Sampling Distributions (7.2)

The video introduces the sampling distribution, first by differentiating it from a sample distribution. A sample distribution involves interpreting data from a single sample, while a sampling distribution is built from multiple simple random samples. An example using population height illustrates this difference.

The process of creating a sampling distribution of the sample mean is explained. It involves defining a population, taking random samples of a specific size (n), calculating the mean (x-bar) for each sample, and plotting these x-bar values to form a frequency distribution. Repeating this process many times results in a sampling distribution, which, with enough data, will be normally distributed due to the central limit theorem.

The video compares the characteristics of a population distribution (mean μ, standard deviation σ) and a sampling distribution. While the mean of the sampling distribution (μx-bar) is equal to the population mean (μ), its standard deviation (called the standard error) is smaller. The standard error is calculated as σ / √n, where n is the sample size. This is because averages are less variable than individual observations.

The video presents the standardization formulas for both population and sampling distributions. For a population, the z-score is (x - μ) / σ. For a sampling distribution, the z-score is (x-bar - μ) / (σ / √n).

A quick recap clarifies the three types: population distribution (measures every individual in the population), sample distribution (measures every individual in a single sample), and sampling distribution (repeatedly takes samples, calculates a statistic, and combines them). The main purpose of sampling distributions is convenience and efficiency, allowing inferences about population parameters without measuring every individual, and calculating probabilities for certain outcomes based on sample size.

The first practice problem asks for the probability that the average height of 10 random Canadians is less than 157 centimeters, given the population mean (160 cm) and standard deviation (7 cm). The solution involves calculating the standard error, standardizing the distribution using the sampling distribution formula, and finding the area under the curve using a z-score table.

The second practice question asks for the proportion of all people with heights greater than 170 centimeters. This problem emphasizes using the population distribution formula for standardization, as it refers to 'all people' rather than a sample mean. The area to the right of the z-score is calculated by subtracting the z-table value from 1.