Data Interpretation — SSC CGL Quantitative Aptitude

DI is the only quant topic with a guaranteed appearance: recent Tier 1 papers carry one full set (3–5 questions on a single chart), and Tier 2 adds statistics-flavoured reads. Nothing in DI is conceptually new — it is percentage, ratio and averages executed against a picture, under time pressure. The aspirants who lose marks here lose them to reading errors and slow arithmetic, never to difficulty.

1. What SSC actually asks

One data set, 3–5 questions. The chart types, in order of frequency:

1. Tables — raw numbers by year/category. Watch the units row ("in thousands", "in lakh").
2. Bar graphs — single or grouped bars; SSC loves "production of two companies across 5 years".
3. Pie charts — shares in % or degrees ($x°=\frac{x}{360}$ of total). Sometimes two pies (two years) — the classic trap set.
4. Histograms & frequency polygons — class intervals and frequencies; Tier 2 leans on these (they sit in the official syllabus under statistics).
5. Line graphs — trends over time; questions about steepest rise or % change.

The five question patterns

2. The pie-chart toolkit

A pie shows parts of a whole, in per cent or degrees.

$\text{Value of a sector}=\frac{\text{sector }\%}{100}\times \text{Total}=\frac{\text{sector degrees}}{360}\times \text{Total}$

Degree ↔ per cent: $1\%=3.6°$, so $18\%=64.8°$ and $90°=25\%$.

The two-pie trap. Two pies (say, expenditure of a family in 2022 and 2023) usually sit on different totals. "Education share fell from 20% to 18%" does not mean education spending fell — if the total rose from ₹40,000 to ₹50,000, spending went from ₹8,000 to ₹9,000, an increase. Every two-pie set contains one question priced exactly on this confusion.

Rule: shares compare within a pie; values compare across pies only after multiplying by each pie's own total.

3. The approximation discipline

DI numbers are chosen so exact calculation is possible but slow. The faster route:

• Compare fractions by cross-multiplication, not division: is $\frac{437}{812}$ bigger than $\frac{389}{705}$? Check $437\times 705$ vs $389\times 812$ — or better, note both are ≈ 54–55% and only then refine if options are close.
• Percentage benchmarks: to find $\frac{347}{1520}$ as %, anchor on 10% = 152 → 20% = 304 → remaining 43 ≈ 2.8% → ≈ 22.8%.
• Option spread decides precision. Options 5% apart → approximate boldly. Options 0.5% apart → compute exactly. Read the options before calculating.
• Averages of a series with a pattern: for consecutive-ish values, average ≈ middle value; only correct for the deviations.

4. Ratio & share moves

• "A as a % of B" = $\frac{A}{B}\times 100$ — the base is B, the thing after "of".
• "By what % is A more than B" = $\frac{A-B}{B}\times 100$ — base is still B ("than B").
• Combined shares: if boys are 60% and 30% of boys pass, passing boys = 18% of the total — multiply shares down the chain.

5. Solved set (bar-graph pattern)

Production (in tonnes) of Company X: 2019: 320, 2020: 400, 2021: 350, 2022: 490, 2023: 540.

Q1. Average annual production? Sum $=320+400+350+490+540=2100$ → average $=420$ tonnes. (Anchor trick: guess 400, deviations −80, 0, −50, +90, +140 sum to +100 → +20 correction.)

Q2. Which year saw the highest percentage increase? 2020: $\frac{80}{320}=25\%$; 2022: $\frac{140}{350}=40\%$; 2023: $\frac{50}{490}\approx 10\%$. Answer: 2022. (Not 2023, despite the biggest bar — % change divides by the smaller base.)

Q3. Production in 2021 is what per cent of total production? $\frac{350}{2100}=\frac{1}{6}=16.67\%$. (The fraction table from the Percentage chapter strikes again.)

Q4. If 2024 production is 15% more than 2023, the 2024 figure is? $540\times 1.15=621$ tonnes. (Multiplying factor — never compute 15% separately and add.)

6. Histograms & frequency polygons (Tier 2 flavour)

• A histogram plots class intervals (e.g., marks 20–30, 30–40) against frequencies; bar area represents frequency (with equal widths, height does).
• A frequency polygon joins the midpoints of histogram bar tops.
• Standard asks: which class is modal (tallest bar), how many observations lie below/above a boundary (add frequencies), and ratio of one class to total.

These questions are pure addition — the risk is misreading interval boundaries ("less than 40" includes the 30–40 class; "40 and above" starts at 40–50).

7. Exam protocol for a DI set

1. 30-second scan first: title, axes, units, legend, footnote. Half of all DI errors are unit errors ("in thousands").
2. Read options before calculating — they set your precision budget.
3. Do the set in order of cheapness: direct reads → single ratios → multi-step combos.
4. Never abandon a set mid-way — the scan cost is sunk; the remaining questions amortise it.
5. Write intermediate sums on paper. Mental accumulation across five bars under a timer is how 420 becomes 410.