DESCRIPTIVE STATISTICS 1
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
X nX fX pX
x₁ n₁ f₁ p₁
x₂ n₂ f₂ p₂
… … … …
xₖ nₖ fₖ pₖ
What does this mean?
X — the values your variable can take (like 0, 1, 2 pets, or "Red", "Blue",
"Green")
nX — absolute frequency: how many times each value appears (n₁,
n₂, ..., nₖ)
fX — relative frequency: each nᵢ divided by the total number of
observations (n)
ni
f i= k
∑ ni
i=1
pX — percentage frequency: each fᵢ multiplied by 100
pi=f i ∙ 100
🔹 Also:
k
n=∑ ni → this is just the total number of observations in your dataset.
i=1
So to recap:
You can’t use this form with continuous variables without grouping
them.
It’s a super useful way to summarize your data and understand how
values are distributed.
CUMULATIVE FREQUENCY DISTRIBUTION
A cumulative frequency tells you how many values are less than or
equal to a certain value.
So instead of just counting how often each individual value occurs (like in a
regular frequency table), you add up all the frequencies up to and including
that value.
🔹 When can you use it?
When your variable X is:
o Ordinal (categories with a meaningful order: e.g., "Low",
"Medium", "High")
o Or quantitative (numeric values)
You can't use it with nominal variables (like "Apple", "Banana") because they
don’t have an order.
X nX (freq) NX (cumulative freq)
x₁ n₁ N₁
x₂ n₂ N₂ = n₁ + n₂
x₃ n₃ N₃ = n₁ + n₂ + n₃
… … …
X nX (freq) NX (cumulative freq)
xₖ nₖ Nₖ = total n
j
N j = ∑ ni
i=1
So, Nj tells you how many data points are ≤ xⱼ.
🔹 Example
Say you asked students how many books they read this month:
Data: 0, 1, 2, 0, 1, 3, 1, 2, 2, 3
First, frequency distribution:
X n
(Books) X
0 2
1 3
2 3
3 2
Now build cumulative frequency:
X n N
(Books) X X
0 2 2
1 3 5
2 3 8
3 2 10
🔹 What does it tell you?
NX = 5 at X = 1 → means 5 students read at most 1 book
NX = 8 at X = 2 → 8 students read 2 or fewer books
NX = 10 at X = 3 → all students, total count
BIVARIATE FREQUENCY DISTRIBUTION
🔹 What is a bivariate frequency distribution?
It’s a table that shows how often combinations of two variables occur
together.
Think of one variable as going across the top (columns) and the other
down the side (rows).
This type of table is also known as a contingency table or cross-
tabulation.
🔹 Notation Breakdown
Let’s say:
Variable X has values: x1, x2, …, xs
Variable Y has values: y1, y2, …, yt
Then:
nij = number of times we observe the pair (xi, yj)
o This is called the bivariate absolute frequency.
o Formally: nij =¿ {( x , y ) : x=x i∧ y= y j }
Here’s what it looks like:
y₁ y₂ … yⱼ … yₜ Row Totals
x₁ n₁₁ n₁₂ … n₁ⱼ … n₁ₜ n₁·
x₂ n₂₁ n₂₂ … n₂ⱼ … n₂ₜ n₂·
… … … … … …
xᵢ nᵢ₁ nᵢ₂ … nᵢⱼ … nᵢₜ nᵢ·
… … … … … …
xₛ nₛ₁ nₛ₂ … nₛⱼ … nₛₜ nₛ·
Col Totals n·₁ n·₂ … n·ⱼ … n·ₜ n
🔹 Explanation of Totals:
nᵢ· = total for row i (sum across y-values for a fixed xᵢ)
t
ni . =∑ n ij
j=1
n·ⱼ = total for column j (sum across x-values for a fixed yⱼ)
s
n. j=∑ nij
i=1
n = grand total (sum of all cells)
s t
n=∑ ∑ nij
i=1 j=1
🔹 Why is this useful? You can study relationships between two variables.
Helps with things like calculating conditional probabilities, correlations, or doing
chi-square tests.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.
FREQUENCY DISTRIBUTION
In descriptive statistics, a frequency distribution is a way of organizing
data to show how often each different value (or range of values) occurs in a
dataset. They can be:
Ungrouped: like the example above, each individual value is shown.
Grouped: used when you have a lot of values, and you group them into
ranges (e.g. 70–79, 80–89, etc.).
Relative frequency: shows percentages instead of counts.
Cumulative frequency: shows the running total of frequencies.
🔹 First, what are we talking about?
We’re looking at how to create a frequency distribution table for a variable
X that can be:
Qualitative (like colors, brands, or categories)
Quantitative discrete (like number of pets, goals in a match, etc.)
🚫 It specifically does not apply to continuous variables.
Why? Because continuous variables (like height, weight, temperature) can take
infinitely many values, so we group them into intervals instead of listing
every possible value.