Standardized Test Z-Score and Percentile Problem

What This Problem Teaches

• Converting raw scores to standardized z-scores using the formula z = (x - μ) / σ
• Using the standard normal distribution table to find cumulative probabilities
• Understanding that "percent above" means finding the upper tail area (1 - cumulative probability)
• Interpreting z-scores as measures of relative performance within a population
• Connecting abstract statistical concepts to real educational assessment scenarios

Visualizing the Distribution

This normal curve shows the distribution of test scores. The shaded red area represents the 6.3% of students who scored higher than 148.

Solution: Method 1 — Z-Score and Standard Normal Table

Step 1 — Calculate the z-score

The z-score formula standardizes any normal distribution to the standard normal distribution (mean = 0, standard deviation = 1):

Where x = 148 (student's score), μ = 125 (population mean), and σ = 15 (population standard deviation):

Step 2 — Interpret the z-score

A z-score of 1.53 means the student scored 1.53 standard deviations above the mean. This is a strong performance — most students score within 2 standard deviations of the mean.

Step 3 — Look up the z-score in the standard normal table

The standard normal table gives us the area to the left of (below) any z-score. For z = 1.53:

This means 93.70% of students scored at or below 148.

Step 4 — Find the area to the right (above)

We want the percentage who scored higher than 148, which is the area to the right of z = 1.53:

Step 5 — Convert to percentage

Converting the probability to a percentage:

Solution: Method 2 — Empirical Rule Estimation

Step 1 — Apply the empirical rule as a starting point

The empirical rule tells us that in a normal distribution:

• About 68% of data falls within 1 standard deviation of the mean (110 to 140)
• About 95% falls within 2 standard deviations (95 to 155)
• About 99.7% falls within 3 standard deviations (80 to 170)

Step 2 — Locate the score within these ranges

The score of 148 is between +1σ (140) and +2σ (155). Since 95% of scores fall between -2σ and +2σ, and the distribution is symmetric, 2.5% of scores are above +2σ (155).

Step 3 — Estimate using the known boundaries

We know:

• 16% of scores are above +1σ (140)
• 2.5% of scores are above +2σ (155)

Since 148 is about halfway between 140 and 155, our answer should be roughly halfway between 16% and 2.5%, or around 9%.

Step 4 — Refine with the exact z-score

For precision, we calculate z = (148 - 125) / 15 = 1.53 and use the standard normal table to get 6.30%, which confirms our estimate was in the right ballpark.

Verification

Let's verify our calculation by checking the z-score and table lookup:

Check the z-score calculation

Check the table lookup

For z = 1.53, the cumulative probability is indeed 0.9370, meaning 93.70% scored at or below 148. Therefore, 100% - 93.70% = 6.30% scored higher. ✓

Sanity check

The score of 148 is significantly above average (125), so it makes sense that only a small percentage scored higher. Our answer of 6.30% seems reasonable.

Common Mistakes to Avoid

The Pattern Behind This

This problem follows the standard template for "upper tail probability" questions in normal distributions:

Where Φ represents the cumulative distribution function of the standard normal distribution.

The key insight is that any normal distribution can be transformed to the standard normal distribution through z-scores, allowing us to use the same table for all normally distributed data. This is why standardized test scores are so useful — they provide a common scale for comparing performance across different tests and populations.

Why This Matters in Education

Understanding percentiles and z-scores is crucial in educational assessment because:

• Standardized testing: SAT, ACT, state achievement tests all report both raw scores and percentiles
• Special education: Eligibility often requires performance below the 16th percentile (z ≈ -1.0)
• Gifted programs: Entry might require performance above the 95th percentile (z ≈ +1.65)
• Research: Educational researchers use these techniques to compare interventions and measure achievement gaps

What If?