Standard Normal vs Normal Distribution
A normal distribution can have any mean and spread. The standard normal has a mean of 0 and a standard deviation of 1. Enter your own numbers and watch both curves give the same probability.
Normal to Standard Normal Converter
Set the mean and standard deviation, pick what you know, and compare the two curves.
Your normal distribution
Standard normal, N(0, 1)
Working appears here.Summary appears here.Your own normal curve and the standard normal are drawn together, with the same area shaded on each.
Go from a value to a z-score, from a z-score back to your units, or across an interval.
The two curve heights are shown, so you can watch f(x) = φ(z) / σ at work.
What Is the Difference Between a Normal and a Standard Normal Distribution?
A normal distribution is a bell-shaped curve that can have any mean and any standard deviation. The standard normal distribution is the one member of that family with a mean of 0 and a standard deviation of 1. Every normal distribution can be converted into it with the z-score formula.
Think of the normal distribution as a family and the standard normal as its reference member. Heights, delivery times and test scores each have their own mean and spread, so each gets its own curve. The standard normal has no units at all. Its horizontal axis counts standard deviations from the mean, which is why it works for every problem.
| Feature | Normal distribution | Standard normal |
|---|---|---|
| Mean | Any value, μ | Exactly 0 |
| Standard deviation | Any positive value, σ | Exactly 1 |
| Notation | X ~ N(μ, σ²) | Z ~ N(0, 1) |
| Horizontal axis | Original units (inches, minutes, points) | z-scores, no units |
| Peak height | 1 / (σ√(2π)) | 0.3989 |
| Probability tables | A different table for every μ and σ | One table covers all |
The tool above shows the link directly. Enter any mean and standard deviation, and it draws your normal curve and the standard normal curve side by side, with the same area shaded on both.
How Do You Use This Standardizer?
Enter the mean and standard deviation of your normal distribution, choose what you know, and the tool converts between the two curves.
- Find z from x. Type a value x. You get its z-score, the area to its left and right, and the position on both curves.
- Find x from z. Type a z-score. You get the matching value in your original units, with the same areas.
- Probability between two values. Type a lower and an upper value. You get both z-scores and the probability that X falls between them.
The quick buttons load the three worked examples from this page. The top chart is your normal curve, the bottom chart is the standard normal, and the shaded areas always match. That match is the whole idea of standardizing.
What Is a Normal Distribution?
A normal distribution is a symmetric, bell-shaped probability distribution described by two numbers: the mean μ, which sets the center, and the standard deviation σ, which sets the spread. Values near the mean are common, and values far from it become rare quickly.
Many measurements land close to this shape, such as adult heights, small measurement errors and the weights of filled packages. The notation X ~ N(μ, σ²) says that X follows a normal distribution with mean μ and variance σ². Some books write the second number as σ instead of σ², so check which one your course uses before you read N(100, 225) as a standard deviation of 225.
Changing μ slides the whole curve left or right without changing its shape. Changing σ stretches or squeezes it. A bigger σ gives a wider and lower curve, a smaller σ gives a narrower and taller one. There is no single normal distribution, only infinitely many, one for each pair of μ and σ. The Gaussian distribution calculator handles any of them.
What Is the Standard Normal Distribution?
The standard normal distribution is the normal distribution with a mean of 0 and a standard deviation of 1. Its variable is usually written Z, and its values are called z-scores. It is also known as the z-distribution.
The density at a point z is φ(z) = e−z²/2 / √(2π), and the area to the left of z is written Φ(z). The curve peaks at 0.3989, sits symmetric around zero, and gives 68.27% of its area between −1 and 1. Printed z tables list Φ(z) for each z. A z table is that list, and the z-score to percentile calculator reads it for you.
Because the mean is 0 and the spread is 1, a z-score has a plain meaning. A z of 1.5 is one and a half standard deviations above the mean. A z of −2 is two below. No units remain, which is why the same table serves heights in inches and delivery times in minutes.
How Do You Turn a Normal Distribution Into a Standard Normal?
Subtract the mean from each value, then divide by the standard deviation: z = (x − μ) / σ. This is called standardizing. It moves the center to 0 and rescales the spread to 1, and it does nothing else to the shape.
Think of it as two steps. Subtracting μ shifts the curve so its center sits at zero. Dividing by σ shrinks or stretches the horizontal axis so one standard deviation becomes one unit. The second formula runs the same trip backwards, from a z-score to a value in original units.
Three things stay the same. The shape is still a bell, the order of the values is unchanged, and every probability is unchanged. The probability that X is below a value equals the probability that Z is below that value's z-score. The things that change are the units, the mean, the standard deviation, and the height of the curve, which the next section explains.
Standardizing does not make data normal. If the original data are skewed, the z-scores are skewed in exactly the same way. It only works as a shortcut to a probability when X really is normal.
Worked Example 1: How Likely Is a Height Below 66 Inches?
About 78.8%. Suppose adult heights in a group are normal with a mean of 64 inches and a standard deviation of 2.5 inches. What share is shorter than 66 inches?
You can also ask for the other side. For heights above 68 inches, z = (68 − 64) / 2.5 = 1.60, and P(X > 68) = 1 − 0.9452 = 0.0548, so about 1 person in 18. Press the first quick button above to see both curves with this area shaded.
Convert the heights to centimeters and the answer does not move. The mean becomes 162.56, the standard deviation becomes 6.35, and 66 inches becomes 167.64 cm. The z-score is still 0.80 and the probability is still 0.7881. That is the point of a unit-free scale.
Worked Example 2: Which Delivery Time Marks the Slowest 5%?
About 41.9 minutes. Delivery times are normal with a mean of 32 minutes and a standard deviation of 6 minutes. Find the time that only 5% of deliveries exceed.
This runs the standardizing formula in reverse. You look up the z-score for the area first, then convert it back to minutes. The inverse normal distribution calculator does the first step directly if you only need the z-score.
Worked Example 3: What Share of Flour Bags Falls Between 490 and 510 Grams?
About 78.9%. A machine fills bags to a mean of 500 grams with a standard deviation of 8 grams. Both limits sit 10 grams from the mean.
The two z-scores are mirror images, so the answer is symmetric. About 21.1% of bags fall outside the range, 10.6% on each side. For a wider look at between-value problems on any curve, the normal CDF calculator takes the same inputs.
What Does the 68-95-99.7 Rule Look Like on Both Curves?
The rule says that about 68% of a normal distribution lies within 1 standard deviation of the mean, 95% within 2 and 99.7% within 3. On the standard normal that is the interval from −1 to 1, −2 to 2 and −3 to 3. On any other normal it is μ ± σ, μ ± 2σ and μ ± 3σ.
| Width | Standard normal | Heights, μ = 64, σ = 2.5 | Share inside |
|---|---|---|---|
| ± 1σ | −1 to 1 | 61.5 to 66.5 | 68.27% |
| ± 2σ | −2 to 2 | 59.0 to 69.0 | 95.45% |
| ± 3σ | −3 to 3 | 56.5 to 71.5 | 99.73% |
The percentages are identical in every row because standardizing preserves areas. Only the endpoints change, from z-scores to real heights.
Why Are the Areas Equal but the Curve Heights Different?
Standardizing keeps areas but changes heights. The height of a normal curve at x equals the standard normal height at the matching z, divided by σ: f(x) = φ(z) / σ. Heights are densities, not probabilities, so they depend on the units.
Take the height example with μ = 64 and σ = 2.5. At x = 66, z = 0.80 and φ(0.80) = 0.2897. The normal curve's height there is 0.2897 / 2.5 = 0.1159. The normal curve is taller than it looks per inch, because inches are small units and a lot of probability is packed into a narrow range. Peak heights follow the same rule: 0.3989 for the standard normal, 0.1596 for σ = 2.5, and 0.0266 for σ = 15.
The areas under both curves are 1, and the area to the left of 66 on one equals the area to the left of 0.80 on the other. The tool shows the density of both curves at your point, so you can watch the division by σ happen.
How Do You Use a Z Table for Any Normal Distribution?
Standardize your value, then read the area to the left of the z-score. A standard normal table has a row for the first decimal of z and a column for the second, and the cell gives the area. Every other normal problem reduces to that lookup.
- Write down μ, σ and the value x.
- Compute z = (x − μ) / σ and round to two decimals.
- Read Φ(z) from the table for the area to the left.
- For the area to the right, take 1 − Φ(z). For an area between two values, subtract the two left-tail areas.
With a mean of 64, a deviation of 2.5 and x = 66, step 2 gives z = 0.80, and the table cell in row 0.8 and column 0.00 reads 0.7881. If z is negative, either read the negative table or use symmetry: Φ(−z) = 1 − Φ(z). The normal curve calculator removes the table step when you want to check your work.
How Do You Compare Scores From Two Different Normal Distributions?
Standardize both scores, then compare the z-scores. A bigger z-score means a result further above its own average, whatever the scale. This is the main practical use of the standard normal.
Say a student scores 78 on a test with a mean of 70 and a standard deviation of 4, and 85 on a second test with a mean of 75 and a standard deviation of 10. The first z-score is (78 − 70) / 4 = 2.00, the 97.7th percentile. The second is (85 − 75) / 10 = 1.00, the 84.1st percentile. The 78 is the stronger result, even though 85 is the larger number. The normal curve calculator converts either z-score to a percentile.
How Is the Standard Normal Different From the t Distribution?
The t distribution looks like the standard normal but has heavier tails. It is used when the population standard deviation is unknown and has to be estimated from a small sample. As the sample grows, the t distribution moves toward the standard normal.
| Distribution | 95% two-sided critical value | P(|T| > 2) |
|---|---|---|
| Standard normal | 1.960 | 0.0455 |
| t, 5 degrees of freedom | 2.571 | 0.1019 |
| t, 10 degrees of freedom | 2.228 | 0.0734 |
| t, 30 degrees of freedom | 2.042 | 0.0546 |
| t, 120 degrees of freedom | 1.980 | 0.0478 |
With only 5 degrees of freedom, a result more than 2 standard errors from the mean happens 10.2% of the time under the t distribution, compared with 4.6% under the standard normal. That extra tail probability is the price of estimating σ. Use the t distribution calculator for small samples, and the standard normal when σ is known or the sample is large.
What Mistakes Do People Make With Standard Normal Problems?
- Mixing up variance and standard deviation. N(100, 225) has a standard deviation of 15, not 225. Divide by 15, not by 225.
- Calling any z-score “standard normal”. Z-scores of skewed data are standardized, but they are not standard normal. The label needs a normal parent distribution.
- Reading a tail the wrong way. Tables give the area to the left. “More than” needs 1 − Φ(z).
- Comparing densities as if they were probabilities. A curve height of 0.1159 is not an 11.6% chance. Only areas are probabilities.
- Using the sample standard deviation as if it were known. For small samples the t distribution is the safer choice.
- Rounding z too early. Rounding to one decimal can shift a probability by more than 0.01. Keep two decimals, or use the tool.
When you need the sample standard deviation first, the standard deviation calculator gives it for any list of numbers.
How Do You Calculate Both Versions in Excel, R and Python?
Every tool has a general function that takes μ and σ, and a shorter standard version that does not. Both give the same answer after standardizing. For X ~ N(64, 2.5²) and x = 66, the area to the left is 0.7881 either way.
| Task | Normal (any μ, σ) | Standard normal |
|---|---|---|
| Excel / Sheets | =NORM.DIST(66,64,2.5,TRUE) | =NORM.S.DIST(0.8,TRUE) |
| R | pnorm(66,64,2.5) | pnorm(0.8) |
| Python (SciPy) | norm.cdf(66,64,2.5) | norm.cdf(0.8) |
| TI-84 | normalcdf(-1E99,66,64,2.5) | normalcdf(-1E99,0.8) |
To go from an area back to a value, use NORM.INV, qnorm, norm.ppf or invNorm. In every package the standard deviation is the argument, not the variance.
Where Do Normal and Standard Normal Distributions Appear in Practice?
Anywhere a quantity clusters around an average. Manufacturers use them to set tolerance limits, teachers use z-scores to compare classes, and analysts use them to flag unusual readings. Standardizing is the common first step in all three.
In quality control, a process with a known mean and spread gives the share of parts outside a tolerance, as in the flour bags above. In education, standardized tests report scores on a scale built to have a fixed mean and spread, so results from different years line up. In science, many test statistics, such as the z statistic for a mean, follow a standard normal distribution when their assumptions hold. Large sample averages also tend toward a normal shape, which is why the standard normal appears so often in confidence intervals. When counts are involved, a normal curve can approximate a binomial, as the binomial probability calculator page explains. The descriptive vs inferential statistics guide shows how such an interval is built from one sample.
What Should You Remember About Normal vs Standard Normal?
- A normal distribution has any mean μ and standard deviation σ. The standard normal has μ = 0 and σ = 1.
- Standardize with z = (x − μ) / σ, and convert back with x = μ + zσ.
- Standardizing keeps the shape, the order and every probability. It changes the units, the center, the spread and the curve height.
- Curve heights are densities. Only areas are probabilities, so compare areas.
- One z table or one calculator then covers every normal distribution.
- The t distribution replaces the standard normal when σ is estimated from a small sample.
Frequently Asked Questions
Is the standard normal distribution a normal distribution?
Yes. It is the normal distribution with a mean of 0 and a standard deviation of 1. Every normal distribution belongs to the same family, and any of them can be converted into the standard normal with z = (x − μ) / σ.
Why is it called standard?
It is the reference version that statisticians standardize everything to. With a mean of 0 and a standard deviation of 1, it has no units, so a single table or formula covers every normal problem. Z-scores are the values on its axis.
What are the mean and variance of a z-score?
For a standardized variable, the mean is 0 and the variance is 1. For the list 4, 8, 6, 5, 9, the z-scores are −1.157, 0.772, −0.193, −0.675 and 1.254. They add to 0, and their sample standard deviation is exactly 1.
Does standardizing change probabilities?
No. The probability that X is below a value equals the probability that Z is below that value's z-score. Standardizing changes the units and the curve height, but areas stay the same, which is why one table works for all normal distributions.
Is standard normal the same as standard deviation?
No. Standard deviation is a measure of spread for any data set. Standard normal is a specific distribution whose standard deviation happens to be 1. Mixing them up is common because both use the word standard.
Can I use z-scores on data that is not normal?
You can always compute z-scores, and they still show how many standard deviations a value sits from the mean. But reading probabilities from the standard normal table is only valid when the data are roughly normal. For skewed data, the probabilities will be off.
When should I use the t distribution instead of the standard normal?
Use t when the population standard deviation is unknown and you estimate it from a sample, especially with fewer than about 30 observations. Its tails are heavier. With 5 degrees of freedom the 95% critical value is 2.571 instead of 1.960.
Sources: NIST/SEMATECH e-Handbook of Statistical Methods, section 1.3.6.6.1 (normal distribution); SciPy documentation, scipy.stats.norm; R documentation, Normal; Microsoft Support, NORM.DIST and NORM.S.DIST functions.
Related Calculators
Ready to run your own numbers? These do the next step, or browse the full list of free statistics calculators:
