Measures of center: Statistical measures that represent the typical or central value of a dataset (mean, median, mode).
- Mean: Sum of all values divided by the number of values
- Median: Middle value when data is ordered from least to greatest
- Mode: Value that appears most frequently in the dataset
- Compare measures to assess symmetry and outliers
Ordered data: 75, 80, 85, 85, 90, 92, 95, 95, 95, 100
Mean = (75 + 80 + 85 + 85 + 90 + 92 + 95 + 95 + 95 + 100) ÷ 10
Mean = 897 ÷ 10 = 89.7
With 10 values (even number), median is average of 5th and 6th values
Median = (90 + 92) ÷ 2 = 91
Mode = 95 (appears 3 times, more than any other value)
| Measure | Value | Interpretation |
|---|---|---|
| Mean | 89.7 | Average score |
| Median | 91 | Middle value |
| Mode | 95 | Most common score |
Sum all values: 75 + 80 + 85 + 85 + 90 + 92 + 95 + 95 + 95 + 100 = 897
Divide by count: 897 ÷ 10 = 89.7
Data is already ordered: 75, 80, 85, 85, 90, 92, 95, 95, 95, 100
With 10 values, median is average of 5th and 6th values
Median = (90 + 92) ÷ 2 = 91
Count frequencies: 75(1), 80(1), 85(2), 90(1), 92(1), 95(3), 100(1)
Mode = 95 (appears 3 times, most frequent)
Mean (89.7) < Median (91) < Mode (95)
Pattern suggests slight left skewness due to lower scores pulling mean down
Mean = 89.7, Median = 91, Mode = 95
• Mean sensitivity: Affected by all values, including outliers
• Median resistance: Only affected by middle values
• Mode uniqueness: May not exist or may have multiple values
Measures of spread: Statistical measures that indicate how spread out or dispersed the values in a dataset are.
Ordered data: 12, 15, 18, 20, 22, 25, 28, 30, 35, 40
Range = Maximum - Minimum = 40 - 12 = 28
Q1 (first quartile) = Median of lower half = (15 + 18) ÷ 2 = 16.5
Q2 (median) = (22 + 25) ÷ 2 = 23.5
Q3 (third quartile) = Median of upper half = (30 + 35) ÷ 2 = 32.5
IQR = Q3 - Q1 = 32.5 - 16.5 = 16
Mean = 24.5
Deviations: (-12.5)², (-9.5)², (-6.5)², (-4.5)², (-2.5)², (0.5)², (3.5)², (5.5)², (10.5)², (15.5)²
Variance = 570.5 ÷ 10 = 57.05
Standard deviation = √57.05 ≈ 7.55
| Measure | Value | Interpretation |
|---|---|---|
| Range | 28 | Total spread from min to max |
| IQR | 16 | Middle 50% of data spread |
| Std Dev | 7.55 | Avg distance from mean |
Range = Maximum value - Minimum value = 40 - 12 = 28
Lower half: 12, 15, 18, 20, 22 → Q1 = (15 + 18) ÷ 2 = 16.5
Upper half: 25, 28, 30, 35, 40 → Q3 = (30 + 35) ÷ 2 = 32.5
IQR = Q3 - Q1 = 32.5 - 16.5 = 16
First, find the mean: (12+15+18+20+22+25+28+30+35+40) ÷ 10 = 24.5
Then, find squared deviations from mean and average them
Standard deviation ≈ 7.55
Range (28) shows total spread across all data
IQR (16) focuses on the middle 50% of data
Standard deviation (7.55) indicates average deviation from mean
Range = 28, IQR = 16, Std Dev ≈ 7.55
• Range sensitivity: Affected only by extreme values
• IQR robustness: Resistant to outliers
• Standard deviation: Uses all data points, measures average variability
Outlier: A data value that is significantly different from other values in the dataset, potentially affecting statistical analysis.
Ordered data: 65, 68, 70, 72, 75, 78, 80, 82, 85, 90, 95, 105
Lower half: 65, 68, 70, 72, 75, 78
Upper half: 80, 82, 85, 90, 95, 105
Q1 = (70 + 72) ÷ 2 = 71
Q3 = (85 + 90) ÷ 2 = 87.5
IQR = Q3 - Q1 = 87.5 - 71 = 16.5
Lower fence = Q1 - 1.5(IQR) = 71 - 1.5(16.5) = 71 - 24.75 = 46.25
Upper fence = Q3 + 1.5(IQR) = 87.5 + 1.5(16.5) = 87.5 + 24.75 = 112.25
Values below 46.25 or above 112.25 are outliers
All values (65-105) are within fences, so NO OUTLIERS
| Measure | With 105 | Without 105 | Difference |
|---|---|---|---|
| Mean | 81.08 | 78.18 | +2.9 |
| Range | 40 | 30 | +10 |
| Std Dev | 12.3 | 10.6 | +1.7 |
65, 68, 70, 72, 75, 78, 80, 82, 85, 90, 95, 105
Lower half: 65, 68, 70, 72, 75, 78 → Q1 = (70 + 72) ÷ 2 = 71
Upper half: 80, 82, 85, 90, 95, 105 → Q3 = (85 + 90) ÷ 2 = 87.5
IQR = 87.5 - 71 = 16.5
Lower fence = 71 - 1.5(16.5) = 46.25
Upper fence = 87.5 + 1.5(16.5) = 112.25
Check each value: All values (65-105) are within [46.25, 112.25]
Conclusion: No outliers according to 1.5 × IQR rule
No outliers detected (all values within fences)
• 1.5 × IQR rule: Outliers are values beyond Q1-1.5(IQR) or Q3+1.5(IQR)
• Outlier impact: Can significantly affect mean and standard deviation
• Robust measures: Median and IQR are less affected by outliers
Mean: The arithmetic average of all values in a dataset.
Median: The middle value when data is arranged in order from least to greatest.
Mode: The value that appears most frequently in a dataset.
Range: The difference between the maximum and minimum values in a dataset.
Interquartile Range (IQR): The range of the middle 50% of data, calculated as Q3 - Q1.
- Organize data: Order values from least to greatest
- Calculate center: Find mean, median, and mode
- Calculate spread: Find range, IQR, and standard deviation
- Detect outliers: Use appropriate methods to identify unusual values
- Interpret results: Analyze how measures relate to each other
- Draw conclusions: Summarize findings about the dataset
Outlier impact: The effect that extreme values have on statistical measures, particularly those that use all data points in their calculation.
| Measure | Without 50 | With 50 | Change |
|---|---|---|---|
| Mean | 17.43 | 22.38 | +4.95 |
| Median | 18 | 18 | 0 |
| Mode | None | None | 0 |
| Range | 15 | 40 | +25 |
| IQR | 10 | 10 | 0 |
| Std Dev | 5.35 | 13.16 | +7.81 |
Mean = (10+12+15+18+20+22+25) ÷ 7 = 122 ÷ 7 ≈ 17.43
Median = 18 (4th value)
Range = 25 - 10 = 15
Mean = (10+12+15+18+20+22+25+50) ÷ 8 = 172 ÷ 8 = 21.5
Median = (18+20) ÷ 2 = 19
Range = 50 - 10 = 40
Without 50: 10, 12, 15, 18, 20, 22, 25
Mean = 122 ÷ 7 ≈ 17.43
Median = 18
Range = 25 - 10 = 15
With 50: 10, 12, 15, 18, 20, 22, 25, 50
Mean = 172 ÷ 8 = 21.5
Median = (18+20) ÷ 2 = 19
Range = 50 - 10 = 40
Mean increased significantly (+4.07)
Median changed slightly (+1)
Range increased dramatically (+25)
Standard deviation increased significantly
Mean is highly sensitive to outliers
Median is resistant to outliers
Range is extremely sensitive to outliers
IQR remains unchanged as it's resistant to outliers
Outlier increases mean from 17.43 to 21.5 (+23.3%)
• Mean sensitivity: Affected by every value in the dataset
• Median resistance: Only affected by middle values
• Range vulnerability: Depends entirely on extreme values
Comprehensive analysis: A thorough examination of data that includes measures of center, spread, shape, and any unusual features.
| Category | Measure | Value |
|---|---|---|
| Center | Mean | 82.11 |
| Median | 82 | |
| Mode | 100 | |
| Spread | Range | 45 |
| IQR | 25 | |
| Std Dev | 14.89 | |
| Outliers | None | Using 1.5×IQR rule |
Q1 = 70 (5th value), Q3 = 95 (14th value), IQR = 95 - 70 = 25
Lower fence = 70 - 1.5(25) = 32.5, Upper fence = 95 + 1.5(25) = 132.5
Mean (82.11) ≈ Median (82), suggesting approximate symmetry
Mode (100) is higher than center, indicating possible right tail
Three perfect scores create a mode at the maximum
Mean = 1478 ÷ 18 ≈ 82.11
Median = (82 + 82) ÷ 2 = 82 (average of 9th and 10th values)
Mode = 100 (appears 3 times)
Range = 100 - 55 = 45
Q1 = 70 (median of lower half), Q3 = 95 (median of upper half)
IQR = 95 - 70 = 25
Lower fence = Q1 - 1.5(IQR) = 70 - 1.5(25) = 32.5
Upper fence = Q3 + 1.5(IQR) = 95 + 1.5(25) = 132.5
All values (55-100) are within fences, so no outliers
Approximately symmetric (mean ≈ median)
Three students scored perfectly (mode at maximum)
Range of 45 points shows moderate variability
Distribution is approximately symmetric with no outliers
• Symmetry indicator: Mean ≈ Median suggests symmetric distribution
• Outlier detection: 1.5 × IQR rule identifies unusual values
• Mode significance: Repeated maximum values may indicate ceiling effect
Measures of center: Values that represent the typical or central value of a dataset (mean, median, mode).
Measures of spread: Values that indicate how spread out or dispersed the data is (range, IQR, standard deviation).
Outliers: Data values that are significantly different from other values in the dataset.
Quartiles: Values that divide the ordered dataset into four equal parts (Q1, Q2, Q3).
Interquartile Range (IQR): The range of the middle 50% of data, calculated as Q3 - Q1.
- Data preparation: Order the data from least to greatest
- Calculate center measures: Mean, median, and mode
- Calculate spread measures: Range, IQR, and standard deviation
- Identify quartiles: Q1, Q2 (median), Q3
- Detect outliers: Use 1.5 × IQR rule or other methods
- Analyze relationships: Compare measures to understand distribution
- Draw conclusions: Interpret results in context
| Measure | Sensitivity to Outliers | When to Use |
|---|---|---|
| Mean | Highly sensitive | Symmetric distributions without outliers |
| Median | Not sensitive | Skewed distributions or with outliers |
| Mode | Not sensitive | Categorical data or identifying common values |
| Range | Highly sensitive | Quick spread estimate (not recommended with outliers) |
| IQR | Not sensitive | Robust measure of spread |
| Standard Deviation | Moderately sensitive | Normal distributions or when all data matters |
Mean: The arithmetic average. Formula: Σx/n. Affected by all values, sensitive to outliers. Best for symmetric distributions.
Median: The middle value when ordered. Not affected by extreme values. Robust measure, good for skewed data.
Mode: The most frequent value. May not exist or may have multiple values. Useful for categorical data.
Range: Maximum - Minimum. Simple to calculate but highly sensitive to outliers.
Interquartile Range (IQR): Q3 - Q1. Measures middle 50% of data, resistant to outliers.
Standard Deviation: Average distance from mean. Uses all data points, affected by outliers.
1.5 × IQR Rule: Values below Q1 - 1.5(IQR) or above Q3 + 1.5(IQR) are considered outliers.
Impact: Outliers can significantly affect mean and standard deviation but have minimal effect on median and IQR.
Context: Always consider if outliers are valid data points or errors before removing them.
Questions & Answers
Question: I'm confused about when to use mean vs median. Why does it matter which one I choose?
Answer: The choice between mean and median depends on your data's distribution and presence of outliers:
Use the mean when:
- Your data is symmetric (bell-shaped)
- You want to account for every data point
- There are no significant outliers
- You need to perform further mathematical calculations
Use the median when:
- Your data is skewed (has a long tail)
- There are outliers that would distort the average
- You want a measure that represents a "typical" value
- You want a measure resistant to extreme values
Example: If household incomes in a neighborhood are $30K, $35K, $40K, $45K, $50K, $55K, $1,000K (millionaire), the mean would be $164K (misleading), but the median would be $45K (more representative of typical income).
Question: What's the difference between range and IQR? When should I use each one?
Answer: Both measure spread, but they do so differently:
Range:
- Calculated as: Maximum - Minimum
- Uses only two data points
- Highly sensitive to outliers
- Provides the total spread of all data
- Simple to calculate but limited in information
Interquartile Range (IQR):
- Calculated as: Q3 - Q1 (middle 50% of data)
- Uses the middle half of all data points
- Resistant to outliers
- Focuses on the core of the distribution
- More robust measure of spread
When to use: Use range for a quick, rough idea of spread. Use IQR when you want a more reliable measure that isn't affected by extreme values. IQR is generally preferred in statistical analysis.
Question: How do I know if a data point is really an outlier or just part of the normal variation?
Answer: There are both mathematical and contextual ways to determine outliers:
Mathematical approach (1.5 × IQR rule):
- Calculate Q1 (first quartile) and Q3 (third quartile)
- Compute IQR = Q3 - Q1
- Calculate Lower Fence = Q1 - 1.5(IQR)
- Calculate Upper Fence = Q3 + 1.5(IQR)
- Any value outside these fences is considered an outlier
Contextual considerations:
- Does the value make sense in the real-world context?
- Could it be a data entry error?
- Is it physically possible?
- Does it significantly affect your analysis?
Important note: The 1.5 × IQR rule is a guideline, not a strict rule. In some cases, extreme values are genuine and important. Always consider the context before deciding to exclude data points.