Central Limit Theorem for Sample Means Study Guide

Introduction to the Central Limit Theorem for Sample Means

• The Central Limit Theorem (CLT) provides a framework for understanding what to expect when collecting data from multiple different random samples.
• While previous studies established that a normal model represents proportions for multiple samples, the normal curve is also effective for representing the mean ($\mu$) of quantitative data across several samples.
• A primary application discussed is evaluating the mean SAT scores for various groups of high school seniors.

Variability of Sample Means

• When drawing random samples, the resulting means ($\bar{x}$) will vary because each sample surveys a different set of people.
• Specific examples of varying sample means include:
  • $1030$
  • $1050$
  • $1035$
  • $1068$
• Despite this variability, if certain conditions are met, the distribution of these sample means will follow a normal distribution.

Required Conditions for the Normal Model

• To ensure that sample means are normally distributed, two primary conditions must be verified:
  • Randomization Condition: This requires that all samples collected must be randomized.
  • Large Enough Condition: The sample size ($n$) must be sufficiently large. While there is no specific numerical cutoff defined, the size should be enough to allow the histogram of means to form a normal distribution.
• If these conditions are satisfied, a histogram of the collected sample means will show a normal distribution with the actual population mean located at the center.

Characteristics of the Normal Model for Means

• In the normal model for the mean, the actual population mean ($\mu$) is positioned at the center.
• Most sample means will fall within three standard deviations ($SD$) above or below this central population mean.
• The Empirical Rule (68-95-99.7 Rule):
  • $68\%$ of sample means lie between one standard deviation below and one standard deviation above the population mean ($\pm 1\,SD$).
  • $95\%$ of sample means lie between two standard deviations below and two standard deviations above the population mean ($\pm 2\,SD$).
  • $99.7\%$ of sample means lie between three standard deviations below and three standard deviations above the population mean ($\pm 3\,SD$).

Calculations for the Normal Model

• Problems typically provide the population mean ($\mu$) and the population standard deviation ($\sigma$).
• Model Mean: The mean of the normal model is identical to the population mean provided.
• Model Standard Deviation Calculation: To find the standard deviation specifically for the normal model, divide the population standard deviation by the square root of the sample size.
• Formula:$\text{Standard deviation for the normal model} = \frac{\sigma}{\sqrt{n}}$

Case Study: College Board SAT Scores

• Population Data:
  • Mean SAT score ($\mu$): $1051$
  • Population Standard Deviation ($\sigma$): $211$
• Sample Data:
  • Random sample size ($n$): $400$ college applicants.
  • Sample mean ($\bar{x}$): $1100$
• Objective: Determine if a sample mean of $1100$ is an unusually high average.
• Condition Check: The sample was randomly collected, and a sample size of $400$ is assumed to be large enough.
• Standard Deviation Calculation:$\text{Standard deviation for the normal model} = \frac{211}{\sqrt{400}} = \frac{211}{20} = 10.55$
• Determining the Normal Curve Ranges:
  • Mean (Center): $1051$
  • $+1\,SD$: $1051 + 10.55 = 1061.55$
  • $+2\,SD$: $1061.55 + 10.55 = 1072.1$
  • $+3\,SD$: $1072.1 + 10.55 = 1082.65$
  • $-1\,SD$: $1051 - 10.55 = 1040.45$
  • $-2\,SD$: $1040.45 - 10.55 = 1029.9$
  • $-3\,SD$: $1029.9 - 10.55 = 1019.35$
• Evaluation:
  • The sample mean of $1100$ lies beyond the $+3\,SD$ marker ($1082.65$).
  • Since $99.7\%$ of all sample means are within three standard deviations, a mean of $1100$ is more than three standard deviations above the mean and is therefore considered unusually high.

Comparison: Individual Data vs. Sample Means

• Distribution of Data (Single Sample): This model looks at individual data points, such as individual SAT scores. These can range from very low to very high, resulting in a wide spread.
• Distribution of Means (Multiple Samples): This model uses the averages calculated from large random samples. This process yields a much smaller spread of data compared to individual scores, as the means of large samples tend to be more consistent.

Summary of Key Findings

• The normal model represents the distribution of means across several samples.
• Critical requirements are the Randomization condition and the Large enough condition.
• The actual population mean is centered in the normal model.
• The standard deviation for the normal model is determined by:     $\text{Standard deviation for the normal model} = \frac{\text{Standard deviation of the population}}{\sqrt{\text{sample size}}}$