NPC Table, Z-Scores, and the Standard Normal Distribution

"The z-score is the common language of all normal distributions — it translates every raw score into a universal position on the bell curve."

The Table of Areas Under the Normal Probability Curve is the psychometrician's most essential tool. Once you understand how to convert raw scores to z-scores and read the NPC table, you can answer a remarkable range of practical questions: What percentage of students scored above a cutoff? How many individuals fall between two score points? What score corresponds to the 90th percentile?

This section bridges theory and practice, teaching you to use the NPC mathematically.

---

1. The Standard Normal Distribution

The NPC table is based on the Standard Normal Distribution — a special case of the normal distribution where:

• Mean (μ) = 0
• Standard Deviation (σ) = 1

Any normal distribution can be transformed into the standard normal by converting raw scores to z-scores. This standardisation allows us to use a single table for all normal distributions, regardless of the original mean and standard deviation.

---

2. The Z-Score Formula

The fundamental transformation from raw score to standard score is:

$z = \frac{X - M}{\sigma}$

Where:

• z = standard score (z-score)
• X = raw score
• M = mean of the distribution
• σ = standard deviation of the distribution

What Does a Z-Score Mean?

Z-score	Interpretation
z = 0	Score exactly at the mean
z = +1.0	Score is 1 SD above the mean
z = −1.5	Score is 1.5 SD below the mean
z = +2.5	Score is 2.5 SD above the mean (high score)
z = −3.0	Score is 3 SD below the mean (very rare)

Key properties of z-scores:

• The mean of all z-scores = 0
• The standard deviation of z-scores = 1
• Positive z indicates score above the mean (right side of curve)
• Negative z indicates score below the mean (left side of curve)
• Since the NPC is symmetric, we use the absolute value of z when consulting the table

---

3. Understanding the NPC Table

3.1 Table Structure

The NPC table provides the area (as a proportion of 10,000 total) between the mean and a z-value. The table has:

• Rows: The first decimal of z (0.0 through 5.0)
• Columns: The second decimal of z (.00 through .09)

x/σ	.00	.01	.02	.03	.04	.05	.06	.07	.08	.09
0.0	0000	0040	0080	0120	0160	0199	0239	0279	0319	0359
0.5	1915	1950	1985	2019	2054	2088	2123	2157	2190	2224
1.0	3413	3438	3461	3485	3508	3531	3554	3577	3599	3621
1.5	4332	4345	4357	4370	4383	4394	4406	4418	4429	4441
2.0	4772	4778	4783	4788	4793	4798	4803	4808	4812	4817
2.5	4938	4940	4941	4943	4945	4946	4948	4949	4951	4952
3.0	4986.5	…	…	…	…	…	…	…	…	4990.0

3.2 How to Read the Table

Step 1: Calculate z = (X − M) / σ
Step 2: Find the row matching the first decimal of z
Step 3: Find the column matching the second decimal of z
Step 4: Read the area (out of 10,000 = percentage out of 100)

Example: Find area between mean and z = 1.56

• Row: 1.5
• Column: .06
• Value: 4406 → means 44.06% of the distribution lies between the mean and z = 1.56

3.3 Converting Table Values to Percentages

Since the table uses a base of 10,000:

• Table value ÷ 100 = percentage
• Table value 3413 → 34.13%
• Table value 4772 → 47.72%

3.4 Key Reference Values to Memorise

Z-value	Area (Mean to z)	Total within ±z	Why important
±1.00	34.13%	68.26%	1σ rule
±1.28	39.97%	79.94%	80th/20th percentile
±1.64	44.95%	89.90%	90th/10th percentile
±1.96	47.50%	95.00%	95% confidence interval
±2.00	47.72%	95.44%	2σ rule
±2.58	49.51%	99.02%	99% confidence interval
±3.00	49.87%	99.73%	3σ rule

---

4. Rules for Using the NPC Table

These four rules prevent errors when consulting the table:

Rule 1: Always Convert to Z-Scores First

Every raw score must be converted to z using z = (X − M) / σ before consulting the table. The table only works with z-scores on the standard normal distribution.

Rule 2: Mean is Always the Reference Point

All table values represent distance from the mean (which = 0). The table gives area between the mean and the z-value, not between 0 and X.

Rule 3: Convert to Percentages

Table values (in units of 10,000) can be divided by 100 to get percentages. This allows direct interpretation.

Rule 4: Use Absolute Value of Z; Account for Sign Separately

Use the absolute value of z to look up the table value. Then apply the sign:

• Positive z: score is above the mean (right side)
• Negative z: score is below the mean (left side)

---

5. Types of NPC Problems and Solution Strategies

Strategy Map

---

6. Step-by-Step Worked Examples

Example A: Percentage Between Two Scores (Opposite Sides of Mean)

Problem: Adjustment test; N = 500, M = 40, σ = 8. What % of students scored between 36 and 48?

Solution:

1. z₁ = (36 − 40)/8 = −4/8 = −0.50
2. z₂ = (48 − 40)/8 = +8/8 = +1.00
3. Table value for |z| = 0.50: 1915 → 19.15%
4. Table value for |z| = 1.00: 3413 → 34.13%
5. Because scores are on opposite sides of the mean: ADD the areas
6. Total = 19.15 + 34.13 = 53.28%

Interpretation: 53.28% of the 500 students scored between 36 and 48 on the adjustment test.

---

Example B: Number of Cases Between Two Scores

Problem: Reading ability test; N = 200, M = 60, σ = 10. How many students scored between 40 and 70?

Solution:

1. z₁ = (40 − 60)/10 = −2.00
2. z₂ = (70 − 60)/10 = +1.00
3. Table value for |z| = 2.00: 4772 → 47.72%
4. Table value for |z| = 1.00: 3413 → 34.13%
5. Scores on opposite sides of mean → ADD: 47.72 + 34.13 = 81.85%
6. Number of cases = (81.85/100) × 200 = 163.7 ≈ 164 students

---

Example C: Percentage Above or Below a Score

Problem: Intelligence test; N = 500, M = 100, σ = 16. How many students have IQ below 80 and above 120?

Solution:

1. z for X = 80: (80 − 100)/16 = −1.25
2. z for X = 120: (120 − 100)/16 = +1.25
3. Table value for |z| = 1.25: 3944 → 39.44%
4. Cases below 80: 50% − 39.44% = 10.56%
5. Cases above 120: 50% − 39.44% = 10.56% (by symmetry)
6. N below 80 = (10.56/100) × 500 = 52.8 ≈ 53 students
7. N above 120 = 53 students

Key insight: The symmetry of the NPC means the same z-value produces equal tails on both sides.

---

Example D: Finding Score Limits for a Percentage

Problem: Achievement test; N = 75, M = 50, σ = 10. What are the score limits containing the middle 60% of students?

Solution:

1. Middle 60% means 30% on each side of the mean
2. Find z such that table value = 3000 (30%): z = ±0.84
3. Lower limit: X₁ = M + z₁ × σ = 50 + (−0.84)(10) = 50 − 8.4 = 41.6 ≈ 42
4. Upper limit: X₂ = M + z₂ × σ = 50 + (+0.84)(10) = 50 + 8.4 = 58.4 ≈ 58
5. Middle 60% scored between 42 and 58

---

Example E: Percentile Rank

Problem: M = 50, σ = 10. Sumit scored 75. Find his percentile rank.

Solution:

1. z = (75 − 50)/10 = +2.50
2. Table value for z = 2.50: 4938 → 49.38%
3. This is the area above the mean (right side)
4. Total area below 75 = 50% (left half) + 49.38% = 99.38% ≈ 99th percentile

Interpretation: Sumit ranks at the 99th percentile — only about 1% of students scored higher.

---

7. The Z-Score Family: Derived Standard Scores

Z-scores serve as the foundation for several derived scoring systems commonly used in psychology:

Score Type	Transformation	Mean	SD	Used in
Z-score	z = (X−M)/σ	0	1	Raw reference
T-score	T = 10z + 50	50	10	MMPI, personality tests
Deviation IQ	IQ = 15z + 100	100	15	Wechsler, Stanford-Binet
Stanine	S = 2z + 5 (bounded 1–9)	5	2	Educational testing
Sten score	2z + 5.5 (bounded 1–10)	5.5	2	Personality inventories (16PF)
Percentile	Area below score	50th	—	All norm-referenced tests

Indian context: NIMHANS neuropsychological batteries and NCERT achievement tests use T-scores and percentiles for reporting, all derived from z-score transformations.

---

8. Common Student Errors and How to Avoid Them

Error	Correct Approach
Forgetting to convert to z before using table	Always use z = (X−M)/σ first
Adding when should subtract (same-side scores)	Draw a sketch — if both z-values are on the same side of the mean, SUBTRACT areas
Subtracting when should add (opposite-side scores)	If z-values are on opposite sides (one negative, one positive), ADD areas
Using negative z directly in table lookup	Use absolute value of z; then account for direction separately
Forgetting to add 50% for "below a positive score"	Sketch the curve; shade the region you want; count systematically

---

9. External Resources

Wikipedia

• Standard Score — Z-score definition and properties
• Standard Normal Table — Reading the NPC table

Educational Videos

• Khan Academy: Z-Scores — Computing and interpreting z-scores
• StatQuest: Normal Distribution, Clearly Explained — Visual explanation
• Professor Leonard: Using z-Tables — Step-by-step table usage

Research Papers

• Field, A. (2023). Discovering Statistics Using IBM SPSS Statistics. Chapter on normal distribution assumptions. Sage Publications.
• Ghasemi, A., & Zahediasl, S. (2012). Normality tests for statistical analysis: A guide for non-statisticians. International Journal of Endocrinology and Metabolism, 10(2), 486–489.

Additional Resources

• Z-Table Calculator — Interactive z-score to area converter
• Khan Academy: Normal Distributions (full unit)
• Statistics LibreTexts: The Standard Normal Distribution
• Real Statistics: Using the Normal Distribution
• PSYC341 Online Statistics: Z-Scores
• Psychometric Resources: Score Conversions

---

10. Self-Assessment Questions

Q1 (Remember): What is the formula for converting a raw score to a z-score? Define each symbol.

Answer: z = (X − M) / σ, where z = standard score, X = raw score, M = mean of the distribution, σ = standard deviation of the distribution.

Q2 (Understand): A student scores z = −1.5 on a psychology test. What does this mean, and is it a good or poor score?

Answer: A z-score of −1.5 means the student scored 1.5 standard deviations below the mean of the group. This places the student in approximately the bottom 6.7% of the distribution (50% − 43.32% = 6.68%), which is a relatively low score. However, whether it is "poor" depends on context — in a very competitive group, −1.5 SD might still represent a passing grade.

Q3 (Apply): An anxiety test has M = 50 and σ = 8. Priya scored 62. Find her percentile rank.

Answer:

1. z = (62 − 50)/8 = 12/8 = +1.50
2. Table value for z = 1.50: 4332 → 43.32%
3. Priya's percentile rank = 50% + 43.32% = 93.32% ≈ 93rd percentile

Priya scores higher than approximately 93% of the standardisation sample.

Q4 (Apply): If M = 100 and σ = 15 (IQ test), how many students in a class of 1000 would score between 85 and 115?

Answer:

1. z for 85 = (85−100)/15 = −1.00 → table value: 34.13%
2. z for 115 = (115−100)/15 = +1.00 → table value: 34.13%
3. Scores on opposite sides of mean → ADD: 34.13 + 34.13 = 68.26%
4. N = (68.26/100) × 1000 = 682.6 ≈ 683 students

Q5 (Analyse): Why is the mean of z-scores always 0, and the SD always 1? What is the practical value of this?

Answer: Z-scores standardise by subtracting the mean (centering the distribution at 0) and dividing by SD (scaling the spread to 1 unit). The practical value is that z-scores from completely different tests (different Ms and SDs) become directly comparable. A z = +1.5 on an anxiety test and z = +1.5 on an IQ test both represent the same relative standing — 1.5 SDs above the respective means. This enables cross-test comparisons and is the foundation of all normalised scoring systems.

---

11. Memory Aids

Formula Mnemonic: "X Minus Mean, Divided by Sigma"

"eXercise Minus Main effort, Divided by Systematic variation" z = (X − M) / σ

Reading the Table: "Row then Column, Area from Mean"

1. Row = first decimal of z (e.g., 1.5 for z = 1.56)
2. Column = second decimal (e.g., .06 for z = 1.56)
3. Area = value at intersection (in units of 10,000 = %)

Percentage Benchmarks — "28, 44, 50"

• From mean to ±1σ: 34.13 (≈ 34)
• From mean to ±2σ: 47.72 (≈ 48)
• From mean to ±3σ: 49.87 (≈ 50, practically)

---

12. Summary

The NPC Table and Z-score system enable a powerful set of calculations:

• Any raw score is converted to z = (X − M) / σ
• The table gives the area between the mean and any z-value (out of 10,000)
• Four rules govern correct table use: convert to z, use mean as reference, convert to percentages, use absolute z with sign applied separately
• Different problem types require different strategies: adding areas (opposite sides), subtracting areas (same side), or reverse-solving for X given a percentile
• Z-scores are the foundation for all standardised scoring systems (T-scores, IQ scales, stanines, percentiles) used in psychology and education

---

Source PDFs:

• 📄 Block-3/Unit-1.pdf - Pages 12–21
• 📚 MPC-006 Statistics in Psychology, Block 3: Normal Distribution