CARIBBEAN EXAMINATIONS COUNCIL

REPORT ON CANDIDATES’ WORK IN THE

CARIBBEAN ADVANCED PROFICIENCY EXAMINATION

®

MAY/JUNE 2020

COMPUTER SCIENCE

Copyright © 2019 Caribbean Examinations Council

St Michael, Barbados
All rights reserved.
 GENERAL COMMENTS

This was the twelfth year in which the revised syllabus was examined. In both Unit 1 and Unit 2, Paper 01
was examined externally by CXC while Paper 03, the School-Based Assessment, was examined by teachers
and moderated by CXC. Due to the COVID-19 situation, Paper 02 was not administered.

Paper 01 consisted of 45 multiple-choice items designed to test candidates’ breadth of coverage of the
syllabus.

Performance on the multiple-choice paper for Unit 1 improved compared to 2019. The mean score for
July 2020 was 69.83 with a standard deviation of 17.62 compared to a mean score of 61.88 and a standard
deviation of 17.53 in 2019.

Performance on the multiple-choice paper Unit 2 also improved compared to 2019. The mean score for
July 2020 was 70.86 with a standard deviation of 16.13 compared to a mean score of 66.61 and a standard
deviation of 17.98 in 2019.

DETAILED COMMENTS

UNIT 1

Paper 03 — School-Based Assessment

Definition of Problem

Many students were able to accurately describe the problem, complete with the description of the current
system. In addition to stating the issue or problem, they also provided the background to the issue or
problem, the recommended solution and examples of what takes place when the problem arises. Some
students only included partial definitions of activities of the proposed system that they intended to create
without completely describing the current system and the problems which are to be solved.

Narrative and Flowchart or Pseudocode

Narratives were fairly well written in most samples. Pseudocode algorithms were generally well done. In
some instances, symbols were incorrectly used in flowcharts.

Coding of Program

Most programs were logically written and properly decomposed. However, some students
  did not use functions as independent units in their programs and yet some teachers still awarded
full marks
 used too few data structures in their programs
 did not demonstrate appropriate use of the concept of structured programming.

Some students did not print their programs from the C compiler, instead choosing to make print screen
copies of the project.

Evidence that Code Matches Algorithm

Most students were able to obtain full marks in this area.

Evidence of File Manipulation

This section was well done. However, some students were awarded marks by the teacher even though
there was no evidence in support of this requirement.

Testing and Presentation

Most of the samples moderated did not have a suitable range of test data. However, a few students

 had a test plan and screenshots while others provided screenshots without any test plan
 did not include all testing criteria (normal, extreme and erroneous) in their test plan
 had test results but did not have a clear test plan
 had no test results but were awarded marks.

UNIT 2

Paper 03 — School-Based Assessment

Students should follow the order laid out in the criteria for marking when arranging the sections of the
SBA.

Teachers should ensure that each SBA is clearly labelled with the student’s name and centre number.

Definition of Problem

Almost all students handled this section fairly well but there were a few who seemed unclear as to what
to include in their definition of problem. A brief description of the context in which the problem occurred
is required but details about the actual problems staff/clients face and proposed steps to correct such
problems must be emphasized.
Analysis Techniques

Most students were able to state the various techniques used for data collection. However, some of them
were not able to justify their selections. Some students misinterpreted justification to mean definition.
Students should clearly explain why the chosen technique was used as it relates to the business/company
institution. Proof of analysis should also be given.

Use of Data Flow and E-R Diagrams

This section of the SBA was satisfactorily done by most students.

Context Level Diagrams

Students were required to give a complete and accurate diagram of all relevant entities and data flows.
In some instances,

 incorrect symbols were used for entities and processes

 data flows were not labelled
 data stores/files were incorrectly included in the context level diagram.

Level 1 Diagrams

Most students are unaware that the Level 1 diagram is an expansion of the context level diagram.

E-R Diagrams

 Most students did not use the correct symbol (a diamond) for a relationship.
 Some students did not include attributes for the entities.

Functional and Non-Functional Requirements

 For the functional requirements students did not clearly state what the system was expected to
do, instead they stated what the user would be doing.
 For the non-functional requirements, students did not state the limitations of the system.

Design Specifications

 Students should not only include screenshots of the interface and report design but also a
justification, in order to gain marks.
 Narratives will not be accepted as an algorithm. Students should submit pseudocodes or
flowcharts.
Coding and Testing

All code must be written in procedural C. No other programming language will be accepted. Code must
also be printed from the compiler, not transferred to a word processor before printing.

 Most students did not include enough screenshots to support the functioning of the code.
 Some codes presented did not match the screenshots given.
 Test plans should be written in a tabular format and should include normal, extreme, erroneous
and incompatible data.