Chapter Overview
🧭Your visual roadmap through the chapter
Measures of Central Tendency
Mean, median, mode, geometric & harmonic mean
Measures of Dispersion
Range, IQR, variance, standard deviation, coefficient of variation
Measures Compared
Choosing the right average and dispersion measure; skewness; box plots
Statistical measures that summarize data's central value (averages) and its spread or variability (dispersion), supporting descriptive and predictive analysis.
Comparing datasets fairly requires knowing not just their typical value but also how consistent or spread out they are.
Calculate mean/median/mode for grouped and ungrouped data; compute variance, SD and coefficient of variation; construct and read box plots and ogives.
Key Definitions
📘Source-based terminology, exam-ready
Descriptive Statistics
The part of statistics that describes or summarizes information from data, e.g. via central value or dispersion.
Key word: describe/summarize
Arithmetic Mean
The sum of the values in a series divided by the count of values — a mathematical representation of the typical value.
Key word: sum ÷ count
Median
The middle value when data is arranged in ascending or descending order.
Key word: middle value
Mode
The most frequently occurring value (or the highest frequency in grouped data). There may be more than one mode, or none at all.
Key word: most frequent
Geometric Mean
Useful when several quantities are multiplied together to arrive at a product, e.g. average growth rate over several years.
Key word: average growth rate
Harmonic Mean
The number of observations divided by the sum of the reciprocals of each value; useful for average speeds.
Key word: average speed
Variance & Standard Deviation
Variance is the average of the squared deviations from the mean. Standard deviation is its square root.
Key word: SD = √variance
Coefficient of Variation
A relative measure of dispersion expressing standard deviation as a percentage of the mean, used to compare variability between datasets.
Key word: relative spread
Core Concept
🎯Explained in exam-oriented language
CONCEPT → Central Tendency
MEANING
A single value that represents the "centre" or typical value of a dataset.
PURPOSE
Compare two or more datasets using the same units.
APPLICATION
Choosing mean, median, mode, geometric or harmonic mean depending on the data's nature.
CONCEPT → Dispersion
MEANING
How spread out or clustered the data is around its central value.
PURPOSE
Reveal consistency that measures of central tendency alone cannot show (e.g. two batsmen with the same average).
APPLICATION
Absolute measures (range, IQR, variance, SD) vs relative measures (coefficient of variation).
Key Rules & Principles
📏Every source rule, exam-ready
RULE 1 Properties of an ideal average
Well defined; based on all values; capable of further algebraic calculation
Not affected by extreme values; stable in repeated sampling
RULE 3 Properties of variance & standard deviation
SD and variance of a constant value are always zero
Independent of a change of origin: SD(x±C)=SD(x)
Affected by a change of scale: SD(Cx)=|C|×SD(x); Var(Cx)=C²×Var(x)
If symmetrical: 68.27% of values lie within ±1 SD, 95.45% within ±2 SD, 99.73% within ±3 SD of the mean
RULE 2 Properties of the arithmetic mean
Sum of deviations from the mean is always zero; sum of squared deviations is always least
Combining group means requires the weighted combined-mean formula, NOT a simple average of the means
Mean of constant values is that constant; mean is affected by a change of origin or scale (Mean(x±c)=Mean(x)±c; Mean(cx)=c×Mean(x))
RULE 4 Empirical relationship & skewness
(Mean − Mode) = 3 × (Mean − Median)
Perfectly symmetric: mean = median
Skewed left/negative: mean < median | Skewed right/positive: mean > median
Formula Zone — Central Tendency
🧮Every formula supported by the source text
MEDIAN (UNGROUPED) & MEAN
Mean formula on the right applies to grouped data using class mid-points (x).
MEDIAN (GROUPED)
L=lower boundary of median class, cf=cumulative frequency before it, f=its frequency, h=its width.
MODE (GROUPED)
fm=frequency of modal class, f1=class before, f2=class after.
COMBINED MEAN & WEIGHTED MEAN
Σwx÷Σw
Never average two group means directly — weight by each group's size (n) or importance (w).
GEOMETRIC MEAN
Use (1+growth rate) as each x for compounding growth rates over n periods.
HARMONIC MEAN
Number of items divided by the sum of the reciprocals of each value. Useful for average speed/rate problems.
Formula Zone —
🧮DISPERSION Every formula supported by the source text
RANGE & INTER-QUARTILE RANGE
For grouped data, range = upper limit of last class − lower limit of first class.
VARIANCE & STANDARD DEVIATION
Second variance formula (grouped) avoids computing each deviation individually.
COEFFICIENT OF VARIATION
Lower CV = greater precision/consistency. Used to compare datasets with different units or scales.
PEARSON'S COEFFICIENT OF SKEWNESS
Sk=0: symmetric. Positive Sk: right-skewed. Negative Sk: left-skewed.
Calculation Roadmap
🗺️General path for any statistical measure question
1· Read Question
Grouped or ungrouped? What measure is needed?
2· Organize Data
Sort, tabulate class/frequency/mid-point
3· Select Formula
Match the measure and data type
4· Substitute
Put known values into the formula
5· Calculate & Check
Solve, recheck, state final answer
Step-by-step Methods
🪜Kept separate — do not mix methods
METHOD A · Median (Grouped Data)
- 1Build the cumulative frequency column; find Σf÷2
- 2Identify the median class (where the cf first reaches/exceeds Σf÷2)
- 3Note L, cf (before median class), f and h for that class
- 4Substitute into Median=L+[(Σf/2−cf)/f]×h
METHOD D · Geometric Mean
- 1Express each period's growth as (1+rate), or note start/end values
- 2Multiply all growth factors together (or divide end value by start value)
- 3Take the n-th root, where n=number of periods
- 4Subtract 1 and convert to a percentage for the average growth rate
METHOD B · Mode (Grouped Data)
- 1Identify the modal class (highest frequency)
- 2Note L, fm (modal class), f1 (class before), f2 (class after), and h
- 3Substitute into Mode=L+[(fm−f1)/((fm−f1)+(fm−f2))]×h
- 4Alternative: draw the histogram and read off where the two diagonal lines cross
METHOD E · Variance & Standard Deviation
- 1Ungrouped: find x̄, then each (x−x̄) and square it
- 2Sum the squared deviations and divide by n for variance
- 3Grouped: use Σfx²/Σf − (Σfx/Σf)² instead
- 4Take the square root of variance for standard deviation
METHOD C · Mean (Grouped / Weighted / Combined)
- 1Grouped: find each class mid-point (x), multiply by frequency (f)
- 2Sum fx and f, then divide (Σfx÷Σf)
- 3Weighted: multiply each value by its weight, sum, divide by Σw
- 4Combined: use each group's mean AND size in the combined-mean formula
METHOD F · Coefficient of Variation & Skewness
- 1Calculate the mean and SD for each dataset being compared
- 2CV = (SD÷mean)×100 for each; lower CV = more consistent
- 3Skewness: find mean, median and SD
- 4Substitute into Sk=3(Mean−Median)/SD; interpret the sign
Different Cases / Scenarios
🔀Recognise the situation, pick the right measure
IF you want the most commonly-used, calculation-friendly average
→ USE the arithmetic mean
IF the distribution is skewed / has extreme values
→ USE the median (not affected by extremes)
IF you want the single most frequently occurring value
→ USE the mode
IF averaging growth/decline rates or ratios over time
→ USE the geometric mean
IF averaging a rate like speed (distance per constant time)
→ USE the harmonic mean
IF some values matter more than others
→ USE the weighted mean
IF combining the means of two different-sized groups
→ USE the combined mean formula — never just average the two means
IF comparing variability between datasets with different units/scales
→ USE the coefficient of variation (relative), not raw SD
Comparison Tables
⚖️Don't confuse these!
| A.M. | G.M. | H.M. | Median | Mode | |
|---|---|---|---|---|---|
| Uses all values? | Yes | Yes | Yes | Middle value only | Highest-frequency value only |
| Effect of extremes | Greatest effect | Less than A.M. | Least effect | No effect | No effect |
| Zero/negative value | Can calculate | Cannot | Cannot | Can calculate | Can calculate |
| Typical usage | Default choice | Growth/decline rates | Rates like speed | Skewed/asymmetric data | Most frequent value needed |
| Measure | Advantage | Disadvantage |
|---|---|---|
| Mean | Uses every value; used extensively elsewhere | Distorted by extreme values in small samples |
| Median | Not influenced by extreme values; an actual value (odd n) | Little use in other statistical applications |
| Mode | Is an actual value | Little use in other applications |
| Range | Easy to calculate and understand | Distorted by extreme figures |
| Semi-IQR | Easy to calculate; ignores extremes | Doesn't use all figures; no insight into clustering |
| Standard Deviation | Uses all values; useful in further applications | Harder to understand/calculate by hand |
Reading Ogives & Box Plots
📌How to interpret the source's graphical tools
Reading an ogive for median/quartiles
The median is the value at Σf÷2 on the cumulative-frequency axis; Q1 and Q3 are read at Σf×25% and Σf×75% respectively, tracing across to the curve then down to the value.
Reading a box-and-whisker plot
The box spans Q1 to Q3 (the middle 50%); the line inside is the median. Whiskers extend from the box to the minimum and maximum values in the data.
Worked Calculation Logic
✍️Method students must reproduce in the exam
GIVEN → FORMULA → SUBSTITUTION → CALCULATION → ANSWER
Group A: 1,200 students, mean score 60. Group B: 400 students, mean score 70. Find the combined mean.
GIVEN: MeanA=60, nA=1,200; MeanB=70, nB=400
FORMULA: Combined Mean = (MeanAnA+MeanBnB)÷(nA+nB)
SUBSTITUTION: = (60×1,200 + 70×400) ÷ (1,200+400) = (72,000+28,000)÷1,600
FINAL ANSWER: Combined mean = 100,000÷1,600 = 62.5 (NOT the simple average of 65)
Exam Question Patterns
📝Direct calculation — mean/median/mode, grouped or ungrouped
Geometric/harmonic mean — growth rates or average speed
Combined/weighted mean — merging groups correctly
Dispersion calculation — range, IQR, variance, SD, CV
Comparison — which dataset is more consistent (CV)
Graph/box-plot reading and skewness identification
Exam Traps
⚠️⚠ WATCH OUT
- ✗ Averaging two group means directly — always use the combined-mean formula weighted by group size.
- ✗ Variance vs SD mix-up — forgetting to take the square root of variance to get SD.
- ✗ Using arithmetic mean for growth rates instead of the geometric mean.
- ✗ Wrong median class — picking the class where cf first reaches, not exceeds, Σf/2 incorrectly.
- ✗ Misreading CV direction — a LOWER coefficient of variation means MORE consistency, not less.
Quick Decision Tree
🌳Chapter 8 in 60 Seconds
⚡Median pos: (n+1)/2
Mean: Σx/n or Σfx/Σf
Mode: L+[(fm−f1)/((fm−f1)+(fm−f2))]×h
G.M.: ⁿ√(x1×x2...xn)
H.M.: n/Σ(1/x)
Variance: Σ(x−x̄)²/n; SD=√variance CV = (SD/mean)×100
Trap: combined mean ≠ simple average
Test yourself on this chapter
Practice 100 MCQs per chapter with answers – Chapter 1 is free for everyone. Enrol on WhatsApp for all 12 chapters, classes and tests on our student portal.