Range: The difference between the maximum and minimum values in a dataset.
Interquartile Range (IQR): The difference between the third quartile (Q3) and first quartile (Q1), representing the spread of the middle 50% of data.
Quartiles: Values that divide ordered data into four equal parts (Q1 = 25th percentile, Q2 = 50th percentile/median, Q3 = 75th percentile).
- Range: Maximum value - Minimum value
- Quartiles: Q1 = median of lower half, Q3 = median of upper half
- IQR: Q3 - Q1
Data is already sorted: 72, 78, 82, 85, 88, 90, 92, 95, 98, 100
Minimum = 72, Maximum = 100
Range = Maximum - Minimum = 100 - 72 = 28
With 10 values (even), split into two halves: 72, 78, 82, 85, 88 and 90, 92, 95, 98, 100
Lower half: 72, 78, 82, 85, 88
Q1 = median of lower half = 82
Upper half: 90, 92, 95, 98, 100
Q3 = median of upper half = 95
IQR = Q3 - Q1 = 95 - 82 = 13
Range = 28, IQR = 13. The range shows the total spread from lowest to highest score, while the IQR shows the spread of the middle 50% of scores (between 82 and 95).
• Range formula: Max - Min
• Quartile calculation: Q1 = median of lower half, Q3 = median of upper half
• IQR formula: Q3 - Q1
Robust measure: A measure that is not significantly affected by outliers or extreme values.
Outlier sensitivity: How much a measure changes when extreme values are present.
Range = 100 - 60 = 40
Q1 = 65 (average of 60 and 70), Q3 = 95 (average of 90 and 100)
IQR = 95 - 65 = 30
Range = 150 - 60 = 90
Q1 = 65 (average of 60 and 70), Q3 = 95 (average of 90 and 150)
IQR = 95 - 65 = 30
Range: Dataset A = 40, Dataset B = 90 (increased by 50)
IQR: Dataset A = 30, Dataset B = 30 (no change)
The value 150 in Dataset B is an outlier compared to the other values
This affects the range but not the IQR
The IQR (30) better represents the typical spread because it's not affected by the outlier
The range is inflated by the extreme value
Dataset A: Range = 40, IQR = 30; Dataset B: Range = 90, IQR = 30. The IQR better represents the typical spread because it's not affected by the outlier (150). The range increases significantly due to the outlier.
• Range sensitivity: Affected by all values, especially outliers
• IQR robustness: Only affected by middle 50% of data
• Outlier impact: Range is more sensitive to extreme values
Percentile: A value below which a certain percentage of data falls.
Quartile range: The interval between Q1 and Q3 containing 50% of the data.
Data is sorted: 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70
Minimum = 15, Maximum = 70
Range = 70 - 15 = 55
With 12 values, split into two halves: 15, 20, 25, 30, 35, 40 and 45, 50, 55, 60, 65, 70
Lower half: 15, 20, 25, 30, 35, 40
Q1 = median of lower half = (25 + 30) ÷ 2 = 27.5
Upper half: 45, 50, 55, 60, 65, 70
Q3 = median of upper half = (55 + 60) ÷ 2 = 57.5
IQR = Q3 - Q1 = 57.5 - 27.5 = 30
Number of values in IQR range (27.5 to 57.5): 30, 35, 40, 45, 50, 55 → 6 values
Percentage = (6 ÷ 12) × 100% = 50%
Range = 55, IQR = 30. Exactly 50% of the data (6 out of 12 values) falls within the IQR. This confirms that the IQR always contains the middle 50% of the data.
• IQR definition: Always contains middle 50% of data
• Quartile calculation: Q1 = 25th percentile, Q3 = 75th percentile
• Percentile interpretation: 50% of data between Q1 and Q3
Full Spread
Middle 50%
Q3 = 75th %
Measures of spread: Statistical measures that describe how data values are distributed or dispersed.
Range: The simplest measure of spread, calculated as the difference between maximum and minimum values.
Interquartile Range (IQR): The range of the middle 50% of data, calculated as Q3 - Q1.
- Range: For a quick overview of total spread, when outliers are not present
- IQR: When you want to measure spread of the central data, when outliers are present
• Range sensitivity: Affected by all values, especially outliers
• IQR robustness: Only affected by middle 50% of data
• Quartile division: Q1 (25%), Q2 (50%), Q3 (75%) percentiles
Linear transformation: Adding or subtracting a constant to each value in a dataset.
Scaling transformation: Multiplying or dividing each value in a dataset by a constant.
Original range = 40, IQR = 15
Let's say min = x, max = x + 40, so range = (x + 40) - x = 40
Let's say Q1 = y, Q3 = y + 15, so IQR = (y + 15) - y = 15
New min = x + 10, New max = (x + 40) + 10 = x + 50
New range = (x + 50) - (x + 10) = 40 (unchanged)
New Q1 = y + 10, New Q3 = (y + 15) + 10 = y + 25
New IQR = (y + 25) - (y + 10) = 15 (unchanged)
New min = 3x, New max = 3(x + 40) = 3x + 120
New range = (3x + 120) - 3x = 120 (multiplied by 3)
New Q1 = 3y, New Q3 = 3(y + 15) = 3y + 45
New IQR = (3y + 45) - 3y = 45 (multiplied by 3)
For linear transformation (adding constant c):
• Range_new = Range_original
• IQR_new = IQR_original
For scaling transformation (multiplying by constant c):
• Range_new = Range_original × c
• IQR_new = IQR_original × c
Adding 10: Range stays 40, IQR stays 15
Multiplying by 3: Range becomes 40×3 = 120, IQR becomes 15×3 = 45
Adding 10 to each value: Range = 40, IQR = 15 (both unchanged). Multiplying each value by 3: Range = 120, IQR = 45 (both multiplied by 3). Linear transformations don't affect spread measures, but scaling transformations do.
• Linear transformation: Spread measures remain unchanged
• Scaling transformation: Spread measures scale by the same factor
• Translation invariance: Adding constants doesn't change spread
Business context: Applying statistical measures to practical business situations.
Typical variation: The usual range of fluctuations in business metrics.
Data is sorted: 18, 22, 25, 28, 30, 32, 35, 38, 40, 42, 45, 120
Minimum = 18, Maximum = 120
Range = 120 - 18 = 102 thousand dollars
Split into two halves: 18, 22, 25, 28, 30, 32 and 35, 38, 40, 42, 45, 120
Lower half: 18, 22, 25, 28, 30, 32
Q1 = (25 + 28) ÷ 2 = 26.5
Upper half: 35, 38, 40, 42, 45, 120
Q3 = (40 + 42) ÷ 2 = 41
IQR = Q3 - Q1 = 41 - 26.5 = 14.5 thousand dollars
The value 120 is significantly higher than other values
This represents the Black Friday sales spike
Range = 102 (heavily influenced by the outlier)
IQR = 14.5 (focuses on regular business days)
The IQR better represents typical daily variation
Range = $102,000, IQR = $14,500. The IQR better represents typical daily sales variation because it excludes the outlier (Black Friday sales). The range is inflated by the extreme value and doesn't reflect regular business patterns.
• Outlier impact: Range is highly sensitive to extreme values
• IQR robustness: Not affected by outliers in tails
• Business decision-making: Choose measures that reflect typical conditions
Range: The difference between the maximum and minimum values in a dataset. Simplest measure of spread.
Interquartile Range (IQR): The difference between the third quartile (Q3) and first quartile (Q1), representing the spread of the middle 50% of data.
Quartiles: Values that divide ordered data into four equal parts: Q1 (25th percentile), Q2 (50th percentile/median), Q3 (75th percentile).
Measures of spread: Statistical measures that describe how data values are distributed or dispersed around the center.
- Range calculation: Find maximum and minimum values, subtract minimum from maximum
- Quartile calculation: Arrange data in order, find Q1 (median of lower half) and Q3 (median of upper half)
- IQR calculation: Subtract Q1 from Q3
• Linear transformation (adding constant c): Range_new = Range_orig, IQR_new = IQR_orig
• Scaling transformation (multiplying by constant c): Range_new = Range_orig × c, IQR_new = IQR_orig × c
• Outlier sensitivity: Range is highly sensitive to outliers, IQR is robust to outliers
• Data usage: Range uses 2 values, IQR uses 50% of data values
Range: Total spread, affected by outliers
IQR: Middle 50% spread, robust to outliers
Quartiles: Divide data into 4 equal parts
Choose based on data characteristics and purpose
Analysis: Different measures of spread provide different perspectives on data variability.
- Range: Total spread from min to max, sensitive to outliers
- IQR: Middle 50% of data, robust to extreme values
- Quartiles: Boundary values for data quarters
- Selection: Depends on distribution shape and analysis goals