Analyzing SAT Score Percentiles Using the Normal Model and the Empirical Rule

Standard SAT Statistics: * Mean ( $\mu$ ): The overall mean for all test takers is approximately $500$ . * Standard Deviation ( $\sigma$ ): The overall standard deviation is approximately $100$ . * Variability: While the mean and standard deviation may differ from these target values in any specific year by small amounts, these figures serve as a reliable overall approximation for analysis.

The Question: Suppose a student earns a score of $600$ on one part of the SAT. The goal is to determine the student's standing (percentile) among all other test takers.
The Problem-Solving Process (Think, Show, Tell): * Step 1: Think (Planning): * Objective: State clearly what is sought. The objective is to determine how a specific SAT score compares with the scores of all other students. * Modeling Requirement: To accomplish this comparison, the distribution of scores must be modeled. * Identifying Variables: Let $y$ represent the variable of interest, which is the SAT score. * Data Type: SAT scores are classified as quantitative data. They do not have meaningful units other than "points."

Nearly Normal Condition: Before applying a normal model, this condition must be checked. * Data-Driven Check: If raw data were available, one would inspect a histogram to verify the shape of the distribution. * Assumed Distribution: In this scenario, while raw data is absent, it is explicitly stated that SAT scores are "roughly unimodal and symmetric," justifying the use of the normal model.
Model Parameters: * The model used is a normal distribution with specified parameters: a mean of $500$ and a standard deviation of $100$ . * This is denoted notationally as $N(500, 100)$ .

Step 2: Show (Mechanics): This phase involves the mathematical calculations and visual representations necessary to reach a conclusion.
Z-score Calculation: * The z-score represents how many standard deviations a value is from the mean. * Formula applied: $z = \frac{y - \mu}{\sigma}$ * Calculation for a score of $600$ : $z = \frac{600 - 500}{100} = 1$ * This result indicates the score of $600$ is exactly $1$ standard deviation above the mean.
Normal Model Visualization: * A simple sketch of the normal curve is used to map the distribution. * The 68-95-99.7 (Empirical) Rule: This rule dictates the distribution of data points within standard deviations: * Within 1 Standard Deviation (Blue): Approximately $68\%$ of the values fall within one standard deviation of the mean ( $400$ to $600$ ). * Within 2 Standard Deviations (Green): Approximately $95\%$ of the values fall within two standard deviations of the mean ( $300$ to $700$ ). * Within 3 Standard Deviations (Red): Approximately $99.7\%$ of the values fall within three standard deviations of the mean ( $200$ to $800$ ). * Individual Placement: By locating the score of $600$ on the sketch, it is confirmed to be exactly at the boundary of the first standard deviation above the mean.

Step 3: Tell (Evaluation): The final step involves interpreting the mathematical findings within the context of the original question.
Calculating the Upper Tail: * Because the normal distribution is symmetric, and $68\%$ of scores are within one standard deviation ( $\pm 1\,\sigma$ ) of the mean, the remaining scores comprise $32\%$ of the population ( $100\% - 68\% = 32\%$ ). * These remaining scores are split equally between the lower tail (more than $1$ SD below the mean) and the upper tail (more than $1$ SD above the mean). * Therefore, the percentage of students scoring higher than $1$ standard deviation above the mean is half of $32\%$ , which is $16\%$ .
Determining Percentile: * If approximately $16\%$ of test takers scored better than $600$ , then the remaining portion of the population scored at or below that value. * Calculation: $100\% - 16\% = 84\%$ * Final Contextual Result: A score of $600$ on the SAT is higher than approximately $84\%$ of all scores on the test.