The Standard Normal Distribution N(0, 1)
Mateo's score in an online racing game is exactly 1 standard deviation above the average. Is that good? What fraction of all players did he beat? Scores like this follow a bell-shaped curve, and once you can read one special bell curve, the standard normal curve, you can answer questions like this for almost any data.
The standard normal curve
In a continuous distribution, the probability of an interval is the area under the curve over that interval, and the total area is $1$. The most important continuous distribution is the normal distribution. Its standard version is the one every table and calculator is built around.
| Interval | Approximate area |
|---|---|
| $-1 < Z < 1$ | $68\%$ |
| $-2 < Z < 2$ | $95\%$ |
| $-3 < Z < 3$ | $99.7\%$ |
Observe the following representation of the standard normal distribution. If we know that the area of the shaded region to the right of $z = 1,96$ is $0,025$, reason…
Observe the following representation of the standard normal distribution. If we know that the area of the shaded region to the right of $z = 1,96$ is $0,025$, reason what the cumulative probability up to that point will be, $P(Z \leq 1,96)$.
Reading the z-table
A standard z-table lists $\Phi(z) = P(Z \le z)$: the area to the left of $z$. You find the row with $z$ to the tenths place and the column with the hundredths digit. Here is a small piece of a full table (use the complete table for your problems).
| $z$ | .00 | .01 | .02 | .03 | .04 | .05 | .06 | .07 | .08 | .09 |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.6 | .7257 | .7291 | .7324 | .7357 | .7389 | .7422 | .7454 | .7486 | .7517 | .7549 |
| 1.3 | .9032 | .9049 | .9066 | .9082 | .9099 | .9115 | .9131 | .9147 | .9162 | .9177 |
| 1.7 | .9554 | .9564 | .9573 | .9582 | .9591 | .9599 | .9608 | .9616 | .9625 | .9633 |
In a reflex training app, users' results have been standardized following a standard normal distribution $Z \sim N(0, 1)$. If a player obtains a standardized score of…
In a reflex training app, users' results have been standardized following a standard normal distribution $Z \sim N(0, 1)$. If a player obtains a standardized score of $z = 1,45$, calculate the probability that a randomly chosen user has a lower score than theirs; that is, calculate $P(Z < 1,45)$.
The time it takes students to complete a logic challenge on TikTok follows a standard normal distribution once the data has been standardized. Calculate the probability…
The time it takes students to complete a logic challenge on TikTok follows a standard normal distribution once the data has been standardized. Calculate the probability that a student takes more time than the mean corresponding to a value of $Z = 0,82$. That is, find $P(Z > 0,82)$.
Given a random variable $Z$ that follows a standard normal distribution $N(0,1)$, use the probability table to calculate the following areas under the curve: a)…
Given a random variable $Z$ that follows a standard normal distribution $N(0,1)$, use the probability table to calculate the following areas under the curve:
- a) $P(Z \leq 1,48)$
- b) $P(Z < 0,35)$
- c) $P(Z \geq 2,12)$
- d) $P(Z > 1,96)$
Graphically represent the area corresponding to part c) on a bell curve.
Negative z-values and symmetry
Many tables list only $z \ge 0$. You do not need more: the bell curve is a mirror image, so the area to the left of $-z$ equals the area to the right of $z$.
A study has been carried out on the monthly savings of young people in a city. After standardizing the variable, it is observed that it follows a $N(0, 1)$ distribution.…
A study has been carried out on the monthly savings of young people in a city. After standardizing the variable, it is observed that it follows a $N(0, 1)$ distribution. Using the symmetry properties of the normal distribution, calculate the probability that a young person has a savings level lower than a standardized value of $z = -1,26$.
In an entrance exam for a vocational training program, the scores fit a standard normal distribution. Knowing that the curve is symmetrical with respect to the y-axis,…
In an entrance exam for a vocational training program, the scores fit a standard normal distribution. Knowing that the curve is symmetrical with respect to the y-axis, determine the probability that a candidate obtains a standardized score greater than $-0,54$.
Making use of the symmetry properties of the standard normal distribution $N(0,1)$, determine the following probabilities for negative values of the variable: a)…
Making use of the symmetry properties of the standard normal distribution $N(0,1)$, determine the following probabilities for negative values of the variable:
- a) $P(Z \leq -0,74)$
- b) $P(Z < -1,25)$
- c) $P(Z \geq -0,58)$
- d) $P(Z > -2,03)$
Area between two z-values
To find the area between $a$ and $b$, take everything to the left of $b$ and remove everything to the left of $a$.
Calculate the probability that a random variable $Z \sim N(0, 1)$ takes values within the interval $[0,45, \, 1,88]$. Remember that to calculate the area between two…
Calculate the probability that a random variable $Z \sim N(0, 1)$ takes values within the interval $[0,45, \, 1,88]$. Remember that to calculate the area between two positive values $a$ and $b$, you must subtract their cumulative probabilities: $P(a < Z < b) = P(Z < b) - P(Z < a)$.
In the quality control of some electronic components, it is accepted that the measurement error follows a $N(0, 1)$ distribution. Find the probability that the…
In the quality control of some electronic components, it is accepted that the measurement error follows a $N(0, 1)$ distribution. Find the probability that the standardized error lies in the symmetrical interval $[-1,5, \, 1,5]$.
Calculate the probability that the random variable $Z \sim N(0,1)$ takes values within the following intervals: a) $P(0,45 \leq Z \leq 1,82)$ b) $P(-1,05 < Z < 2,15)$ c)…
Calculate the probability that the random variable $Z \sim N(0,1)$ takes values within the following intervals:
- a) $P(0,45 \leq Z \leq 1,82)$
- b) $P(-1,05 < Z < 2,15)$
- c) $P(-2,30 \leq Z \leq -0,90)$
Explain by means of a drawing why in case c) the result must be equal to the probability of the corresponding positive interval.
Symmetric intervals, two tails, and independence
Intervals of the form $-k < Z < k$ appear constantly. Symmetry gives a shortcut, and it explains why.
When two events $A$ and $B$ are independent, $P(A \text{ and } B) = P(A)\cdot P(B)$. This lets you combine normal probabilities for different people.
Calculate in a reasoned way, using the properties of the standard normal distribution $Z \sim N(0, 1)$, the value of the following probabilities without directly using…
Calculate in a reasoned way, using the properties of the standard normal distribution $Z \sim N(0, 1)$, the value of the following probabilities without directly using the table for negative values (only use the table for positive values):
- a) $P(|Z| \ge 1,25)$
- b) $P(-0,5 \le Z \le 0,5)$
- c) $P(Z \le -1,96 \cup Z \ge 1,96)$
Prove algebraically that for any variable $Z \sim N(0, 1)$ and any real value $k > 0$, it holds that $P(-k \le Z \le k) = 2 \cdot \Phi(k) - 1$. Subsequently, use this…
Prove algebraically that for any variable $Z \sim N(0, 1)$ and any real value $k > 0$, it holds that $P(-k \le Z \le k) = 2 \cdot \Phi(k) - 1$.
Subsequently, use this formula to find the probability that $Z$ is at a distance from the mean greater than $1,645$ standard deviations.
A study has been carried out on the reaction time of teenagers to a visual stimulus. After standardizing the data to a $N(0, 1)$, it is observed that $15,87\%$ of the…
A study has been carried out on the reaction time of teenagers to a visual stimulus. After standardizing the data to a $N(0, 1)$, it is observed that $15,87\%$ of the subjects have a very slow reaction time (high values of $Z$).
If we choose two teenagers at random, what is the probability that both have a value of $Z$ between $-1$ and $1$? Assume that the reactions of both are independent.
Working backward: from probability to z
Sometimes you know the probability and need the z-value. Now you search inside the table for the area, then read the row and column to build $z$.
Given a standard normal distribution $Z \sim N(0, 1)$, use the table in reverse to find the value of the constant $k$ such that the probability of obtaining a value less…
Given a standard normal distribution $Z \sim N(0, 1)$, use the table in reverse to find the value of the constant $k$ such that the probability of obtaining a value less than it is exactly $0,9750$. That is, look for $k$ in: $P(Z < k) = 0,9750$.
In the standard normal distribution $N(0,1)$, it is common to have to take the "reverse path": knowing the probability and looking for the value of the abscissa. Find…
In the standard normal distribution $N(0,1)$, it is common to have to take the "reverse path": knowing the probability and looking for the value of the abscissa. Find the value of the constant $k$ in each of the following cases:
- a) $P(Z \leq k) = 0,8749$
- b) $P(Z < k) = 0,9913$
- c) $P(Z \leq k) = 0,0516$
- d) $P(Z < k) = 0,3336$
Find the value of the abscissa $k$ in a distribution $N(0,1)$ knowing that: a) The area to the right of $k$ is $0,1515$. b) The area to the right of $k$ is $0,9418$. c)…
Find the value of the abscissa $k$ in a distribution $N(0,1)$ knowing that:
- a) The area to the right of $k$ is $0,1515$.
- b) The area to the right of $k$ is $0,9418$.
- c) $P(Z > k) = 0,0228$
Percentiles and z-scores
A percentile tells you the percentage of the data that falls below a value. For $Z \sim N(0, 1)$, the $p$th percentile is the value $z_p$ with $P(Z < z_p) = \dfrac{p}{100}$. So $\Phi(z)$ is the percent of the population below $z$, and going backward gives you the $z$ for a given percentile.
Two students from different high schools want to compare their results on an English exam. As it is not the same exam, they use the standardized score $Z$ (which follows…
Two students from different high schools want to compare their results on an English exam. As it is not the same exam, they use the standardized score $Z$ (which follows a $N(0,1)$) to know who is in a better position with respect to their group.
-
Student A has a standardized score of $z_A = 1,2$.
-
Student B has a standardized score of $z_B = 0,8$.
Calculate the percentage of students below each of them in their respective school and interpret who has achieved a more outstanding result.
In the world of esports, the skill of players of a famous video game is measured with a standardized score that follows a $N(0, 1)$. The developer company has decided to…
In the world of esports, the skill of players of a famous video game is measured with a standardized score that follows a $N(0, 1)$. The developer company has decided to award an exclusive skin only to players who exceed the 99th percentile. On the other hand, players who are below the value $-2,33$ enter a "basic training" program.
What is the probability that a randomly chosen player does not receive the skin but also does not have to enter the training program?
Critical values: the middle area
A common question is: what symmetric interval $-k$ to $k$ captures a given probability $C$? The leftover probability $1 - C$ is split evenly between the two tails, which is why this value is often written $z_{\alpha/2}$ with $\alpha = 1 - C$. These values are called critical values and appear in confidence intervals.
In advanced statistical studies, symmetrical intervals with respect to the origin that enclose a given probability are often sought. Find the positive value of $k$ such…
In advanced statistical studies, symmetrical intervals with respect to the origin that enclose a given probability are often sought. Find the positive value of $k$ such that:
- a) $P(-k \leq Z \leq k) = 0,90$
- b) $P(-k < Z < k) = 0,95$
- c) $P(-k \leq Z \leq k) = 0,99$
Note: These values of $k$ are known as critical values and are fundamental for calculating confidence intervals.
A streaming platform classifies its users according to their "loyalty index", which follows a distribution $N(0, 1)$. The system considers "VIP Users" to be those who…
A streaming platform classifies its users according to their "loyalty index", which follows a distribution $N(0, 1)$. The system considers "VIP Users" to be those who are in the central interval that concentrates $68,26\%$ of the population. However, for an aggressive marketing campaign, they want to redefine this group.
Find the symmetrical values $-z_{\alpha/2}$ and $z_{\alpha/2}$ that enclose a probability of $0,98$. Explain what percentage of users would be left out of this new group at each end.
Find a positive value $m$ such that the area enclosed under the standard normal curve $N(0, 1)$ between $-m$ and $m$ is three times the total area of the two outer tails…
Find a positive value $m$ such that the area enclosed under the standard normal curve $N(0, 1)$ between $-m$ and $m$ is three times the total area of the two outer tails (the sum of the areas to the left of $-m$ and to the right of $m$).
Setting up equations with $\Phi$
Harder problems describe a relationship between two areas. The trick is to let $p = \Phi(k)$ be the unknown. Then every area can be written with $p$ and the complement rule, and you solve a linear equation. Finally you go backward in the table to find $k$.
Given a variable $Z \sim N(0,1)$, calculate the value of $a$ knowing that the probability of the variable being greater than $a$ is seven times greater than the…
Given a variable $Z \sim N(0,1)$, calculate the value of $a$ knowing that the probability of the variable being greater than $a$ is seven times greater than the probability of it being less than $a$, where $a$ is a negative value.
Use the relationship $P(Z < a) + P(Z > a) = 1$ and the standard normal tables to help you.
In a standard normal distribution $N(0, 1)$, it is known that there is a positive value $k$ such that the probability that the variable $Z$ is greater than $k$ is…
In a standard normal distribution $N(0, 1)$, it is known that there is a positive value $k$ such that the probability that the variable $Z$ is greater than $k$ is exactly a quarter of the probability that $Z$ is less than $k$. Using the symmetry properties of the bell curve and the distribution tables, calculate the value of this parameter $k$ to two decimal places.
Given a random variable $Z \sim N(0, 1)$, calculate the value of the abscissa $a$ knowing that the following probability relationship holds:…
Given a random variable $Z \sim N(0, 1)$, calculate the value of the abscissa $a$ knowing that the following probability relationship holds:
$P(Z < a) = 3 \cdot P(a < Z < 1,5)$
Note: You will need to set up an equation using the distribution function $\Phi(z)$ and solve by looking it up in the table.
Summary
Back to the start: Mateo's score is $z = 1$, and $\Phi(1) = 0.8413$. He beat about $84\%$ of all players.