Examiners’ Report

Principal Examiner Feedback

Summer 2023

Pearson Edexcel International GCSE

In Computer Science (4CP0)
Paper 02 Application of Computational Thinking
Edexcel and BTEC Qualifications

Edexcel and BTEC qualifications are awarded by Pearson, the UK’s largest awarding
body. We provide a wide range of qualifications including academic, vocational,
occupational and specific programmes for employers. For further information visit our
qualifications websites at www.edexcel.com or www.btec.co.uk. Alternatively, you can
get in touch with us using the details on our contact us page at
www.edexcel.com/contactus.

Pearson: helping people progress, everywhere

Pearson aspires to be the world’s leading learning company. Our aim is to help everyone
progress in their lives through education. We believe in every kind of learning, for all
kinds of people, wherever they are in the world. We’ve been involved in education for
over 150 years, and by working across 70 countries, in 100 languages, we have built
an international reputation for our commitment to high standards and raising
achievement through innovation in education. Find out more about how we can help
you and your students at: www.pearson.com/uk

Summer 2023
Publications Code 4CP0_02_2306_ER
All the material in this publication is copyright
© Pearson Education Ltd 2023
SPECIFIC COMMENTS

Written response questions

• There was one multiple-choice questions included in the question paper, 2(a)(ii). This
question was answered well with most candidates knowing that a function returns a
result.

• Single mark questions were also generally well answered.

o In question 1(b) most pupils got most or all available marks. The most
common errors were pupils describing what a comment was for 1(b)(i),
rather than giving the specific comment from the code, and not being able
to identify a logical operator for 1(b)(ii). Pupils scored well in 1(b)(v)
indicating that most pupils are aware of the difference between a
parameter and an argument.

o 1(c) was not as well answered. Although most candidates knew what a
constant was for 1(c)(i), most could not articulate a suitable reason for
using a constant, often simply repeating their previous answer and so not
gaining a mark for 1(c)(ii).

o 2(a)(iii) was well answered with most candidates understanding the scope
of a local variable.

• Multiple mark questions were generally less well done with many answers
lacking the detail required for an expanded response.
o In 1(a) many candidates lost the mark for identifying a process. Some
just gave incorrect answers such as ‘debit – credit’, with many giving a
vague response such as ‘a calculation’ or ‘adding’. The question was
looking for a specific process that occurred in the table, such as
‘balance + credit’. A description would also have been acceptable such
as ‘adding a credit to the balance’.

o For question 2(a)(i), most candidates knew what a subprogram library

was and had some idea about why to use them, with most scoring at
least 1 mark. For the few who scored no marks, answers were
generally vague benefits such as ‘it is faster’. Many candidates didn’t
score the second mark as their second answer was a rephrasing of
 their first response e.g. ‘..saves time as it is already coded..’ and ‘..you
don’t have to code it yourself..’.

o 2(b)(i) was very well answered with most candidates able to identify
valid, boundary, and invalid test data to gain all 3 marks. Where
marks were lost it was most often due to candidates not properly
reading the question, which asked for ‘numerical data’, and giving
text or another data type as invalid data.

o Question 3(a) was the least well answered question in the paper.
While responses indicated that candidates knew what a trace table
was, most were not able to identify the characteristics of one, with
many responses stating the benefits or purpose of a trace table.
Some candidates did confuse trace tables with truth tables.

o Question 3(b) was very well answered as most candidates could

follow the given algorithm to correctly identify the outputs. A few
candidates lost marks by simply stating a number e.g. ‘2’ rather than
the full output ‘State 2’, but this was rare. The most commonly lost
mark seemed to be with the first output which many candidates gave
as ‘State 3’.

o Candidates were comfortable completing Caesar Cypher encodings,

with most scoring both marks in 4(a)(i) and 4(a)(ii). While 4(a)(ii) was
also quite well done, with over half of candidates getting both marks,
most marks were lost in this part of the question. Most candidates
could identify where the error had occurred; the letter Y being
incorrectly encoded, but many failed to sufficiently explain the
reason. Candidates were expected to identify that the error had been
caused by not handling the end of the alphabet correctly, i.e. should
have moved to letter a. Instead, many assumed that there was simple
human error e.g. ‘they did a -3 instead of a +5’.

o Question 5 was the most difficult written-response question, and

most candidates struggled to score well in this question. Many
candidates seemed unsure of how to present their response for 5(a).
Candidates needed to show the merging of groups of elements in
their response. Many candidates failed to show any kind of grouping
and simply showed a single progressively sorted list but often with no
indication why positions were changing. Most did show some sort of
 grouping but lost a mark by sorting 3 individual elements being
grouped together in one step, rather than the 2 steps that would be
required in a merge sort.

o 5(b) was also not well answered by most candidates. It is difficult to

ascertain whether candidates were not well versed in the use of a
temporary variable for swapping the values inside two existing
variables, or if it was simply too difficult to work out which line could
be changed without the need to change a second line. Some
candidates did identify a change in lines 12 or 13 that could have
been part of a solution, but line 11 was the only line that could be
changed on its own to give a full solution.
Coding response questions

Examiners found that candidates responded very well to the coding challenges presented in
the question paper. There were many candidates who scored close to full marks on these
questions. Most candidates made some attempt at every question.

2(b)(ii) Most candidates scored most or all marks in this question. Many inventive
ways were found of correcting the errors in the code. The answers in the
mark scheme are indicative of expected responses, but any response that
solved the given problem was awarded the marks.
Some candidates reversed the entire selection statement rather than
reversing the condition that compared the stock level to the required level.
Many candidates divided the required stock calculation by -1000 rather
than swap the values in the calculation i.e. ‘weightNeed = (inStock -
requiredWeight) / -1000’, to get both the third and fourth mark-points.
Where marks were lost, it was often as a result of candidates removing the
division by zero, most likely as this caused a run-time error.

3(c) Although this question was generally well answered, candidates seem to
have found it more difficult to implement the logic of a flowchart compared
to similar questions in previous papers which use pseudocode. Many
candidates followed the order of the flowchart and added the code for each
element as they came to it, rather than looking at the algorithm as a whole.
This resulted in many candidates missing the fact that loops were required,
and so only included selection statements. Relatively few candidates had
solutions that successfully repeated until a zero was entered as the base
value. For those that did, most did not have the terminating condition on
the while statement itself (while base !=0), but would use ‘while True’ followed
by a selection statement which would ‘break’ the loop ‘if base == 0’. Both
methods were acceptable for the marks available.
Many solutions lost marks as candidates did not use an appropriate error
message, with many just being ‘error’. Where error messages are requested
in coding tasks, candidates should give some indication to the user of what
the problem is. In this instance an appropriate error message would have
indicated that the exponent must not be a negative number e.g. ‘The
exponent cannot be a negative value’.
Examiners noted that many responses held the user inside a loop until a
positive exponent value was entered. While this was not the logic of the
flowchart, it was encouraging to see effective use of loops here.
Another common mistake came from candidates not properly reading the
question. The second bullet point in the program requirements instructed
candidates that the base and exponent, as well as the answer, should be
included in the final output. Many candidates only output the answer and
omitted the base and exponent values.
4(b) Another question that was generally well answered. Candidates are
comfortable with taking inputs and most were able to cast them to the
correct data types. Most candidates were able to reverse the letters of the
input word which was almost always achieved through string indexing.
Many candidates were also able to get the whole number part of the float
by casting it to an int, however some did lose this mark by leaving the input
as a string and using string indexing to take only the first digit rather than
the whole number. This would not have worked on floats of 10 or higher.
About half of the candidates successfully cast the number parts back to
strings and used concatenation to join the parts of the key together. Some
candidates left all three parts as strings from input and so still got the
marks for concatenating these parts together.
Many candidates were not able to correctly check for a word of exactly two
characters. Most candidates did get the length of the word but many didn’t
seem to be familiar with the not operator (similar to its omission in Q3(c)) so
relied on a more complex condition e.g. ‘if len(word) <= 2 and len(word) > 1’
rather than ‘if len(word) != 2’. This additional complexity often introduced
logical errors which resulted in candidates losing marks.

5(c) Although many candidates did get most or all marks in this question, most
candidates found this question too difficult and gained very few marks as a
result. Most candidates were able to open the file for writing (append mode
was also acceptable) however many forgot to close the file after processing.
The provided data, tbl_tuna, was a two-dimensional array and most
candidates were not able to use 2D indexing to access the individual
elements inside the array. It is unclear whether candidates were not aware
of the structure of a csv file, or if they were hampered by this inability to
access the elements of a two-dimensional array.
Most candidates did use a for loop to iterate over the table, but many
candidates didn’t use the ‘in’ keyword, instead using ‘in range’ set to the
length of the array, which added to the difficulty of achieving a working
program. Most responses that wrote anything to a file consisted of a
number followed by one or more of the 1D arrays inside tbl_tuna.
For those who were able to perform 2D indexing, almost all were able to
correctly construct each line, including commas and line break, and
successfully write these to the file.
6 It was good to see that almost all candidates made a reasonable attempt at
completing this programming task, although a few of those attempts
amounted to little more than hard-coding a series of output statements.
This question required candidates to iterate over several one-dimensional
arrays to combine the data for calculations and outputting. A peculiarity
noted by many examiners was that many candidates started by combining
the data from each of these arrays into a single two-dimensional array
before doing the requested processing on the 2D array rather than the
original arrays.
Most candidates were able to iterate over the arrays, calculate daily milk
volumes per breed and output the required data in rows. Most were also
able to calculate the total daily milk yield either as a running total or using
the sum function on the completed array. Many candidates were not able
to append data to an array, although some created a new array and
appended the data to it rather than in the empty array that was provided.
Most candidates struggled to create suitable code for selecting the best
breed. Some did identify the correct breed, but their code would not work
on alterative data sets e.g. some hard coded the care rating as 1 and only
checked for this value rather than finding the lowest number in the array.
Others hard coded the count-controlled loop rather than finding the lengths
of the given arrays.
A number of methods were used for finding the best breed including
comparisons joined with logical operators, comparisons within nested
selection statements, and values being appended to new arrays and sorted.
For the more capable candidates who were able to create working
solutions, marks were often lost in the levels-based part of the mark
scheme. This was for two main reasons. Most candidates created a new
loop to complete each part of the program rather than completing each of
the tasks within the one loop. We permitted up to two separate loops while
still allowing full marks in that section.
The second common issue was the lack of commenting. It was surprising to
see how many candidates who obtained most marks in this question but
had no comments at all. A very small number of candidates overused
commenting, making code more difficult to read. Good commenting should
be used to separate and explain blocks of code.
Pearson Education Limited. Registered company number 872828
with its registered office at 80 Strand, London, WC2R 0RL, United Kingdom