Data Distributions and Visualization
Focus on Continuous Variables and the Normal Distribution
October 6, 2025
Data Distributions October 6, 2025 1 / 15
Outline
1 Continuous Random Variables
2 Probability Density Function (PDF)
3 The Normal Distribution
4 Standardization Z-scores
5 Percentiles of a Normal Distribution
6 Practice Problems
7 Conclusion
Data Distributions October 6, 2025 2 / 15
1. Continuous Random Variables
A random variable X is continuous if its possible values comprise an
entire interval on the number line.
Data Distributions October 6, 2025 3 / 15
1. Continuous Random Variables
A random variable X is continuous if its possible values comprise an
entire interval on the number line.
Unlike discrete variables, the probability that a continuous variable
takes on any single specific value is zero.
Data Distributions October 6, 2025 3 / 15
1. Continuous Random Variables
A random variable X is continuous if its possible values comprise an
entire interval on the number line.
Unlike discrete variables, the probability that a continuous variable
takes on any single specific value is zero.
Key Idea: For any specific value c, P(X = c) = 0.
Data Distributions October 6, 2025 3 / 15
1. Continuous Random Variables
A random variable X is continuous if its possible values comprise an
entire interval on the number line.
Unlike discrete variables, the probability that a continuous variable
takes on any single specific value is zero.
Key Idea: For any specific value c, P(X = c) = 0.
Implication: P(a ≤ X ≤ b) is the same as P(a < X < b).
Data Distributions October 6, 2025 3 / 15
1. Continuous Random Variables
A random variable X is continuous if its possible values comprise an
entire interval on the number line.
Unlike discrete variables, the probability that a continuous variable
takes on any single specific value is zero.
Key Idea: For any specific value c, P(X = c) = 0.
Implication: P(a ≤ X ≤ b) is the same as P(a < X < b).
Probability is found over intervals, not at specific points.
Data Distributions October 6, 2025 3 / 15
2. Probability Density Function (PDF)
The probability distribution of a continuous random variable X is
described by its Probability Density Function (PDF), denoted as f (x).
Properties of a PDF
Non-negativity: f (x) ≥ 0 for all values of x.
Total Area is 1: The total area under the curve of f (x) must equal 1.
Z ∞
f (x) dx = 1
−∞
Data Distributions October 6, 2025 4 / 15
2. Probability Density Function (PDF)
The probability distribution of a continuous random variable X is
described by its Probability Density Function (PDF), denoted as f (x).
Properties of a PDF
Non-negativity: f (x) ≥ 0 for all values of x.
Total Area is 1: The total area under the curve of f (x) must equal 1.
Z ∞
f (x) dx = 1
−∞
Calculating Probabilities with the PDF
The probability that X falls within an interval [a, b] is the area under the
PDF curve between a and b.
Z b
P(a ≤ X ≤ b) = f (x) dx
a
Data Distributions October 6, 2025 4 / 15
3. The Normal Distribution
The normal distribution is the most important probability distribution in
statistics. It’s defined by its mean µ (center) and its standard deviation σ
(spread).
Notation: We write X ∼ N(µ, σ 2 ) to say that X follows a normal
distribution with mean µ and variance σ 2 .
Data Distributions October 6, 2025 5 / 15
3. The Normal Distribution
The normal distribution is the most important probability distribution in
statistics. It’s defined by its mean µ (center) and its standard deviation σ
(spread).
Notation: We write X ∼ N(µ, σ 2 ) to say that X follows a normal
distribution with mean µ and variance σ 2 .
Properties of the Normal Distribution
The curve is bell-shaped and perfectly symmetric around its mean µ.
The mean, median, and mode are all equal.
The curve approaches the horizontal axis but never touches it.
Data Distributions October 6, 2025 5 / 15
4. The Standard Normal Distribution
The standard normal distribution is a special case of the normal
distribution with a mean of 0 and a standard deviation of 1.
Notation: The standard normal variable is usually denoted by Z , so
Z ∼ N(0, 1).
Data Distributions October 6, 2025 6 / 15
4. The Standard Normal Distribution
The standard normal distribution is a special case of the normal
distribution with a mean of 0 and a standard deviation of 1.
Notation: The standard normal variable is usually denoted by Z , so
Z ∼ N(0, 1).
Its PDF is denoted by ϕ(z) and its Cumulative Distribution Function
(CDF) is denoted by Φ(z), which gives the area to the left of a given
z-value.
Data Distributions October 6, 2025 6 / 15
4. The Standard Normal Distribution
The standard normal distribution is a special case of the normal
distribution with a mean of 0 and a standard deviation of 1.
Notation: The standard normal variable is usually denoted by Z , so
Z ∼ N(0, 1).
Its PDF is denoted by ϕ(z) and its Cumulative Distribution Function
(CDF) is denoted by Φ(z), which gives the area to the left of a given
z-value.
Standard Normal Table (Z-table): We use a Z-table (or software)
to find probabilities for the standard normal distribution. The table
typically provides values for Φ(z) = P(Z ≤ z).
Data Distributions October 6, 2025 6 / 15
5. Standardization and Z-scores
We can relate any normal distribution X ∼ N(µ, σ 2 ) to the standard
normal distribution Z ∼ N(0, 1) using a process called standardization.
Data Distributions October 6, 2025 7 / 15
5. Standardization and Z-scores
We can relate any normal distribution X ∼ N(µ, σ 2 ) to the standard
normal distribution Z ∼ N(0, 1) using a process called standardization.
The Z-score Formula
Any value x from the normal distribution can be converted to a z-score
using the formula:
X −µ
Z=
σ
Data Distributions October 6, 2025 7 / 15
5. Standardization and Z-scores
We can relate any normal distribution X ∼ N(µ, σ 2 ) to the standard
normal distribution Z ∼ N(0, 1) using a process called standardization.
The Z-score Formula
Any value x from the normal distribution can be converted to a z-score
using the formula:
X −µ
Z=
σ
The z-score measures how many standard deviations the value x is
from the mean µ.
A positive z-score means x is above the mean.
A negative z-score means x is below the mean.
Data Distributions October 6, 2025 7 / 15
How to Calculate Probabilities
1 Identify the mean µ and standard deviation σ for your variable X .
2 State the probability you want to find (e.g., P(X < a), P(X > b), or
P(a < X < b)).
X −µ
3 Standardize your x-values into z-scores using the formula Z = σ .
4 Rewrite the probability in terms of Z .
5 Look up the probability in the standard normal table (Z-table) or use
a calculator.
Data Distributions October 6, 2025 8 / 15
How to Calculate Probabilities
1 Identify the mean µ and standard deviation σ for your variable X .
2 State the probability you want to find (e.g., P(X < a), P(X > b), or
P(a < X < b)).
X −µ
3 Standardize your x-values into z-scores using the formula Z = σ .
4 Rewrite the probability in terms of Z .
5 Look up the probability in the standard normal table (Z-table) or use
a calculator.
Example: If X ∼ N(69, 3.52 ) find P(X > 72)
1 µ = 69, σ = 3.5.
72−69
2 Standardize x = 72: Z = 3.5 ≈ 0.86.
3 The problem becomes P(Z > 0.86).
4 Using the Z-table: P(Z > 0.86) = 1 − P(Z ≤ 0.86) = 1 − Φ(0.86).
Data Distributions October 6, 2025 8 / 15
6. Percentiles of a Normal Distribution
Sometimes you need to work backward: given a probability or percentile,
find the corresponding value of X .
Steps to find a percentile
1 Find the z-score that corresponds to the given cumulative probability
from the Z-table.
2 Unstandardize the z-score to find the X value using the rearranged
formula:
X = µ + Zσ
Data Distributions October 6, 2025 9 / 15
6. Percentiles of a Normal Distribution
Sometimes you need to work backward: given a probability or percentile,
find the corresponding value of X .
Steps to find a percentile
1 Find the z-score that corresponds to the given cumulative probability
from the Z-table.
2 Unstandardize the z-score to find the X value using the rearranged
formula:
X = µ + Zσ
Find the 95th percentile of X ∼ N(100, 152 )
We need the z-score such that the area to its left is 0.95. From the
Z-table, this z is approximately 1.645.
Data Distributions October 6, 2025 9 / 15
6. Percentiles of a Normal Distribution
Sometimes you need to work backward: given a probability or percentile,
find the corresponding value of X .
Steps to find a percentile
1 Find the z-score that corresponds to the given cumulative probability
from the Z-table.
2 Unstandardize the z-score to find the X value using the rearranged
formula:
X = µ + Zσ
Find the 95th percentile of X ∼ N(100, 152 )
We need the z-score such that the area to its left is 0.95. From the
Z-table, this z is approximately 1.645.
Use the formula:
X = 100 + (1.645)(15) ≈ 124.675
Data Distributions October 6, 2025 9 / 15
Problem 1: Normal Distribution Probability
Problem: The lifetime of a certain type of electronic component is
normally distributed with a mean (µ) of 500 hours and a standard
deviation (σ) of 25 hours. What is the probability that a randomly
selected component will last for more than 540 hours?
Data Distributions October 6, 2025 10 / 15
Problem 1: Normal Distribution Probability
Problem: The lifetime of a certain type of electronic component is
normally distributed with a mean (µ) of 500 hours and a standard
deviation (σ) of 25 hours. What is the probability that a randomly
selected component will last for more than 540 hours?
Solution:
We have µ = 500, σ = 25. We want to find P(X > 540).
First, standardize the value 540 to a z-score:
540 − 500 40
Z= = = 1.60
25 25
The problem is now to find P(Z > 1.60).
Using a standard normal table, Φ(1.60) = P(Z ≤ 1.60) ≈ 0.9452.
Since the total area is 1, the probability is:
P(Z > 1.60) = 1 − P(Z ≤ 1.60) = 1 − 0.9452 = 0.0548
Data Distributions October 6, 2025 10 / 15
Problem 2: Working Backwards (Percentiles)
Problem: Scores on a standardized exam are normally distributed with a
mean (µ) of 150 and a standard deviation (σ) of 18. To be accepted into
a top program, a student must score in the top 10%. What is the
minimum score required to be accepted?
Data Distributions October 6, 2025 11 / 15
Problem 2: Working Backwards (Percentiles)
Problem: Scores on a standardized exam are normally distributed with a
mean (µ) of 150 and a standard deviation (σ) of 18. To be accepted into
a top program, a student must score in the top 10%. What is the
minimum score required to be accepted?
Solution:
”Top 10%” corresponds to the 90th percentile. We need to find the
score x such that P(X ≤ x) = 0.90.
Find the z-score for a cumulative area of 0.90. From a Z-table,
z ≈ 1.28.
Now, unstandardize this z-score to find the corresponding exam score
x:
X = µ + Z σ = 150 + (1.28)(18) = 150 + 23.04 = 173.04
Answer: Since scores are likely integers, the minimum required score
is 174.
Data Distributions October 6, 2025 11 / 15
Problem 3: Interpreting a Normal Probability Plot
Problem: An engineer collects 50 data points on the strength of a new
alloy. When she creates a normal probability plot, she observes that the
points form a distinct curve that is concave down. Does the data appear
to be normally distributed? If not, what does the pattern suggest?
Data Distributions October 6, 2025 12 / 15
Problem 3: Interpreting a Normal Probability Plot
Problem: An engineer collects 50 data points on the strength of a new
alloy. When she creates a normal probability plot, she observes that the
points form a distinct curve that is concave down. Does the data appear
to be normally distributed? If not, what does the pattern suggest?
Solution:
No, the data does not appear to be normally distributed. For the data
to be normal, the points on the plot should fall close to a straight line.
Data Distributions October 6, 2025 12 / 15
Problem 3: Interpreting a Normal Probability Plot
Problem: An engineer collects 50 data points on the strength of a new
alloy. When she creates a normal probability plot, she observes that the
points form a distinct curve that is concave down. Does the data appear
to be normally distributed? If not, what does the pattern suggest?
Solution:
No, the data does not appear to be normally distributed. For the data
to be normal, the points on the plot should fall close to a straight line.
The distinct concave-down curve is a classic sign of positive skew
(or right-skewed) data.
This means the distribution has a long tail to the right, and the
assumption of normality is likely violated.
Data Distributions October 6, 2025 12 / 15
Problem 4: Identifying a Lognormal Distribution
Problem: A scientist analyzes pollutant concentrations. A normal
probability plot of the original data is clearly curved. However, after taking
the natural logarithm (ln) of each value, a new plot of the transformed
data is very linear. What can be concluded?
Data Distributions October 6, 2025 13 / 15
Problem 4: Identifying a Lognormal Distribution
Problem: A scientist analyzes pollutant concentrations. A normal
probability plot of the original data is clearly curved. However, after taking
the natural logarithm (ln) of each value, a new plot of the transformed
data is very linear. What can be concluded?
Solution:
The scientist can conclude that the original data for pollutant
concentration most likely follows a lognormal distribution.
Data Distributions October 6, 2025 13 / 15
Problem 4: Identifying a Lognormal Distribution
Problem: A scientist analyzes pollutant concentrations. A normal
probability plot of the original data is clearly curved. However, after taking
the natural logarithm (ln) of each value, a new plot of the transformed
data is very linear. What can be concluded?
Solution:
The scientist can conclude that the original data for pollutant
concentration most likely follows a lognormal distribution.
The definition of a lognormal distribution is that the natural
logarithm of the data, Y = ln(X ), is normally distributed.
Data Distributions October 6, 2025 13 / 15
Problem 4: Identifying a Lognormal Distribution
Problem: A scientist analyzes pollutant concentrations. A normal
probability plot of the original data is clearly curved. However, after taking
the natural logarithm (ln) of each value, a new plot of the transformed
data is very linear. What can be concluded?
Solution:
The scientist can conclude that the original data for pollutant
concentration most likely follows a lognormal distribution.
The definition of a lognormal distribution is that the natural
logarithm of the data, Y = ln(X ), is normally distributed.
The fact that the normal probability plot for the logged data is linear
provides strong evidence for this conclusion.
Data Distributions October 6, 2025 13 / 15
Summary of Key Concepts
Continuous Variables: Probabilities are areas under a PDF curve,
not values at a single point.
Normal Distribution: A symmetric, bell-shaped distribution defined
by its mean (µ) and standard deviation (σ).
Standardization: The z-score (Z = X σ−µ ) is a powerful tool to
convert any normal distribution to the standard normal distribution
(N(0, 1)), allowing us to use Z-tables.
Working Backwards: We can use percentiles and z-scores to find
data values (X = µ + Z σ) corresponding to a given probability.
Assessing Normality: Tools like normal probability plots help us
visually check if our data fits a normal distribution, which is a key
assumption for many statistical methods.
Data Distributions October 6, 2025 14 / 15
Data Distributions October 6, 2025 15 / 15