Descriptive Statistics
Descriptive Statistics
Descriptive statistics summarizes and organizes data to make it easy to understand.
Why is it important?
• Helps to quickly understand large datasets
• Identifies patterns and trends
• Supports decision-making before deeper analysis
Descriptive Statistics
Key Measures
1. Measures of Central Tendency (Center)
• Mean → Average value
• Median → Middle value (robust to outliers)
• Mode → Most frequent value
2. Measures of Dispersion (Spread)
• Range → Max − Min
• Variance → Data variability
• Standard Deviation → Spread around mean
Arithmetic Mean (average)
The arithmetic mean is the sum of all values divided by the number of values.
Data: 2, 4, 6, 8
Mean = (2 + 4 + 6 + 8) / 4 = 5
Arithmetic Mean
When to Use?
• Data is numerical (quantitative)
• Data is evenly distributed (no extreme outliers)
• Useful for overall average performance
◦e.g., average marks, income, temperature
When NOT to Use?
• When data has outliers (mean gets distorted)
• For skewed distributions
Geometric Mean
The geometric mean is the n-th root of the product of n values, used for multiplicative data.
Example
Data: 2, 8
GM = √(2 × 8) = √16 = 4
Geometric Mean
When to Use?
• For growth rates (finance, population, investments)
• When values multiply over time
• For ratios or percentages
When NOT to Use?
• When data contains zero or negative values
• For simple additive averages (use arithmetic mean instead)
Harmonic Mean
The harmonic mean is the reciprocal of the average of reciprocals of the data values.
It is useful for rates and ratios.
Data: 2, 4
HM = 2 ÷ (1/2 + 1/4) = 2 ÷ (3/4) = 2.67
Harmonic Mean
When to Use?
• For rates, speed, or ratios
◦e.g., average speed, efficiency, price per unit
• When smaller values have more weight
When NOT to Use?
• When data has zero values (division by zero)
• For simple additive data (use arithmetic mean instead)
Key Idea:
“Harmonic mean gives a meaningful average for rates and ratios, emphasizing smaller values more.”
Median
What is it?
The median is the middle value of a dataset when the numbers are arranged in ascending order.
• It divides data into two equal halves.
• Robust to outliers (unlike mean).
Median
1. Odd Number of Observations
• Middle value = Median
Example: Data = 3, 5, 7
• Middle value = 5 → Median = 5
2. Even Number of Observations
• Median = Average of the two middle values
Example: Data = 2, 4, 6, 8
• Middle values = 4, 6
• Median = (4 + 6)/2 = 5
Make sure that the data are in ascending order, if it is not, then sort it
Mode
What is it?
The mode is the value that occurs most frequently in a dataset.
• Can have no mode, one mode, or multiple modes (bimodal, multimodal).
• Works for numerical and categorical data.
How to Find Mode
Example 1: Single Mode
Data: 2, 3, 4, 4, 5
• Most frequent value = 4 → Mode = 4
Example 2: No Mode
Data: 1, 2, 3, 4, 5
• No repeating values → Mode = No Mode(it is not 0)
Example 3: Bimodal
Data: 2, 2, 3, 3, 4
• Two most frequent values = 2 and 3 → Mode = 2, 3
Finding outlier using IQR method
• arrange data: 3, 6, 7, 12, 15, 17, 30
• Find Q1 and Q3 (note: Q2 is aka Median value)
• Q1 (median of Lower half excluding Q2: 3,6,7 —>6) (refer to class note)
• Q3 (Median of upper half excluding Q2: 15,17,30 —>17) (refer to class note)
• Find IQR (Q3-Q1) (17-6 =11)
• Lower range: Q1-1.5*IQR= 6-1.5*11= -10.5
• Upper range: Q3+1.5*IQR= 17+1.5*11=33.5
• Any value < lower range and > upper range is outlier (here is no outlier)
Even though 30 looks large, it is NOT an outlier mathematically.
Range
Range is the difference between the maximum and minimum values in a dataset.
It is the simplest measure of variability
Range
🔹 Advantages
• Very easy to calculate
• Quick idea of data spread
🔹 Limitations (Important for Teaching)
• Uses only two values (min & max)
• Ignores all other data points
• Highly affected by outliers
🔹 When to Use
• For quick comparison of datasets
• When rough idea of spread is enough
Variance (Measure of Spread)
Definition
Variance measures how far data points are spread out from the mean.
It tells us how much the values vary (deviate) from the average.
Key Idea
• Small variance → data points are close to mean
• Large variance → data points are widely spread
Population Variance
Sample Variance
•
Standard Deviation (S.D.)
Use when: You have data for the entire population
Standard Deviation (S.D.)
Sample S.D.
Use when: You are working with a sample
Note:
Why Square Root?
• Variance is in squared units
• Standard deviation converts it back to original units
Example: marks instead of marks²
Interpretation (Very Important for Slides)
• Small S.D. → data is close to mean (consistent)
• Large S.D. → data is spread out (variable)