Unit 10: Estimating Means with Confidence

The Unique Properties of the t-Distribution: Students will understand that when the population standard deviation is unknown, they must use a t-distribution, which is shorter and has thicker tails than the standard Normal curve to account for extra uncertainty.

The Concept of Degrees of Freedom: Knowledge that the specific shape of the t-distribution depends on the sample size and that as the sample size increases, the t-distribution becomes nearly identical to the Normal distribution.

The Difference Between Standard Deviation and Standard Error: The understanding that standard error is an estimate of the variability of the sample mean based on the data actually collected.

The Requirements for Independence: The logic behind the Ten Percent Rule and why it is necessary to ensure that observations within a sample do not influence one another. The Importance of Robustness: Knowing that t-procedures are generally reliable even if the population is not perfectly bell-shaped, provided there are no extreme outliers.

Identify the Correct Procedure: Students will distinguish between a one-sample interval for a single average and a two-sample interval for the difference between two independent averages.

Verify Conditions for Inference: For every problem, they will check for Randomness, the Ten Percent Rule, and the Normal or Large Sample condition.

Graph and Analyze Small Datasets: When sample sizes are small, students will create and inspect dotplots or boxplots to ensure the data does not show extreme skewness or outliers.

Calculate Intervals Using Technology: They will use calculators or statistical tables to find critical values and determine the lower and upper bounds of their estimate.

Assess the plausibility of parameter estimates: Use an interval to determine whether or not a claim about a parameter is supported.

Students will demonstrate their mastery by writing formal interpretations of confidence intervals that correctly identify the population parameter and the level of uncertainty in the context of the problem. They will be able to explain how changing the confidence level or the sample size will impact the width and precision of their estimate. In a comparative setting, they will use a calculated interval to determine if it is plausible that two populations have the same average by checking if a difference of zero is included within their findings. Finally, they will justify the use of specific statistical models by providing a written defense of the conditions they checked, particularly when dealing with small or non-Normal datasets.