Unit 6: Numerical Ability

I. Foundations of Numerical Ability

Numerical ability is the capacity to understand quantitative information, perform accurate calculations, judge approximate magnitude, identify mathematical relationships, and draw conclusions from data. It combines knowledge of arithmetic with logical judgment: a solver must choose an appropriate method, apply it correctly, and assess whether the result is reasonable.

• Core number systems: Numerical work commonly uses natural numbers, integers, rational numbers, irrational numbers, and real numbers.
  • Natural numbers are counting numbers: (1,2,3,\ldots).
  • Integers include negative numbers, zero, and positive numbers.
  • Rational numbers can be expressed as (p/q), where (p,q\in\mathbb Z) and (q\ne0).
• Order of operations: Brackets and powers are evaluated before multiplication or division, which precede addition or subtraction. Operations of equal priority are normally processed from left to right.
• Equivalent representations: Fractions, decimals, percentages, and ratios may describe the same quantity; for example, (3/4=0.75=75\%).
• Units and dimensions: Quantities can be combined meaningfully only when their units are compatible. Thus, (2\text{ m}+50\text{ cm}=2.5\text{ m}), after conversion.
• Accuracy principle: Exact computation is preferred when the available information and purpose require a precise answer; estimation is preferred for rapid judgment, checking, or incomplete data.
• Reasonableness principle: A numerical result should be checked for sign, size, unit, and consistency with the original conditions.
• Interpretive principle: Data do not explain themselves; conclusions depend on scale, definitions, time period, sample size, and the distinction between absolute and relative change.

II. Numerical Computation — Exact Manipulation of Quantities

A. Numerical Computation

Numerical computation is the systematic use of arithmetic operations and mathematical relationships to obtain an exact or appropriately rounded result.

• Fundamental operations: Addition combines quantities, subtraction finds a difference, multiplication represents repeated scaling, and division distributes or compares quantities.
  • For (a) and (b), the basic expressions are (a+b), (a-b), (ab), and (a/b), with (b\ne0) for division.
• Fractions: Addition and subtraction require a common denominator, while multiplication operates directly on numerators and denominators.

a/b + c/d = (ad + bc)/bd
a/b × c/d = ac/bd
a/b ÷ c/d = ad/bc

Here (a,c) are numerators; (b,d) are non-zero denominators. In the division formula, (c\ne0).

• Percentages: A percentage expresses a quantity per hundred.

Percentage = (part / whole) × 100
New value after r% increase = original value × (1 + r/100)
New value after r% decrease = original value × (1 - r/100)

Here (r) is the percentage rate. Successive percentage changes must be multiplied, not simply added.

• Ratio and proportion: A ratio (a:b) compares quantities in compatible units. The proportion (a:b=c:d) implies (ad=bc).
• Powers and roots: Exponent laws simplify repeated multiplication.

a^m × a^n = a^(m+n)
a^m / a^n = a^(m-n), a ≠ 0
(a^m)^n = a^(mn)

Here (a) is the base and (m,n) are exponents.

• Average: The arithmetic mean equals the sum of observations divided by their number.

Mean = (x₁ + x₂ + ... + xₙ) / n

Here (x_1,\ldots,x_n) are observations and (n) is their count.

• Worked example: A product marked at (₹2{,}000) receives a (15\%) discount and then (18\%) tax. Its final price is
  [
  2000(1-0.15)(1+0.18)=2000(0.85)(1.18)=₹2{,}006.
  ]
  The tax and discount do not cancel because they apply to different base amounts.

B. Applications and Limitations

Numerical computation supports precise quantitative decisions, but its reliability depends on correct inputs, operations, and units.

• Applications: Exact calculations are essential in accounting, measurement, rates, interest, inventory, scientific formulas, and allocation problems.
• Error control: Intermediate values should retain sufficient decimal places; premature rounding can distort the final answer.
• Computational check: Inverse operations provide quick verification; if (248\div8=31), then (31\times8) should return (248).
• Limitation: A precisely calculated result may still be misleading when the source data are inaccurate, outdated, or defined ambiguously.

III. Numerical Estimation — Approximation and Magnitude

A. Numerical Estimation

Numerical estimation produces a value close enough to the exact answer for rapid calculation, planning, comparison, or error detection.

• Rounding: A number is rounded to a stated place by examining the digit immediately to its right. Thus, (47.386) rounded to two decimal places is (47.39).
• Significant figures: Significant figures begin with the first non-zero digit. The value (0.004786) to two significant figures is (0.0048).
• Compatible numbers: Values may be replaced by nearby numbers that are easy to calculate; for example, (598\div29.8) is approximately (600\div30=20).
• Order of magnitude: Writing a value as (a\times10^n), where (1\le |a|<10), reveals its scale. For example, (52{,}000=5.2\times10^4).
• Bounds: Rounding implies an interval rather than a perfectly exact value. If a length is (12.4\text{ cm}) to the nearest (0.1\text{ cm}), then
  [
  12.35\le L<12.45,
  ]
  where (L) is the actual length in centimetres.
• Error measures: Absolute error measures numerical distance, while relative and percentage errors compare that distance with the actual or accepted value.

Absolute error = |approximate value - actual value|
Relative error = absolute error / |actual value|
Percentage error = relative error × 100

• Worked example: Estimate (19.8\times5.12). Using compatible values,
  [
  19.8\times5.12\approx20\times5=100.
  ]
  The exact value should therefore be near (100); an answer such as (10.1376) would immediately indicate a misplaced decimal point.

B. Applications and Limitations

Estimation is most useful when speed and plausibility matter more than an exact final value.

1. Effective use: Budget forecasting, mental arithmetic, measurement planning, and checking calculator output benefit from rounded values and magnitude comparisons.
2. Potential misuse: Estimation is unsuitable when small differences affect safety, dosage, legal payment, engineering tolerance, or final financial settlement.

• Accumulated error: Repeatedly using rounded intermediate values may produce a substantial final deviation.
• Direction of bias: Rounding every value upward systematically overestimates a total; balanced rounding reduces this bias.
• Required precision: The context determines acceptable accuracy. Reporting a population estimate to the nearest thousand may be sensible, while reporting a machine component to the nearest metre is not.

IV. Numerical Reasoning — Relationships, Constraints, and Conclusions

A. Numerical Reasoning

Numerical reasoning is the process of identifying quantitative patterns, translating conditions into mathematical relationships, and deriving logically justified conclusions.

• Translation into equations: Verbal relationships become variables and operations. If one number is five more than twice another, the relation is (x=2y+5), where (x) and (y) represent the two numbers.
• Rate reasoning: Rate connects two quantities measured in different units.

Rate = quantity / time
Distance = speed × time
Work completed = work rate × time

Units must remain consistent; a speed in kilometres per hour requires time in hours to produce kilometres.

• Proportional reasoning: Direct proportion has the form (y=kx), while inverse proportion has the form (y=k/x). Here (k) is a constant of proportionality.
• Sequence recognition: A pattern may depend on constant differences, constant ratios, alternating operations, squares, cubes, or combinations of earlier terms.
  • Arithmetic sequence: (a_n=a_1+(n-1)d).
  • Geometric sequence: (a_n=a_1r^{n-1}).
  • Here (a_n) is the (n)th term, (a_1) the first term, (d) the common difference, and (r) the common ratio.
• Constraint checking: A candidate answer must satisfy every stated condition, including positivity, integrality, range, and divisibility.
• Worked example: A vehicle travels (120\text{ km}) at (60\text{ km/h}) and another (120\text{ km}) at (40\text{ km/h}). Total time is (2+3=5) hours, so
  [
  \text{average speed}=\frac{240}{5}=48\text{ km/h}.
  ]
  The arithmetic mean (50\text{ km/h}) is incorrect because equal distances require time-weighted reasoning.

B. Applications and Limitations

Numerical reasoning converts information into defensible decisions, provided that the assumed relationship matches the situation.

• Applications: It supports scheduling, comparison of plans, resource allocation, trend recognition, and elimination of impossible outcomes.
• Sanity checks: Boundary cases test logic; a percentage cannot exceed (100\%) when it represents a subset of a fixed whole.
• Correlation caution: Two variables moving together does not establish that one causes the other.
• Pattern caution: A short sequence can fit multiple rules; the simplest consistent rule may be useful, but it is not uniquely proved without further conditions.

V. Data Interpretation — Extracting Meaning from Quantitative Displays

A. Data Interpretation

Data interpretation involves reading tables, charts, graphs, and summaries accurately, then calculating comparisons or trends supported by the displayed information.

• Tables: Row and column headings define each value; totals should be checked against categories, units, and time periods.
• Bar charts: Bar length represents magnitude. A truncated vertical axis can visually exaggerate small differences.
• Line graphs: Points show values over an ordered variable, commonly time; slope indicates the direction and rate of change.
• Pie charts: Each sector represents a proportion of the whole.

Sector angle = (category value / total value) × 360°

• Absolute and relative change: Absolute change measures the numerical difference, while percentage change expresses that difference relative to the original value.

Absolute change = new value - old value
Percentage change = [(new value - old value) / old value] × 100

• Central tendency: Mean uses every value, median is the middle ordered value, and mode is the most frequent value. An extreme observation affects the mean more strongly than the median.
• Worked example: Quarterly sales are (120,150,135,) and (195) units. Total sales equal (600), mean quarterly sales equal (150), and the increase from the first to fourth quarter is
  [
  \frac{195-120}{120}\times100=62.5\%.
  ]

B. Applications and Limitations

Sound interpretation requires attention to both calculated values and the way the data were collected and presented.

• Comparison basis: Percentages must share a clear denominator; (20\%) of customers and (20\%) of revenue describe different populations.
• Weighted data: Group averages should be combined using group sizes, not by taking an unweighted mean of the averages.
• Missing context: Sample size, source, collection method, inflation, seasonal effects, and category definitions can alter the conclusion.
• Outliers: Extreme values may represent genuine events or errors; they should be investigated before exclusion.
• Unsupported extrapolation: A past trend does not guarantee continuation beyond the observed interval.
• Presentation integrity: Conclusions should state the relevant quantity, unit, period, and comparison base so that the numerical claim can be independently understood.