Study guide

7.1 Statistical Tables: Rapid Extraction, Percentages, and Proportions

Common Recruitment Examination CRE

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: CRE Structure, Eligibility, Administration, and the BLNST

1.1 What the CRE Is, Who Can Sit It, and Which Grades Need It 1.2 Paper Formats, Grading, Equivalent Results, and Result Certificates 1.3 Sittings, Applications, Exam-Day Rules, and Pacing 1.4 The BLNST, Your Results, and What Happens After the CRE

2 Chapter 2: Use of English — Reading Comprehension

2.1 Reading Strategy, Gist Identification, and Passage Architecture 2.2 Detail Retrieval, Drawing Inferences, and Author Tone 2.3 Vocabulary in Context, Nuance, and Figurative Language

3 Chapter 3: Use of English — Error Identification & Grammar Foundations

3.1 Grammatical Errors: Subject-Verb Agreement, Tense, and Mood 3.2 Syntax Errors: Misplaced Modifiers, Parallelism, and Prepositions 3.3 Lexical and Stylistic Errors: Diction, Redundancy, and Word Choice

4 Chapter 4: Use of English — Sentence Completion & Paragraph Improvement

4.1 Sentence Completion: Structural Connectives, Subordinate Clauses, and Collocations 4.2 Paragraph Improvement: Cohesion, Flow, and Sentence Combining 4.3 Paragraph Improvement: Revision for Conciseness, Tone, and Clarity

5 Chapter 5: Aptitude Test — Verbal & Deductive Reasoning

5.1 Verbal Reasoning: True, False, and Cannot Tell Methodology 5.2 Deductive Reasoning: Categorical Logic, Syllogisms, and Conditionals 5.3 Relational Logic, Sequencing, and Analytical Elimination

6 Chapter 6: Aptitude Test — Data Sufficiency & Numerical Reasoning

6.1 Data Sufficiency: The CRE Two-Clue Format 6.2 Data Sufficiency: Hidden Anchors, Redundant Clues, and Other Traps 6.3 Numerical Reasoning: Arithmetic, Geometric, and Multi-Tier Sequences

7 Chapter 7: Aptitude Test — Interpretation of Tables & Graphs

7.1 Statistical Tables: Rapid Extraction, Percentages, and Proportions 7.2 Bar Charts, Line Graphs, and Trend Extrapolation 7.3 Multi-Data Synthesis, Composite Tables, and Common Distractors

8 Chapter 8: Use of Chinese — Reading Comprehension and Character Recognition (閱讀理解與字詞辨識)

8.1 Use of Chinese: Article Reading (文章閱讀) 8.2 Use of Chinese: Short Extract Reading (片段╱語段閱讀) 8.3 Use of Chinese: Wrongly Written Characters (字詞辨識:錯別字) 8.4 Use of Chinese: Matching Simplified and Traditional Characters (字詞辨識:簡化字)

9 Chapter 9: Use of Chinese — Sentence Analysis and Usage (句子辨析與詞句運用)

9.1 Use of Chinese: Structural Sentence Errors (句子辨析:語病) 9.2 Use of Chinese: Logical Errors and Ambiguity (句子辨析:邏輯錯誤與歧義) 9.3 Use of Chinese: Near-Synonyms and Word Choice (詞句運用:詞語選用) 9.4 Use of Chinese: Idioms, Connectives, and Polite Forms (詞句運用:成語、關聯詞與敬謙辭) 9.5 Use of Chinese: Sentence Filling and Sentence Ordering (詞句運用:句子填充與排序)

7.1 Statistical Tables: Rapid Extraction, Percentages, and Proportions

Key Takeaways Explain This Topic AI

The Interpretation of Tables and Graphs part has 8 of the Aptitude Test's 35 questions, giving about 77 seconds per question; a permitted calculator may be used in this paper.

The 15-second Pre-Reading Protocol requires inspecting table titles, column/row headers, units of measurement, and explanatory footnotes before reading the question stems.

Percentage change calculations require strict adherence to the base formula (New - Old) / Old * 100%, where confusing the initial base period with the terminal period is the most common computational trap.

Percentage points (the absolute arithmetic difference between two percentage rates) and percentage change (the relative proportional shift) are distinct mathematical concepts frequently tested in distractor design.

Estimation shortcuts such as benchmark fractions (10%, 5%, 1%), cross-multiplication and order-of-magnitude bounds eliminate distractors faster than keying every figure into a calculator.

Last updated: October 2026

7.1 Statistical Tables: Rapid Extraction, Percentages, and Proportions

The Aptitude Test (AT) of the Common Recruitment Examination is a high-speed, 45-minute screening examination comprising 35 multiple-choice questions. Among its five core disciplines, the Interpretation of Tables and Graphs module accounts for 8 questions (approximately 23% of the paper). The CSB's guidance notes allow a permitted calculator (cordless, silent, and without printing, graphic or word-display, or dictionary functions) in the Aptitude Test. Even so, keying in every figure costs time. Strong candidates extract the right cells, estimate to eliminate options, and use the calculator only for the final step.

With an average allocation of only 77.1 seconds per question , success on this module does not depend on laborious, multi-decimal precision. The CSB's own sample uses a labour-force table of unemployment by age group, much like an official statistical publication. The questions test your capacity to isolate the correct data cells rapidly, identify underlying mathematical relationships, apply estimation shortcuts, and reject mathematically flawed distractors.

The Anatomy of Complex Statistical Tables

Statistical tables in this part resemble real policy briefings. Unlike elementary school math problems with isolated figures, CRE tables feature multi-row, multi-column matrices containing mixed metrics: absolute counts, financial values, index points, and percentage distributions. A typical table exhibits four distinct structural components:

The Universe & Title: Defines the geographic scope, target population, and temporal window (e.g., "Quarterly Composite Household Survey: Labor Force and Median Monthly Employment Earnings by Selected Economic Activity, 2021–2025" ).

Column and Row Headers: Categorical taxonomies often organized in hierarchical tiers (e.g., broad economic sector → \rightarrow → sub-industry → \rightarrow → gender breakdown).

Units and Denominations: Crucial scale markers prominently placed at the top-right corner of the table or within individual column headers (e.g., "In thousands unless otherwise specified" , "HK$ Millions" , or "Percentage share (%)" ).

Footnotes and Qualifier Symbols: Notations marked by asterisks, daggers, or lowercase letters providing critical exclusion rules, provisional estimates, or base-year shifts.

Important

The Unit Blindness Trap: Ignoring column-header unit markers is one of the commonest calculation errors. For example, if a row lists employment as 285.4 under a column header marked Thousands of Persons , the absolute number of individuals is 285,400 285{,}400 285 , 400 , not 285 285 285 or 28,540 28{,}540 28 , 540 . Always circle or note the unit before commencing arithmetic.

The 15-Second Pre-Reading Protocol

Untrained candidates immediately read the question stem, glance frantically at the table, and attempt to search for matching numbers. This undisciplined approach leads to cognitive overload and misreading. High-scoring candidates execute the 15-Second Pre-Reading Protocol before reading Question 1 of any tabular data set: [Step 1: Scope Scan] --> Read Title & Time Horizon (e.g., 2021-2025 fiscal years)

  • [Step 2: Units Check] --> Verify Measurement Units (HK\$ Millions vs Thousands; % vs Counts)
  • [Step 3: Axis Inventory] --> Identify Rows (Variables/Sectors) vs Columns (Years/Demographics)
  • [Step 4: Footnote Alert] --> Read Footnotes for Revisions, Exclusions, or Base Index Years

By spending 15 seconds establishing the mental geography of the exhibit, you avoid most misread row and column errors and know where to retrieve values when reading the stems.

Core Mental Arithmetic & Shortcut Formulas

Whether or not you use a calculator, you need the core formulas of statistical analysis and the shortcuts that make them fast.

1. Percentage Change

The standard formula for percentage change measures the relative shift from a baseline period to a terminal period:

Percentage Change = ( Value New − Value Old Value Old ) × 100 % = ( Δ V V Old ) × 100 % \text{Percentage Change} = \left( \frac{\text{Value}_{\text{New}} - \text{Value}_{\text{Old}}}{\text{Value}_{\text{Old}}} \right) \times 100\% = \left( \frac{\Delta V}{V_{\text{Old}}} \right) \times 100\% Percentage Change = ( Value Old ​ Value New ​ − Value Old ​ ​ ) × 100% = ( V Old ​ Δ V ​ ) × 100%

Warning

The Base-Period Reversal Trap: The denominator is always the initial (older) value, never the terminal (newer) value. If an agency's headcount falls from 800 800 800 in 2023 to 600 600 600 in 2024, the percentage decrease is 600 − 800 800 = − 200 800 = − 25.0 % \frac{600 - 800}{800} = -\frac{200}{800} = -25.0\% 800 600 − 800 ​ = − 800 200 ​ = − 25.0% . Distractors routinely use the new value as the denominator ( − 200 600 = − 33.3 % \frac{-200}{600} = -33.3\% 600 − 200 ​ = − 33.3% ) to snare careless candidates.

2. Percentage Points vs. Percentage Change

A distinction that table and graph questions often test is the difference between an absolute change in percentage rates and a relative percentage change :

Percentage Points (p.p.): The simple arithmetic difference between two percentage values: Δ p.p. = Rate New − Rate Old \Delta_{\text{p.p.}} = \text{Rate}_{\text{New}} - \text{Rate}_{\text{Old}} Δ p.p. ​ = Rate New ​ − Rate Old ​

Relative Percentage Change (%): The proportional growth of the rate relative to its initial value: Relative Change = ( Rate New − Rate Old Rate Old ) × 100 % \text{Relative Change} = \left( \frac{\text{Rate}_{\text{New}} - \text{Rate}_{\text{Old}}}{\text{Rate}_{\text{Old}}} \right) \times 100\% Relative Change = ( Rate Old ​ Rate New ​ − Rate Old ​ ​ ) × 100%

Illustration: If the unemployment rate in a district rises from 4.0 % 4.0\% 4.0% in 2022 to 5.0 % 5.0\% 5.0% in 2023:

The rate rose by 1.0 1.0 1.0 percentage point ( 5.0 % − 4.0 % = 1.0 p.p. 5.0\% - 4.0\% = 1.0\text{ p.p.} 5.0% − 4.0% = 1.0 p.p. ).

The rate rose by 25.0 % 25.0\% 25.0% in relative terms ( 5.0 − 4.0 4.0 × 100 % = 1.0 4.0 × 100 % = 25.0 % \frac{5.0 - 4.0}{4.0} \times 100\% = \frac{1.0}{4.0} \times 100\% = 25.0\% 4.0 5.0 − 4.0 ​ × 100% = 4.0 1.0 ​ × 100% = 25.0% ).

Confusing these two descriptions is among the most common distractor traps in civil service aptitude tests.

3. Base-Period Indexing

Statistical tables frequently express economic time series as an index, where a designated baseline period is assigned a value of 100.0 100.0 100.0 :

Index t = ( Value t Value Base ) × 100 \text{Index}_{t} = \left( \frac{\text{Value}_{t}}{\text{Value}_{\text{Base}}} \right) \times 100 Index t ​ = ( Value Base ​ Value t ​ ​ ) × 100

To find the percentage change between any two non-base periods t 1 t_1 t 1 ​ and t 2 t_2 t 2 ​ , you cannot simply subtract the index points unless t 1 t_1 t 1 ​ is the base year! For any two arbitrary years:

Percentage Change t 1 → t 2 = ( Index t 2 − Index t 1 Index t 1 ) × 100 % \text{Percentage Change}_{t_1 \rightarrow t_2} = \left( \frac{\text{Index}_{t_2} - \text{Index}_{t_1}}{\text{Index}_{t_1}} \right) \times 100\% Percentage Change t 1 ​ → t 2 ​ ​ = ( Index t 1 ​ ​ Index t 2 ​ ​ − Index t 1 ​ ​ ​ ) × 100%

4. Estimation Shortcuts

Estimation is often faster than keying numbers into a calculator, and it is the quickest way to eliminate options:

Shortcut Technique Mathematical Logic Practical Application on Exam

10% / 1% Partitioning Any number X X X can be decomposed into 10 % = X / 10 10\% = X / 10 10% = X /10 and 1 % = X / 100 1\% = X / 100 1% = X /100 . To calculate 16 % 16\% 16% of 420 420 420 : take 10 % ( 42 ) + 5 % ( 21 ) + 1 % ( 4.2 ) = 67.2 10\% (42) + 5\% (21) + 1\% (4.2) = 67.2 10% ( 42 ) + 5% ( 21 ) + 1% ( 4.2 ) = 67.2 .

Benchmark Fractions Convert recurring decimals to memorized fractional equivalents. 1 6 ≈ 16.7 % \frac{1}{6} \approx 16.7\% 6 1 ​ ≈ 16.7% , 1 7 ≈ 14.3 % \frac{1}{7} \approx 14.3\% 7 1 ​ ≈ 14.3% , 1 8 = 12.5 % \frac{1}{8} = 12.5\% 8 1 ​ = 12.5% , 1 9 ≈ 11.1 % \frac{1}{9} \approx 11.1\% 9 1 ​ ≈ 11.1% , 1 12 ≈ 8.33 % \frac{1}{12} \approx 8.33\% 12 1 ​ ≈ 8.33% .

Cross-Multiplication Comparison To determine if a b > c d \frac{a}{b} > \frac{c}{d} b a ​ > d c ​ , evaluate whether a × d > b × c a \times d > b \times c a × d > b × c . Which is larger: 43 127 \frac{43}{127} 127 43 ​ or 51 149 \frac{51}{149} 149 51 ​ ? 43 × 149 = 6,407 43 \times 149 = 6{,}407 43 × 149 = 6 , 407 ; 127 × 51 = 6,477 127 \times 51 = 6{,}477 127 × 51 = 6 , 477 . Thus 51 149 > 43 127 \frac{51}{149} > \frac{43}{127} 149 51 ​ > 127 43 ​ .

Proportional Truncation When answer options are separated by > 5 % >5\% > 5% , truncate the last 2 digits of numerator and denominator. 14,892 43,120 ≈ 149 431 ≈ 150 430 = 15 43 ≈ 34.9 % \frac{14{,}892}{43{,}120} \approx \frac{149}{431} \approx \frac{150}{430} = \frac{15}{43} \approx 34.9\% 43 , 120 14 , 892 ​ ≈ 431 149 ​ ≈ 430 150 ​ = 43 15 ​ ≈ 34.9% (Exact: 34.54 % 34.54\% 34.54% ).

Loading diagram... Rapid Statistical Table Deconstruction & Arithmetic Workflow

The Official Sample

The CSB's sample table shows unemployment by age group from 1992 to 2000. Bold figures are the numbers of unemployed persons in thousands , and the figures in brackets are unemployment rates. A note defines the unemployment rate as the number unemployed in an age group divided by the labour force of the same age group, times 100%.

Question 1: During the period from 1992 to 2000, which age group consistently had the highest unemployment rate? Compare the bracketed rates within each year: the 15-19 group is highest in every year, so the answer is 15-19 . The absolute numbers are a distractor, because the 20-29 group has the most unemployed people.

Question 2: Estimate the total number of persons in the labour force in 1999. The overall row shows 57.3 thousand unemployed at a 1.9% rate. Labour force = 57,300 ÷ 1.9% ≈ 3,015,789, so the nearest option is 3,020,000 .

The sample shows the two habits this part rewards: read the note (units and definitions), and reverse a formula when the question asks for the base rather than the rate.

Realistic Examination Scenario & Worked Examples

To practise statistical interpretation problems, examine the following synthetic data set modeled on employment and earnings surveys conducted across major industries in Hong Kong.

  • Exhibit 1: Employment and Earnings in Selected Economic Sectors (2022 vs. 2025)
  • (Note: Data modeled for civil service examination preparation purposes)

Economic Sector 2022 Persons Engaged (Thousands) 2022 Median Monthly Earnings (HK$) 2025 Persons Engaged (Thousands) 2025 Median Monthly Earnings (HK$)

  • Financing and Insurance 270.0 30,000 297.0 33,600
  • Information and Communications 120.0 25,000 138.0 28,500
  • Retail, Accommodation & Food Services 500.0 15,000 475.0 16,200
  • Public Administration & Social Services 410.0 28,000 451.0 30,800
  • Total (Selected Sectors) 1,300.0 — 1,361.0 —

Footnote: Persons engaged counts are rounded to the nearest hundred persons. Median monthly earnings include basic salary, cost-of-living allowances, and year-end bonuses, rounded to the nearest hundred HK dollars.

Worked Problem 1: Percentage Change in Sectoral Employment

Question: Which economic sector experienced the highest percentage growth in persons engaged between 2022 and 2025?

Step-by-Step Mathematical Derivation:

Isolate Target Data Cells: We examine the "Persons Engaged" columns for 2022 and 2025 across all four sectors.

Compute Absolute Changes ( Δ V = V 2025 − V 2022 \Delta V = V_{2025} - V_{2022} Δ V = V 2025 ​ − V 2022 ​ ):

  • Financing and Insurance: 297.0 − 270.0 = + 27.0 297.0 - 270.0 = +27.0 297.0 − 270.0 = + 27.0 thousand
  • Information and Communications: 138.0 − 120.0 = + 18.0 138.0 - 120.0 = +18.0 138.0 − 120.0 = + 18.0 thousand

Retail, Accommodation & Food Services: 475.0 − 500.0 = − 25.0 475.0 - 500.0 = -25.0 475.0 − 500.0 = − 25.0 thousand (Contracted, eliminate immediately)

Public Administration & Social Services: 451.0 − 410.0 = + 41.0 451.0 - 410.0 = +41.0 451.0 − 410.0 = + 41.0 thousand

Compute Percentage Change ( Δ V V 2022 × 100 % \frac{\Delta V}{V_{2022}} \times 100\% V 2022 ​ Δ V ​ × 100% ):

Financing and Insurance: 27.0 270.0 × 100 % = 1 10 × 100 % = 10.0 % \frac{27.0}{270.0} \times 100\% = \frac{1}{10} \times 100\% = 10.0\% 270.0 27.0 ​ × 100% = 10 1 ​ × 100% = 10.0%

Information and Communications: 18.0 120.0 × 100 % = 18 120 = 3 20 = 15.0 % \frac{18.0}{120.0} \times 100\% = \frac{18}{120} = \frac{3}{20} = 15.0\% 120.0 18.0 ​ × 100% = 120 18 ​ = 20 3 ​ = 15.0%

Public Administration: 41.0 410.0 × 100 % = 1 10 × 100 % = 10.0 % \frac{41.0}{410.0} \times 100\% = \frac{1}{10} \times 100\% = 10.0\% 410.0 41.0 ​ × 100% = 10 1 ​ × 100% = 10.0%

Distractor Analysis: Many candidates mistakenly choose Public Administration & Social Services because its absolute gain ( 41,000 41{,}000 41 , 000 persons) was the largest in the table. However, because its baseline was also large ( 410,000 410{,}000 410 , 000 ), its percentage growth was only 10.0 % 10.0\% 10.0% . Information and Communications achieved a 15.0 % 15.0\% 15.0% growth rate, making it the highest.

Worked Problem 2: Sector Share and Percentage Point Shift

Question: By how many percentage points did the employment share of Financing and Insurance within the total selected sectors change between 2022 and 2025?

Step-by-Step Mathematical Derivation:

Determine 2022 Employment Share: Share 2022 = 270.0 1,300.0 × 100 % = 27 130 × 100 % \text{Share}_{2022} = \frac{270.0}{1{,}300.0} \times 100\% = \frac{27}{130} \times 100\% Share 2022 ​ = 1 , 300.0 270.0 ​ × 100% = 130 27 ​ × 100% Mental division: 130 × 0.20 = 26.0 130 \times 0.20 = 26.0 130 × 0.20 = 26.0 . Remaining 1.0 / 130 ≈ 0.0077 1.0 / 130 \approx 0.0077 1.0/130 ≈ 0.0077 . Thus, Share 2022 ≈ 20.77 % \text{Share}_{2022} \approx 20.77\% Share 2022 ​ ≈ 20.77% .

Determine 2025 Employment Share: Share 2025 = 297.0 1,361.0 × 100 % \text{Share}_{2025} = \frac{297.0}{1{,}361.0} \times 100\% Share 2025 ​ = 1 , 361.0 297.0 ​ × 100% Mental division: 1,361 × 0.20 = 272.2 1{,}361 \times 0.20 = 272.2 1 , 361 × 0.20 = 272.2 . Difference: 297.0 − 272.2 = 24.8 297.0 - 272.2 = 24.8 297.0 − 272.2 = 24.8 . 24.8 / 1,361 ≈ 0.0182 24.8 / 1{,}361 \approx 0.0182 24.8/1 , 361 ≈ 0.0182 . Thus, Share 2025 ≈ 21.82 % \text{Share}_{2025} \approx 21.82\% Share 2025 ​ ≈ 21.82% .

Calculate Percentage Point Difference: Δ p.p. = 21.82 % − 20.77 % = + 1.05 percentage points \Delta_{\text{p.p.}} = 21.82\% - 20.77\% = +1.05\text{ percentage points} Δ p.p. ​ = 21.82% − 20.77% = + 1.05 percentage points

Distractor Analysis: A distractor will often offer 21.82 − 20.77 20.77 × 100 % ≈ + 5.06 % \frac{21.82 - 20.77}{20.77} \times 100\% \approx +5.06\% 20.77 21.82 − 20.77 ​ × 100% ≈ + 5.06% . That represents the relative percentage change in share , not the change in percentage points requested by the stem.