Statistics

Meaning, Need and Real-Life Significance

Measures of central tendency (mean, median, mode) provide only a single representative value, but real-world datasets require deeper analysis. Two datasets may have identical mean yet behave very differently in terms of spread. This variability is captured using dispersion.

Dispersion quantifies the degree of scatter of observations around a central value. It is fundamental in data science, economics, physics experiments, error analysis, and competitive exams.

For IIT-JEE and NEET aspirants, dispersion is crucial in:

• Error and uncertainty analysis in Physics
• Data interpretation (DI) questions
• Statistical reasoning in advanced problems
• Comparing consistency of datasets

Formal Definition

A measure of dispersion is a statistical quantity that describes the spread of data values relative to a central value such as mean or median.

Mathematically, dispersion represents the extent of deviation:

Dispersion ∝ Deviation of observations from central value

Characteristics of a Good Measure of Dispersion

• Clearly and rigidly defined
• Simple to compute and interpret
• Based on maximum observations
• Not overly affected by extreme values (outliers)
• Suitable for algebraic manipulation
• Stable across different samples

In competitive exams, standard deviation is preferred because it satisfies most of these properties.

Types of Dispersion

Dispersion is broadly classified into:

• Absolute Measures – expressed in same units as data
• Relative Measures – expressed as ratios or percentages

Absolute Measures of Dispersion

1. Range = Largest value − Smallest value
2. Quartile Deviation = (Q₃ − Q₁) / 2
3. Mean Deviation = Average of absolute deviations
4. Standard Deviation = Root mean square deviation

Illustrative Example (Board + JEE Level)

Consider two datasets:

A = {10, 10, 10, 10, 10}
B = {5, 8, 10, 12, 15}

Both have the same mean = 10, but:

• Dataset A → No variation (zero dispersion)
• Dataset B → High variation (large dispersion)

This clearly shows why dispersion is necessary beyond averages.

Advanced Insight for IIT / NEET Aspirants

Standard deviation plays a key role in:

• Normal distribution and probability
• Error propagation in experiments
• Statistical mechanics (Physics)
• Machine learning and data variance

Important identity frequently used in exams:

Variance = E(X²) − [E(X)]²

Quick Revision Points

• Dispersion measures variability, not central value
• Same mean ≠ same distribution
• Range is simplest but least reliable
• Standard deviation is most important for exams

Definition and Concept

Range is the simplest measure of dispersion and represents the spread of a dataset using only two extreme observations. It is defined as the difference between the largest and smallest values.

\[ \mathrm{Range} = L - S \]

where L is the largest value and S is the smallest value in the dataset.

Range gives a quick snapshot of variability but does not reflect how values are distributed between extremes.

Visual Understanding of Range

Formulas (Ungrouped & Grouped Data)

For individual data:

\[ \mathrm{Range} = \max(x_i) - \min(x_i) \]

For continuous frequency distribution:

\[ \mathrm{Range} = \text{Upper boundary of highest class} - \text{Lower boundary of lowest class} \]

Illustrative Example (Board + JEE Level)

Find the range of the dataset:

5, 8, 12, 15, 20

Here, Largest value (L) = 20 and Smallest value (S) = 5

Range = 20 − 5 = 15

Interpretation: The dataset spans 15 units, indicating moderate dispersion.

Coefficient of Range (Relative Measure)

Since range depends on units, we use a unit-free measure for comparison:

\[ \mathrm{Coefficient\ of\ Range} = \frac{L - S}{L + S} \]

This allows comparison of variability across different datasets or units.

Example:

L = 50, S = 10

Coefficient = (50 − 10) / (50 + 10) = 40 / 60 = 0.67

Merits of Range

• Extremely simple and easy to compute
• Provides a quick estimate of spread
• Useful in preliminary data analysis
• Applied in quality control and weather reports

Limitations of Range

• Depends only on two extreme values
• Highly sensitive to outliers
• Ignores distribution of intermediate values
• Not suitable for detailed statistical analysis

Advanced Insight for Competitive Exams

• Range is often used in quick elimination in MCQs
• Useful in comparing variability rapidly in DI sets
• Appears in assertion-reason type questions
• Acts as a base concept before studying variance and standard deviation

Shortcut: If max and min are same → Range = 0 (no dispersion)

Definition and Core Idea

Quartile Deviation (Q.D.), also known as the Semi-Interquartile Range, measures the dispersion of the central 50% of a dataset. It is based on quartiles and is less affected by extreme values.

\[ \mathrm{Quartile\ Deviation} = \frac{Q_3 - Q_1}{2} \]

where:

• Q₁ = First Quartile (25th percentile)
• Q₃ = Third Quartile (75th percentile)

It effectively ignores extreme values and focuses on the middle portion of the data.

Visual Interpretation (Middle 50% Spread)

Steps to Calculate Quartile Deviation

1. Arrange data in ascending order
2. Find Q₁ (position = (n+1)/4)
3. Find Q₃ (position = 3(n+1)/4)
4. Apply formula: (Q₃ − Q₁) / 2

Note: For grouped data, quartiles are obtained using cumulative frequency.

Illustrative Example (Board + JEE Level)

Find Quartile Deviation of the dataset:

2, 4, 6, 8, 10, 12, 14, 16

Total observations (n) = 8

Q₁ position = (8+1)/4 = 2.25 → approx 4
Q₃ position = 3(8+1)/4 = 6.75 → approx 12

Quartile Deviation = (12 − 4) / 2 = 4

Coefficient of Quartile Deviation

This is a relative measure used to compare variability across datasets.

\[ \mathrm{Coefficient\ of\ Q.D.} = \frac{Q_3 - Q_1}{Q_3 + Q_1} \]

Example:

Q₁ = 10, Q₃ = 30

Coefficient = (30 − 10) / (30 + 10) = 20 / 40 = 0.5

Merits

• Not affected significantly by extreme values (robust measure)
• Suitable for skewed distributions
• Useful in open-ended class intervals
• Represents central data spread effectively

Limitations

• Ignores 50% of observations
• Not suitable for algebraic manipulation
• Less precise compared to standard deviation

Advanced Insight for IIT / NEET Aspirants

• Closely related to box plots and interquartile range (IQR)
• Used in detecting outliers: values beyond Q₁ − 1.5(IQR) or Q₃ + 1.5(IQR)
• Important in data interpretation and statistical reasoning
• Forms conceptual base for robust statistics

Shortcut: If Q₁ = Q₃ → Quartile Deviation = 0 (no spread in middle data)

Definition and Interpretation

Mean Deviation (M.D.) is the average of the absolute deviations of observations from a central value such as mean or median. It provides a more representative measure of dispersion than range and quartile deviation because it considers all observations.

\[ \mathrm{Mean\ Deviation\ about\ }A = \frac{1}{n}\sum |x_i - A| \]

where A can be mean or median.

The use of absolute values ensures that negative and positive deviations do not cancel out.

Visual Understanding of Mean Deviation

Types of Mean Deviation

Mean Deviation about Mean

\[ \mathrm{M.D.}(\overline{x}) = \frac{1}{n}\sum |x_i - \overline{x}| \]

Mean Deviation about Median

\[ \mathrm{M.D.}(M) = \frac{1}{n}\sum |x_i - M| \]

Important: Mean deviation is minimum when calculated about the median.

Step-by-Step Calculation

1. Find the mean or median of the dataset
2. Calculate deviations (xᵢ − A)
3. Take absolute values |xᵢ − A|
4. Find their sum
5. Divide by total observations (n)

Illustrative Example (Board + JEE Level)

Find mean deviation about mean for the dataset:

2, 4, 6, 8, 10

Mean = (2+4+6+8+10)/5 = 6

Deviations: |2−6|=4, |4−6|=2, |6−6|=0, |8−6|=2, |10−6|=4

Sum of deviations = 12

Mean Deviation = 12 / 5 = 2.4

Mean Deviation for Frequency Distribution

\[ \mathrm{M.D.} = \frac{1}{N}\sum f_i |x_i - A| \]

where:

• fᵢ = frequency
• xᵢ = class mark
• N = total frequency

Merits

• Based on all observations
• Simple to understand and compute
• Less affected by extreme values compared to standard deviation
• Provides better representation than range

Limitations

• Absolute values make algebraic treatment difficult
• Less precise than standard deviation
• Not widely used in higher statistics

Advanced Insight for IIT / NEET Aspirants

• Median minimizes mean deviation → useful theoretical result
• Used in robust statistics and error estimation
• Frequently appears in assertion-reason and conceptual MCQs
• Acts as bridge between quartile deviation and standard deviation

Key Insight: Sum of absolute deviations is minimum about median

Why Grouped Data Needs Special Treatment

In real-life datasets (JEE/NEET level), values are often grouped into frequencies or class intervals. Instead of working with individual observations, we use frequency-weighted averages.

The idea remains the same: measure average deviation from a central value, but now each observation contributes according to its frequency.

Discrete Frequency Distribution

A discrete distribution consists of distinct values \(x_i\) with corresponding frequencies \(f_i\).

\[ \overline{x} = \frac{\sum f_i x_i}{\sum f_i} = \frac{1}{N}\sum f_i x_i \]

\[ \mathrm{M.D.}(\overline{x}) = \frac{1}{N} \sum f_i |x_i - \overline{x}| \]

Solved Example (Board + JEE Level)

Given data:

\(N = 6\)

\(\overline{x} = (2×1 + 4×2 + 6×2 + 8×1)/6 = 30/6 = 5\)

Deviations: |2−5|=3, |4−5|=1, |6−5|=1, |8−5|=3

Weighted sum = 1×3 + 2×1 + 2×1 + 1×3 = 10

Mean Deviation = 10 / 6 = 1.67

Mean Deviation about Median

Steps:

• Find cumulative frequency
• Locate median using \(N/2\)
• Apply formula:

\[ \mathrm{M.D.}(M) = \frac{1}{N} \sum f_i |x_i - M| \]

Continuous Frequency Distribution

In continuous data, values are grouped into class intervals. We assume all values in a class are concentrated at the class midpoint (xᵢ).

\[ x_i = \frac{\text{lower limit + upper limit}}{2} \]

\[ \mathrm{M.D.} = \frac{1}{N} \sum f_i |x_i - A| \]

Shortcut Method (Highly Important for JEE)

To reduce calculations, we use assumed mean \(a\) and class width \(h\).

\[ d_i = \frac{x_i - a}{h} \]

\[ \overline{x} = a + \frac{\sum f_i d_i}{N} \times h \]

\[ \mathrm{M.D.} = \frac{\sum f_i |d_i|}{N} \times h \]

This method saves time and avoids large calculations in MCQs.

Mean Deviation about Median (Continuous Case)

First find median using:

\[ \text{Median} = l + \frac{\frac{N}{2} - C}{f} \times h \]

• l = lower boundary of median class
• f = frequency of median class
• C = cumulative frequency before median class
• h = class width

Then apply:

\[ \mathrm{M.D.}(M) = \frac{1}{N} \sum f_i |x_i - M| \]

Concept Visualization

Exam Strategy (Boards + JEE + NEET)

• Always prefer median when data has outliers
• Use shortcut method in MCQs to save time
• Carefully compute class midpoints in continuous data
• Watch absolute values – common mistake area

High-weight concept in CBSE boards and appears frequently in JEE Main statistics questions.

Definition and Statistical Importance

Standard Deviation (σ) is the most powerful and widely used measure of dispersion. It measures the average spread of data from the mean by considering the square of deviations.

\[ \sigma = \sqrt{\frac{1}{N}\sum (x_i - \overline{x})^2} \]

It removes the issue of sign cancellation (unlike mean deviation) and gives more weight to larger deviations.

Key Insight: Larger spread → Larger σ → More variability

Visual Interpretation

Formulas (All Cases)

Ungrouped Data

\[ \sigma = \sqrt{\frac{1}{N}\sum (x_i - \overline{x})^2} \]

Discrete Frequency Distribution

\[ \sigma = \sqrt{\frac{1}{N}\sum f_i (x_i - \overline{x})^2} \]

Shortcut (Computational Formula)

\[ \sigma = \sqrt{\frac{\sum f_i x_i^2}{N} - \left(\frac{\sum f_i x_i}{N}\right)^2} \]

This is the most used formula in competitive exams.

Step-by-Step Calculation (Exam Oriented)

1. Find mean \(\overline{x}\)
2. Compute deviations \(x_i - \overline{x}\)
3. Square each deviation
4. Multiply by frequency (if any)
5. Find average and take square root

Solved Example (Board + JEE Level)

Find standard deviation of:

2, 4, 6, 8

Mean = (2+4+6+8)/4 = 5

Deviations: -3, -1, 1, 3

Squares: 9, 1, 1, 9 → Sum = 20

σ = √(20/4) = √5 ≈ 2.236

Continuous Frequency Distribution

Use class midpoints \(x_i\) and same formulas:

\[ \sigma = \sqrt{\frac{1}{N}\sum f_i (x_i - \overline{x})^2} \]

Step-Deviation Method (Very Important for JEE)

Choose assumed mean \(a\), class width \(h\):

\[ d_i = \frac{x_i - a}{h} \]

\[ \sigma = h \sqrt{\frac{\sum f_i d_i^2}{N} - \left(\frac{\sum f_i d_i}{N}\right)^2} \]

Reduces heavy calculations significantly in MCQs.

Important Properties (Highly Tested)

• σ ≥ 0 always
• σ = 0 → all values equal
• Independent of change of origin
• Affected by change of scale

If \(y = \frac{x - a}{b}\), then σ_y = σ_x/b

Variance (σ²)

Variance is the square of standard deviation:

\[ \sigma^2 = \frac{1}{N}\sum (x_i - \overline{x})^2 \]

Shortcut identity: σ² = E(X²) − [E(X)]²

Advanced Insight for IIT / NEET Aspirants

• Foundation of Normal Distribution
• Used in error analysis in Physics
• Core concept in probability & statistics
• Appears frequently in JEE Main & Advanced

68–95–99 rule (conceptual): Most values lie within few σ of mean

Exam Strategy

• Prefer shortcut formula in MCQs
• Use symmetry to reduce calculations
• Watch square terms carefully
• Memorize variance identity

Find the mean deviation about the mean for the following data:

6, 7, 10, 12, 13, 4, 8, 12

Visual Insight

Solution

First, calculate the mean:

\[ \overline{x} = \frac{6+7+10+12+13+4+8+12}{8} = \frac{72}{8} = 9 \]

Now compute deviations and absolute deviations:

Sum of absolute deviations:

\[ \sum |x - \overline{x}| = 22 \]

Mean deviation about mean:

\[ \mathrm{M.D.}(\overline{x}) = \frac{22}{8} = \frac{11}{4} = 2.75 \]

Final Answer

Mean Deviation about Mean = 2.75

Exam Insight (Boards + JEE + NEET)

• Always convert deviations to absolute values
• Watch arithmetic errors in summation
• Decimal answer is acceptable unless fraction asked
• This type is frequently asked in CBSE boards

Shortcut Tip: If values are symmetric around mean → M.D. becomes easier

Find the mean deviation about the mean for the following data:

12, 3, 18, 17, 4, 9, 17, 19, 20, 15, 8, 17, 2, 3, 16, 11, 3, 1, 0, 5

Smart Strategy (For Large Data)

• Compute mean first
• Group similar values mentally
• Avoid sign mistakes → directly take absolute values
• Check symmetry around mean (here mean = 10 simplifies work)

Solution

Total number of observations:

N = 20

Calculate mean:

\[ \overline{x} = \frac{\sum x_i}{N} = \frac{200}{20} = 10 \]

Now compute absolute deviations:

Sum of absolute deviations:

\[ \sum |x - \overline{x}| = 124 \]

Mean deviation:

\[ \mathrm{M.D.}(\overline{x}) = \frac{124}{20} = 6.2 \]

Concept Visualization

Final Answer

Mean Deviation about Mean = 6.2

Exam Insight (Boards + JEE + NEET)

• Large datasets test calculation accuracy
• Mean often chosen as integer to simplify work
• Expect MCQs where only final value matters
• Symmetry around mean can reduce effort

Pro Tip: If data repeats patterns, group them mentally to speed up summation

Find the mean deviation about the median for the following data:

3, 9, 5, 3, 12, 10, 18, 4, 7, 19, 21

Key Concept

Mean deviation about median requires:

• Sorting the data
• Finding median correctly
• Taking absolute deviations from median

Important: Median minimizes total absolute deviation → very important theoretical result

Solution

Arrange data in ascending order:

\[ 3,\,3,\,4,\,5,\,7,\,9,\,10,\,12,\,18,\,19,\,21 \]

Total observations:

N = 11

Median position:

\[ \frac{N+1}{2} = \frac{12}{2} = 6 \]

Median = 6th observation = 9

Now compute absolute deviations:

Sum of absolute deviations:

\[ \sum |x - M| = 58 \]

Mean deviation about median:

\[ \mathrm{M.D.}(M) = \frac{58}{11} \approx 5.27 \]

Visual Interpretation

Final Answer

Mean Deviation about Median = 5.27

Exam Insight (Boards + JEE + NEET)

• Always sort data before finding median
• Check odd/even case carefully
• Median-based deviation is preferred when outliers exist
• Frequently asked in CBSE board exams

Pro Tip: Median gives minimum absolute deviation → often used in proofs and MCQs

Find mean deviation about the mean for the following data:

Method Strategy (Highly Important for Boards + JEE)

• Use tabular approach to avoid calculation errors
• Compute mean first using Σfᵢxᵢ
• Then calculate |xᵢ − x̄| and multiply with fᵢ

Solution Table

Step 1: Calculate Mean

\[ N = \sum f_i = 40 \]

\[ \overline{x} = \frac{\sum f_i x_i}{N} = \frac{300}{40} = 7.5 \]

Step 2: Mean Deviation Calculation

\[ \mathrm{M.D.}(\overline{x}) = \frac{1}{N}\sum f_i |x_i - \overline{x}| \]

\[ \mathrm{M.D.} = \frac{92}{40} = \frac{23}{10} = 2.3 \]

Visual Interpretation

Final Answer

Mean Deviation about Mean = 2.3

Exam Insight

• Tabular method is safest for board exams
• Always compute Σfᵢxᵢ carefully
• Check totals before final division
• Very common CBSE + JEE Main question pattern

Pro Tip: Keep mean as fraction (7.5) to avoid rounding errors

Find the mean deviation about the median for the following data:

Method Strategy

• Use cumulative frequency to find median
• Handle even N carefully
• Then apply tabular method for deviation

Step 1: Cumulative Frequency Table

Step 2: Find Median

\[ N = \sum f_i = 30 \]

Since N is even:

Median = average of 15th and 16th observations

From cumulative frequency: both 15th and 16th observations lie at 13

\[ M = 13 \]

Step 3: Mean Deviation Table

Step 4: Final Calculation

\[ \mathrm{M.D.}(M) = \frac{1}{N} \sum f_i |x_i - M| \]

\[ \mathrm{M.D.} = \frac{149}{30} \approx 4.97 \]

Visual Insight

Final Answer

Mean Deviation about Median = 4.97

Exam Insight

• Even N → always check two middle positions
• Median from cumulative frequency is critical
• Tabular method ensures accuracy
• Very common CBSE board + JEE Main question

Pro Tip: Median-based deviation is best when data is skewed

Find the mean deviation about the mean for the following grouped data:

Concept Strategy (Continuous Data)

• Convert class intervals → class marks (midpoints)
• Apply discrete frequency formulas
• Use tabular method for accuracy

Step 1: Complete Table

Step 2: Calculate Mean

\[ N = 40 \]

\[ \overline{x} = \frac{1800}{40} = 45 \]

Central class (40–50) naturally contains the mean → symmetric distribution hint

Step 3: Mean Deviation

\[ \mathrm{M.D.}(\overline{x}) = \frac{1}{N} \sum f_i |x_i - \overline{x}| \]

\[ \mathrm{M.D.} = \frac{400}{40} = 10 \]

Graphical Insight

Final Answer

Mean Deviation about Mean = 10

Exam Insight (Boards + JEE + NEET)

• Always convert intervals → midpoints correctly
• Symmetry reduces calculation effort
• Table method is scoring in CBSE exams
• Frequently asked in Statistics chapter

Pro Tip: If distribution is symmetric → mean lies at center → deviations mirror

Calculate the mean deviation about median for the following data:

Concept Strategy

• Find median using cumulative frequency
• Identify median class correctly
• Use class marks for deviation calculation

Step 1: Complete Table

Step 2: Find Median

\[ N = 50,\quad \frac{N}{2} = 25 \]

Median class = 20–30 (since cumulative frequency just exceeds 25)

\[ M = l + \frac{\frac{N}{2} - c}{f} \times h \]

\[ M = 20 + \frac{25 - 13}{15} \times 10 = 20 + 8 = 28 \]

Step 3: Mean Deviation

\[ \mathrm{M.D.}(M) = \frac{1}{N} \sum f_i |x_i - M| \]

\[ \mathrm{M.D.} = \frac{508}{50} = 10.16 \]

Graphical Insight

Final Answer

Mean Deviation about Median = 10.16

Exam Insight

• Median class identification is critical
• Always use correct lower boundary (l)
• Use class marks for deviation
• Very high probability question in boards

Pro Tip: Errors usually occur in median formula substitution

Find the variance of the following data:

6, 8, 10, 12, 14, 16, 18, 20, 22, 24

Key Insight (Very Important)

The data forms an Arithmetic Progression (A.P.) symmetric about the center. Hence, mean lies at the center → simplifies calculations.

Symmetry Trick: Mean = middle value = 15

Variance Formula

\[ \sigma^2 = \frac{1}{N} \sum (x_i - \overline{x})^2 \]

Step-Deviation Table

Step 1: Mean

\[ \overline{x} = 15 \]

(Either by symmetry or calculation)

Step 2: Variance

\[ \sigma^2 = \frac{330}{10} = 33 \]

Step 3: Standard Deviation

\[ \sigma = \sqrt{33} \approx 5.74 \]

Visual Symmetry Insight

Final Answer

Variance = 33
Standard Deviation ≈ 5.74

Exam Insight (High Value)

• AP data → always check symmetry
• Mean often lies at center
• Squares increase rapidly → careful calculation
• Very common in JEE Main MCQs

Pro Tip: Symmetric data → Σ(x − x̄) = 0 automatically

Calculate the mean, variance and standard deviation for the following distribution:

Concept Strategy (High Scoring)

• Convert classes → class marks (xᵢ)
• Use step-deviation method (saves time)
• Choose assumed mean from central class

Here: a = 65, h = 10

Step 1: Working Table

Step 2: Mean

\[ \overline{x} = a + \frac{\sum f_i d_i}{N} \times h \]

\[ \overline{x} = 65 + \frac{-15}{50} \times 10 = 62 \]

Step 3: Variance

\[ \sigma^2 = \frac{1}{N} \sum f_i (x_i - \overline{x})^2 \]

\[ \sigma^2 = \frac{10050}{50} = 201 \]

Step 4: Standard Deviation

\[ \sigma = \sqrt{201} \approx 14.18 \]

Graphical Insight

Final Answer

Mean = 62
Variance = 201
Standard Deviation ≈ 14.18

Exam Insight (Very Important)

• Always choose central class as assumed mean
• Step-deviation reduces heavy calculations
• Variance questions are common in JEE Main
• Watch sign of Σfᵢdᵢ carefully

Pro Tip: If Σfᵢdᵢ ≈ 0 → mean ≈ assumed mean

Find the standard deviation for the following data:

Concept Strategy

• Use shortcut formula (computational formula)
• Compute Σfᵢxᵢ and Σfᵢxᵢ²
• Avoid direct deviation method (saves time)

Step 1: Correct Working Table

Correction: Missing value was f×x² = 640 (for x = 8)

Step 2: Apply Shortcut Formula

\[ \sigma = \sqrt{\frac{\sum f_i x_i^2}{N} - \left(\frac{\sum f_i x_i}{N}\right)^2} \]

Step 3: Calculation

\[ \frac{\sum f_i x_i^2}{N} = \frac{9652}{48} = 201.08 \]

\[ \left(\frac{\sum f_i x_i}{N}\right)^2 = \left(\frac{614}{48}\right)^2 = (12.79)^2 = 163.58 \]

\[ \sigma = \sqrt{201.08 - 163.58} = \sqrt{37.5} \approx 6.12 \]

Visual Insight

Final Answer

Standard Deviation ≈ 6.12

Exam Insight

• Shortcut formula is fastest in MCQs
• Always verify table values (common error area)
• Avoid rounding too early
• Highly expected in JEE Main & Boards

Pro Tip: Use σ² = E(X²) − [E(X)]² for fastest solving