0.01)

• Target proportion: 0.499 (from sample)

• Required sample size: 9,604 students

Additional info: The sample size increases as the desired margin of error decreases or the confidence level increases.

Using StatCrunch for Confidence Intervals and Sample Size

Steps for Confidence Interval Calculation

1. Go to Stat → Proportion Stats → One Sample → With Summary

2. Enter the number of successes and total observations

3. Select Confidence interval for p and set the confidence level

4. Choose the Standard-Wald method

5. Click Compute to obtain the interval

Steps for Sample Size Calculation

1. Go to Stat → Proportion Stats → One Sample → Width/Sample Size

2. Enter the confidence level, target proportion, and desired width

3. Click Compute to obtain the required sample size

Summary Table: Confidence Interval Components

Component	Description
Sample Proportion (\( \hat{p} \))	Estimate of population proportion from sample
Margin of Error (E)	Maximum likely error in estimate
Confidence Interval	Range: \( \hat{p} - E < p < \hat{p} + E \)
Confidence Level	Probability that the interval contains the true proportion (e.g., 95%)
Sample Size (n)	Number of observations required for desired accuracy

Conclusion

Estimating a population proportion using confidence intervals is a key statistical skill. It requires understanding the underlying assumptions, correctly interpreting the interval, and determining the appropriate sample size for reliable results. Tools like StatCrunch facilitate these calculations, but a solid grasp of the concepts ensures accurate and meaningful statistical inference.