सांख्यिकी

हिंदी में पढ़ें

Statistics — Turning Numbers into Answers

What You Will Master in This Chapter

By the end of this chapter, you will be able to:

1. Compute and interpret mean, median, and mode for raw, discrete, and grouped data.
2. Measure spread using range, variance, and standard deviation — and know when each matters.
3. Use the coefficient of variation to compare datasets with different units.
4. Handle combined means, missing-frequency problems, and weighted averages — the exam favourites.
5. Understand skewness and the empirical relationship between mean, median, and mode.
6. Spot the traps examiners set in SSC CGL Tier-II, IBPS PO Mains, CAT, and CSAT statistics questions.

The Hook: The Average That Almost Sank an Air Force

In the 1950s, the US Air Force designed cockpits based on the average body dimensions of thousands of pilots. The result? The cockpit fit almost nobody. Hardly any pilot was actually "average" on all measurements. The average — the most trusted number in statistics — had failed completely.

That is the lesson of this chapter: statistics is not about computing numbers; it is about knowing what a number hides. A mean of 50 can come from (48, 52) or from (0, 100). Same average, wildly different reality. Exams like SSC CGL Tier-II and IBPS PO Mains test exactly this — not whether you can divide, but whether you understand what the division *means*.

"Statistics Is Just Averages, Right?" — The Struggle Is Real

Most aspirants feel statistics is formula-mugging: memorise σ = √(Σx²/n), plug, chug, move on. Then the exam asks: *"The mean of 40 observations was 36. Later it was found that one observation, 23, was wrongly copied as 32. Find the correct mean."* — and the formula alone doesn't save you.

So we will build this chapter differently: small chunks, one idea at a time, with the "why" before the "how".

Chunk 1: What Statistics Actually Does

Statistics is the science of collecting, organising, analysing, and interpreting data to make decisions under uncertainty. For your exams, it splits into two branches:

• Descriptive statistics — summarising data you already have (mean, median, mode, standard deviation). This forms the vast majority of what SSC, Banking, and CSAT ask.
• Inferential statistics — drawing conclusions about a population from a sample (probability distributions, estimation). Only the basics appear, mostly in CAT-level papers.

Data itself comes in three forms:

🧠 Active Recall: Before reading on — what is the difference between descriptive and inferential statistics? Say it aloud in one sentence. If you can't, reread the two bullets above.

Chunk 2: The Mean — and Its Secret Weakness

The arithmetic mean is the balance point of data:

$$\bar{x} = \frac{\sum x_i}{n} \quad \text{or for frequency data:} \quad \bar{x} = \frac{\sum f_i x_i}{\sum f_i}$$

Key properties the exam loves:

1. Algebraic sum of deviations from the mean is always zero: Σ(x − x̄) = 0. This single fact solves dozens of "correction" questions.
2. Every value changes ⇒ mean changes. If each observation is increased by k, the mean increases by k. If each is multiplied by k, the mean is multiplied by k.
3. The mean is sensitive to outliers. One billionaire entering a village of 1,000 people can double the "average income" — without anyone actually getting richer.

The classic trap — wrong observation correction:
Mean of 40 observations = 36, but 23 was misread as 32.

• Wrong total = 40 × 36 = 1440
• Correct total = 1440 − 32 + 23 = 1431
• Correct mean = 1431 / 40 = 35.775

Notice: we never needed the individual observations. Total = mean × n is the master key.

Combined mean of two groups:
$$\bar{x}_{combined} = \frac{n_1\bar{x}_1 + n_2\bar{x}_2}{n_1 + n_2}$$

🧠 Active Recall: A class of 30 boys has mean weight 60 kg and 20 girls have mean weight 50 kg. What is the combined mean? (Try it: (30×60 + 20×50)/50 = 56 kg. Did you get it?)

Chunk 3: Median — The Middle That Ignores Extremes

The median is the middle value when data is arranged in order. It divides data into two equal halves.

• Raw data: Arrange in ascending order.
• If n is odd → median = value of the ((n+1)/2)th item.
• If n is even → median = average of the (n/2)th and (n/2 + 1)th items.
• Grouped data:
  $$\text{Median} = L + \frac{\frac{N}{2} - C}{f} \times h$$
  where L = lower limit of median class, N = total frequency, C = cumulative frequency before the median class, f = frequency of the median class, h = class width.

Why exams love the median: it is not affected by extreme values. If the largest value in a dataset doubles, the mean jumps but the median doesn't move. Questions testing this conceptual point appear regularly in CSAT and Banking exams.

🧠 Active Recall: Data: 3, 7, 9, 12, 100. Mean = 26.2, Median = 9. Which one better represents a "typical" value here, and why?

Chunk 4: Mode and the Golden Triangle

The mode is the most frequent value. For grouped data:

$$\text{Mode} = L + \frac{f_1 - f_0}{2f_1 - f_0 - f_2} \times h$$

where f₁ = frequency of modal class, f₀ = frequency of the class before it, f₂ = frequency of the class after it.

The empirical relationship (memorise this — it is asked directly):

$$\boxed{\text{Mode} = 3\,\text{Median} - 2\,\text{Mean}}$$

This holds for moderately skewed distributions and is a one-line answer to many SSC CGL questions: *"If mean = 40 and median = 42, find mode."* → Mode = 3(42) − 2(40) = 46.

🎵 Mnemonic — "M3M2": *"Mode needs 3 Medians but pays only 2 Means."* Say it once and the formula is yours forever.

🧠 Active Recall: Without looking — write the empirical relation between mean, median, and mode. Now check.

Chunk 5: Measuring Spread — Variance and Standard Deviation

Two datasets can share a mean yet be completely different. Spread measures fix this blindness.

• Range = Maximum − Minimum. Quick, but uses only two values.
• Variance (σ²):
  $$\sigma^2 = \frac{\sum (x_i - \bar{x})^2}{n} = \frac{\sum x_i^2}{n} - \bar{x}^2$$
• Standard deviation (σ) = √variance. It is in the same units as the data, which is why it is preferred for interpretation.

Properties that become exam questions:

1. Adding a constant k to every value → σ unchanged (spread doesn't move).
2. Multiplying every value by k → σ becomes |k|·σ.
3. σ is always ≥ 0; σ = 0 only when all values are identical.

Coefficient of Variation (CV) — for comparing consistency across datasets with different means or units:

$$CV = \frac{\sigma}{\bar{x}} \times 100\%$$

Lower CV = more consistent. This is the standard answer to "which player/team/machine is more consistent?" questions.

🧠 Active Recall: Player A: mean 50, σ = 10. Player B: mean 40, σ = 6. Who is more consistent? (CV_A = 20%, CV_B = 15% → B. Did you compare CVs and not just σ?)

Chunk 6: Skewness — Which Way Does the Data Lean?

• Symmetrical: Mean = Median = Mode.
• Positively skewed (right-skewed): Mean > Median > Mode. A long tail of high values (e.g., income data) drags the mean up.
• Negatively skewed (left-skewed): Mean < Median < Mode.

🎵 Mnemonic: *"The Mean is a greedy child — it always runs toward the tail."* Right tail → mean runs right (highest). Left tail → mean runs left (lowest).

🧠 Active Recall: In a positively skewed distribution, arrange mean, median, and mode in descending order. (Answer: Mean > Median > Mode.)

Chunk 7: The Examiner's Favourite Traps

1. Wrong-observation corrections — always work through the *total*, never the list.
2. Combined mean ≠ average of means — you must weight by group size. Averaging 60 and 50 to get 55 is wrong when groups are unequal.
3. Median of even n — students forget to average the two middle values.
4. "More consistent" questions — compare CV, not σ, when means differ.
5. Effect of transformations — adding a constant changes the mean but not σ; multiplying changes both.
6. Negative-worded stems — "which of the following is NOT affected by extreme values?" (Answer: median/mode, not mean.)

Summary Table

🔁 Review This Chapter After 2 Days

Statistics fades fast because it is formula-dense. After 48 hours, close the book and test yourself on: (1) the empirical relation Mode = 3Median − 2Mean, (2) the grouped median and mode formulas, (3) the effect of adding vs. multiplying a constant on mean and σ, and (4) the CV rule for consistency. If any of these feel shaky, reread only that chunk — then move to the flashcards and practice sets.