Chapter 7: Confidence Interval Estimates and Population Proportions

Foundations of Inferential Statistics

• Definition of Inferential Statistics:

  • Inferential statistics involves using sample data to make generalizations, conclusions, and formal decisions about an entire population.
  • It bridges the gap between observed sample metrics and unknown population characteristics.

• Primary Purposes of Inferential Statistics:

  • Estimating Population Parameters: Using sample statistics to approximate true, unknown population values, such as population proportions ($p$) or population means ($\mu$).
  • Testing Hypotheses: Formally evaluating claims or hypotheses regarding the values of population parameters using observed sample evidence.

• Point Estimates:

  • Definition: A point estimate is a single numerical value utilized to approximate or estimate a population parameter.

Estimating a Population Proportion

• The Sample Proportion as an Estimator:

  • The sample proportion, denoted by $\hat{p}$ ("p-hat"), is calculated using the formula:         $\hat{p} = \frac{x}{n}$         where $x$ is the number of successes (or individuals exhibiting a given trait) and $n$ is the total sample size.
  • The sample proportion $\hat{p}$ is considered the best point estimate of the population proportion $p$.

• Limitations of Point Estimates:

  • A point estimate provides no measurement or indication of its accuracy or precision.
  • It gives no inherent indication of how close the sample proportion $\hat{p}$ actually is to the true population proportion $p$.

• Applied Example:

  • Consider investigating what proportion of NGU students have a tattoo, where $p$ denotes the true proportion of all NGU students with a tattoo.
  • The calculated sample proportion $\hat{p}$ serves as the point estimate for $p$, but does not reveal how far the estimate may deviate from the actual population value.

Confidence Intervals and Confidence Levels

• Confidence Interval (CI) Estimate:

  • Definition: A confidence interval estimate is a range of values used to estimate the true value of an unknown population parameter.
  • Function: Instead of assigning a single value, a confidence interval establishes an interval of numbers bounded by a lower limit and an upper limit to account for sampling variability.

• Confidence Level:

  • Definition: The confidence level is the probability that the confidence interval actually contains the true value of the population parameter.
  • Common Expressions: Often expressed as a percentage or decimal probability, such as $0.95$ or $95\%$.

The Significance Level ($\alpha$) and Critical Values

• The Role of Alpha ($\alpha$):

  • The Greek letter alpha ($\alpha$) denotes the complement or opposite of the confidence level.
  • Mathematical Relationship:$\alpha = 1 - \text{Confidence Level}$
  • Example Calculation: For a $90\%$ confidence level:         $\alpha = 1 - 0.90 = 0.10$         which is equivalent to $10\%$.

• Critical $z$ Values:

  • Definition: Critical $z$ values (often denoted as $z_{\alpha/2}$) are standard scores ($z$-scores) positioned on the borderlines of the standard normal distribution.
  • Separation of Regions: These values separate sample statistics ($z$-scores) that are significantly high or significantly low from those that are not statistically significant.
  • Distributional Characteristics: The critical boundary values isolate an area of $\frac{\alpha}{2}$ in each of the two tails of the standard normal distribution, leaving the central confidence level area equal to $1 - \alpha$.