HKDSE · Analysis & Prediction

2022 HKDSE Information and Communication Technology Past Paper Analysis

The 2022 DSE ICT exam features high-utility computational thinking problems, intricate storage calculations, and precise SQL spreadsheet logic that strongly separated high-scoring candidates.

Updated: 11 Jun 2026

Analysis Sample paper Solution

Difficulty 3.5/5

No official paper links

Total marks

145

Duration

210 mins

Most tested

Relational Databases, SQL and Algorithmic Thinking

Paper breakdown

Paper 1 (Compulsory): 100 marks · 120 minsPaper 2 (Elective Representative): 45 marks · 90 mins

Top chapters

SQL (Databases) · 20 marksAlgorithm Design · 16 marksData Organisation and Data Control · 16 marksNetworking and Internet Basics · 15 marksRelational Databases Concepts · 14 marks

DSE 2022 ICT Paper Analysis: Deep Technical Insight

The 2022 HKDSE ICT examination presents a well-balanced yet challenging combination of fundamental concepts and rigorous technical applications. Candidates faced several classic traps that differentiated elite performers from average students. The overall difficulty was elevated by analytical questions requiring deep comprehension rather than rote memorisation, notably in storage bin-packing calculations, algorithm tracing, and cell/database references.

Key Areas of Assessment & Marks Distribution

In Paper 1 Section A (MC), question performance varied significantly. Standard knowledge-based queries on networking, basic hardware, and social implications saw high correct rates (exceeding 80%). However, mathematical and algorithmic questions saw sharp declines. For instance, Question 20 on storage calculations had a correct rate of only 7%. Candidates failed to realize that individual 300GB files cannot be partitioned across multiple 2TB drives, a typical bin-packing problem. Similarly, Question 11 on character representation (25% correct rate) highlighted gaps in basic machine organisation principles.

In Section B (Structured), major marks were allocated to relational database/spreadsheet logic and algorithmic design. The VLOOKUP formula in Question 2(b) required absolute referencing for the table array (e.g., $A$2:$B$5); omitting the dollar signs resulted in widespread loss of marks when candidates auto-filled the formula. In Question 5 (Algorithm tracing), understanding loop execution counts and trace table dry-runs was paramount to securing full marks.

Examiner Pitfalls & Misconceptions

Vague Explanations: Many candidates used superficial words like "fast" or "secure" without clarifying the underlying mechanism (e.g., "higher clock rate" or "SSL encryption").
Database vs. Spreadsheet Confusion: Candidates frequently mixed SQL syntax with spreadsheet functions, demonstrating a lack of understanding of structured databases vs. flat-file models.
Networking Attributes: A common misconception remains that 5GHz Wi-Fi is superior in every scenario, ignoring its shorter range and poorer wall-penetration capabilities compared to 2.4GHz.

Preparation Strategy & Future Prediction

To master future exams, students must build rigorous practice routines. First, do not skip trace tables. Always draw a structured dry-run table for algorithm loops. Second, pay close attention to SQL query syntax, especially nested subqueries and grouping rules (e.g., using aggregate functions alongside grouped attributes). With the rapid progression of AI technology, we predict a significant shift towards questions testing machine learning applications, ethical implications of AI/LLMs, and data security (RAID configurations, backup policies, and cloud protocols). Regular exposure to actual past paper scenarios is the most effective way to identify and correct these conceptual traps.

Charts

Marks by chapter

See where the marks were concentrated so revision time goes to the highest-value topics.

Question type mix

Compare the mark share of each paper section and question type.

Multiple-choice40 marks · 40 questions
Compulsory Structured60 marks · 5 questions
Elective Structured45 marks · 3 questions

Mark accessibility

Estimate which marks were basic, mid-level, or high-difficulty.

79%within easy or medium reach

easy: 50 marksmedium: 65 markshard: 30 marks

Difficulty trend

Compare difficulty across recent years.

Command word frequency

Spot common command words so answers match the expected response style.

Skill weighting

Shows the skill mix this paper tested most heavily.

Study ROI

Bigger bubbles recur more often; higher bubbles carry more marks, helping you rank revision priorities.

SQL Queries & Subqueries

20 marks · Difficulty 3 · recurrence 5%

Spreadsheet Functions & References

12 marks · Difficulty 2.5 · recurrence 5%

Algorithm Design & Trace Tables

16 marks · Difficulty 4.2 · recurrence 5%

Networking Protocols & Wi-Fi Frequencies

15 marks · Difficulty 2.8 · recurrence 5%

Technological Innovations (Social Implications)

12 marks · Difficulty 1.8 · recurrence 4%

Time vs marks

Compare marks with suggested time allocation to plan exam pacing.

marksmins

Next-year prediction

Topics worth watching next year, with the reason shown directly below each bar.

SQL Subqueries & Views in Databases90%

Always featured in Paper 2A, candidates repeatedly struggle with operators like '>= ALL' and the modular logic of utilizing created Views.

Trace Tables for 2D Arrays & Multi-nested Loops88%

A key filter of high-performing candidates. Complex spiral indexing and multi-conditional checks (such as those in 2022 Q4) will continue to recur.

Generative AI Ethical Boundaries & Prompt Safety85%

The chatbot question in 2022 was introductory. Since then, generative AI (LLMs) and deep fakes have dominated, making them highly probable for forthcoming contextual examination questions.

Cumulative marks ladder

The line is your running mark total question by question; dashed lines are the estimated grade cut-offs. See which question the line crosses your target grade at, so you know how far you must answer cleanly and which questions decide a band.

Topic-year heatmap

Compare topic weight by year to spot recurring and returning areas.

2021

2022

Data Representation (Information Processing)

Algorithm Design (Computational Thinking and Programming)

Basic Machine Organisation (Computer System Fundamentals)

Relational Databases Concepts (Databases)

Networking and Internet Basics (Internet and its Applications)

Data Manipulation and Analysis (Information Processing)

Technological Innovations (Social Implications)

Threats and Security on the Internet (Internet and its Applications)

Health and Ethical Issues (Social Implications)

System Software (Computer System Fundamentals)

Examiner insights

Official report

Common pitfalls

Candidates struggled with bin-packing logic in MCQ Q20. Many simply divided the total gigabytes by 2,000 without realizing individual 300GB video files cannot be split across separate hard disks.
In Excel formulas (Paper 1 Section B Q2b), candidates consistently forgot to apply absolute referencing ($) to the lookup table array (using A2:B5 instead of $A$2:$B$5), leading to auto-fill execution errors.
In complete SQL queries, candidates often grouped by correct attributes but completely missed matching conditions or forgot to specify the relational JOIN criteria in the WHERE clause.

Misconceptions

Many candidates falsely believed that the primary function of CAPTCHA is to encrypt password data rather than distinguishing automated bots from human users.
Candidates frequently assumed that a 5GHz Wi-Fi signal has higher penetration power through physical obstacles (like thick concrete walls) than 2.4GHz because of its higher frequency.
Many students assumed open-source software is completely equivalent to freeware, omitting that open-source guarantees access to original source code for modifications.

Where marks are lost

Superficial answers like 'it is faster' or 'more secure' without explaining technical causes (such as 'shorter computational cycle', 'higher network bandwidth' or 'SSL key encryption').
Omitting the 'GROUP BY' clause when select attributes list both raw column names and SQL aggregate functions (like MAX or SUM), which causes a query validation error.
Failing to correctly trace the dry-run of variable bounds, resulting in off-by-one errors in array iterations and condition checks within nested loops.

Estimated grade cut-off

Updated: 11 Jun 2026

Source: DSE Treasure — DSE Cut-off (2012–2025, all subjects)

Level	Mark	Percentage
5**	—	87%
5*	—	82%
5	—	80%
4	—	72%
3	—	64%

Estimated cut-offs only. The HKEAA does not publish official raw-mark cut-offs for the HKDSE; these figures are reconstructions from the cited source and may differ from the actual boundaries.