Unit 6: Data Interpretation and Data Sufficiency

I. Orientation: Principles of Quantitative Interpretation

Data interpretation is the systematic extraction, comparison, and evaluation of numerical information presented through tables, charts, and graphs. Data sufficiency is related but distinct: it determines whether the information supplied is adequate to answer a question, without necessarily calculating the final answer.

• Core objective: Convert visual or tabular information into relevant quantities, identify relationships, and reach conclusions supported by the given data.
• Independent reading: Interpret only the values, units, labels, scales, legends, and conditions actually provided; external assumptions can invalidate a conclusion.
• Units and scale: Check whether figures represent units, thousands, percentages, degrees, currency, index values, or cumulative totals.
• Comparison basis: Distinguish absolute difference from relative or percentage difference.
  • Absolute difference: (A-B)
  • Relative difference with (B) as the base: ((A-B)/B)
• Common formulas:

Percentage = (Part / Whole) x 100
Percentage change = ((New value - Old value) / Old value) x 100
Ratio of A to B = A / B or A:B
Average = Sum of observations / Number of observations

• Approximation convention: Round only after completing intermediate calculations unless the data itself is approximate.
• Visual discipline: A taller bar, wider sector, or steeper line suggests a relationship, but numerical conclusions must come from the marked scale.
• Sufficiency principle: A statement is sufficient only when it produces one definite answer under every case consistent with that statement.

II. Graphical and Tabular Data Interpretation: Reading, Comparing, and Calculating

Data-interpretation formats organize the same underlying quantities in different visual structures, so the appropriate method depends on how categories, time, composition, frequency, and multiple variables are represented.

A. Bar graph-based problems

A bar graph represents category values by rectangular bars whose lengths are proportional to the corresponding quantities.

• Structural elements: Identify the category axis, value axis, scale interval, units, title, and legend before calculating.
• Major forms:
  • Simple bar graph: One value is shown for each category.
  • Grouped bar graph: Two or more related series are placed side by side.
  • Stacked bar graph: Each bar shows a total divided into components.
  • Percentage bar graph: Every bar has the same total length, normally representing (100\%).
• Value extraction: If one scale division represents 20 units and a bar reaches the fourth division, its value is (4 \times 20=80).
• Comparison measures:

Difference = Larger value - Smaller value
Ratio = First value / Second value
Percentage excess = (Difference / Smaller value) x 100

• Worked example: If sales are 240 units in Year 1 and 300 units in Year 2, the increase is (300-240=60) units, while the percentage increase is ((60/240)\times100=25\%).
• Common caution: A truncated vertical axis can exaggerate visual differences; calculate from axis values rather than apparent bar lengths.
• Efficiency method: Record only the bars relevant to the required comparison, especially when a grouped graph contains several series.

B. Tabular based problems

A table arranges numerical information in rows and columns, allowing precise lookup and comparison across multiple categories.

• Table structure: Read the title, row headings, column headings, units, footnotes, and whether totals are row-wise or column-wise.
• Direct and derived values:
  • Direct value: Read from a single cell.
  • Derived value: Obtained by adding, subtracting, averaging, or comparing several cells.
• Row and column totals:

Row total = Sum of entries across the selected row
Column total = Sum of entries down the selected column
Share of a cell = (Cell value / Relevant total) x 100

• Worked example: A table gives quarterly outputs of 120, 150, 130, and 160 units. Annual output is (120+150+130+160=560) units, and the fourth quarter contributes ((160/560)\times100\approx28.57\%).
• Missing entries: If the total is known, a missing cell equals the total minus the sum of all known cells in the same row or column.
• Weighted interpretation: An overall average may require weights. If groups contain different numbers of observations, averaging their averages directly is generally incorrect.

Weighted average = Sum of (value x weight) / Sum of weights

• Common caution: “Total,” “average,” and “percentage” may use different bases; confirm the relevant denominator before comparing cells.

C. Pie-chart based problems

A pie chart displays the proportional composition of a whole as sectors of a circle totaling (360^\circ) or (100\%).

• Sector relationship:

Sector angle = (Category value / Total value) x 360 degrees
Category percentage = (Sector angle / 360 degrees) x 100
Category value = (Sector angle / 360 degrees) x Total value

• Useful conversions: (1\%) corresponds to (3.6^\circ), while (90^\circ), (120^\circ), and (180^\circ) correspond to (25\%), (33\frac{1}{3}\%), and (50\%).
• Worked example: If a (72^\circ) sector represents transport expenditure in a total budget of 50,000, its share is (72/360=20\%), and the expenditure is (0.20\times50{,}000=10{,}000).
• Ratio comparison: Sector values are proportional to sector angles. Two sectors of (60^\circ) and (90^\circ) have the ratio (60:90=2:3).
• Multiple pie charts: Compare actual values only after accounting for each chart’s total. Equal percentages from totals of 400 and 600 represent different quantities.
• Common caution: Rounded percentages may sum to 99% or 101%; this can be a rounding effect rather than a data error.

D. Line graph based problems

A line graph plots values against an ordered variable, usually time, and connects successive points to reveal movement and trend.

• Axes and intervals: Confirm the time interval on the horizontal axis and the numerical scale and unit on the vertical axis.
• Direction of movement:
  • An upward segment indicates an increase.
  • A downward segment indicates a decrease.
  • A horizontal segment indicates no change.
• Magnitude and rate: Slope represents change per unit on the horizontal axis.

Slope = (Change in vertical value) / (Change in horizontal value)
Average annual change = (Final value - Initial value) / Number of intervals

• Worked example: If output rises from 200 units in 2021 to 260 units in 2024, the absolute rise is 60 units, the percentage rise is (30\%), and the average annual rise across three intervals is (60/3=20) units.
• Multiple lines: Use the legend to track each series; intersections indicate equal plotted values at that point, not necessarily equal totals over the full period.
• Trend versus fluctuation: A series may have an overall upward trend despite temporary declines between adjacent points.
• Common caution: The steepness of lines cannot be compared across graphs with different axis scales unless the scales are normalized.

E. Mixed graph-based problems

Mixed graphs combine two or more formats, such as a bar graph with a line graph or a table with a pie chart, to represent related variables.

• Integration principle: Interpret each component independently before connecting values through a common category, year, or identifier.
• Dual-axis graphs: A bar series may use the left axis while a line series uses the right axis; assigning a value to the wrong axis produces a false result.
• Cross-format calculation: One component may supply a total while another supplies a percentage or ratio needed to divide that total.
• Worked example: Bars show total production of 800 units, while a line shows a (15\%) defect rate. Defective output is (800\times0.15=120) units, and acceptable output is (800-120=680) units.
• Consistency checks: Component values should reconcile with stated totals, subject to rounding. Percentages representing a complete whole should be approximately (100\%).
• Common caution: Similar colors or markers may represent different variables in separate components; consult every legend and axis label.

F. Histogram based problems

A histogram represents the frequency distribution of continuous data through adjacent rectangular bars corresponding to class intervals.

• Histogram distinction: Unlike an ordinary bar graph, its bars normally touch because intervals such as (10\text{–}20) and (20\text{–}30) form a continuous scale.
• Equal class widths: When all intervals have equal width, bar height may directly represent frequency.
• Unequal class widths: Bar area, not height alone, must be proportional to frequency.

Frequency density = Frequency / Class width
Frequency = Frequency density x Class width

• Worked example: A class interval (20\text{–}30) has frequency 40, so its density is (40/10=4). An interval (30\text{–}50) with frequency 60 has density (60/20=3); although its frequency is larger, its bar is shorter.
• Modal class: For equal widths, the tallest bar identifies the class with the greatest frequency; for unequal widths, compare frequency densities.
• Class boundaries: Inclusive classes such as (10\text{–}19) and (20\text{–}29) may be converted to continuous boundaries (9.5\text{–}19.5) and (19.5\text{–}29.5).
• Limitation: A histogram shows grouped frequencies, so exact individual observations cannot usually be recovered.

III. Data Sufficiency: Determining Whether Information Is Adequate

Data-sufficiency analysis tests whether one or more statements uniquely determine the requested quantity or decision under the conditions in the question.

A. Check sufficiency of data

Checking sufficiency requires testing the statements separately and jointly for uniqueness, rather than merely attempting to accumulate information.

• Initial step: Identify exactly what must be determined: a numerical value, a comparison, or a definite yes-or-no answer.
• Statement testing procedure:
  1. Use Statement I alone and disregard Statement II completely.
  2. Use Statement II alone and disregard Statement I completely.
  3. If neither is sufficient alone, combine them.
  4. Check whether even the combined statements allow multiple valid answers.
• Uniqueness rule: A statement is sufficient when all cases satisfying it yield the same answer. It need not reveal every unknown variable.
• Yes-or-no questions: A definite “yes” or a definite “no” is sufficient; a statement allowing both outcomes is insufficient.
• Worked example: To determine (x), suppose Statement I gives (x^2=25), while Statement II gives (x>0).
  • Statement I is insufficient because (x=5) or (x=-5).
  • Statement II is insufficient because infinitely many positive values are possible.
  • Together, the statements force (x=5), so the combined data are sufficient.
• Domain restrictions: Conditions such as “(x) is an integer,” “(n) is positive,” or “the triangle is right-angled” can turn an otherwise ambiguous statement into a sufficient one.
• Equation count caution: Two equations are not automatically sufficient for two unknowns; they may be dependent, inconsistent, or nonlinear with multiple solutions.
• Avoided assumptions: A diagram need not be drawn to scale, people need not be distinct unless stated, and quantities need not be positive or integral without an explicit condition.
• Final classification: Record whether Statement I alone, Statement II alone, either statement alone, both statements together, or even both together are sufficient; match this logical result to the answer convention used in the given assessment.