Mathematics: Paper 9709/11 Paper 1 Pure Mathematics 1

Cambridge International Advanced Subsidiary and Advanced Level
9709 Mathematics November 2021

Principal Examiner Report for Teachers
MATHEMATICS
Paper 9709/11
Paper 1 Pure Mathematics 1
Key messages
The question paper contains a statement in the rubric on the front cover that ‘no marks will be given for
unsupported answers from a calculator.’ This means that clear working must be shown to justify solutions,
particularly in syllabus items such as quadratic equations and trigonometric equations. For quadratic
equations, for example, it would be necessary to show factorisation, use of the quadratic formula or
completing the square as stated in the syllabus. Using calculators to solve equations and writing down only
the solution is not sufficient for certain marks to be awarded. It is also insufficient to quote only the quadratic
formula, candidates need to show values substituted into it. When factorising, candidates should ensure that
the factors always expand to give the coefficients of the quadratic equation.
General comments
Some very good responses were seen but the paper proved challenging for a number of candidates. For this
paper, the knowledge and use of basic algebraic and trigonometric methods from IGCSE or O Level is
expected, as stated in the syllabus.
Comments on specific questions
Question 1
In both parts (a) and (b), it was necessary to evaluate the terms; in some cases this process led to errors in
coefficients or in the position of the x in the middle term in (a). Some candidates omitted the first term, 1, in
part (b).
In part (c), the best answers identified the three combinations of terms resulting in terms in x, added them
together then isolated the coefficient. Weaker responses contained a full multiplication of the two
expansions, but this tended to introduce errors and required the desired terms to be identified.
Question 2
It was essential to equate the curve and the line first, then use the discriminant to form a quadratic equation
in k and solve the inequality. Sign errors were common, leading to the wrong quadratic equation. Some
candidates chose the wrong regions for the inequality while others omitted to solve an inequality, giving only
the critical values.
Question 3
This question required candidates to factorise the trigonometric expression to form two brackets, the simplest
sin θ
method being to factorise in its original form. However, most candidates replaced tanθ with which,
cos θ
although possible to factorise, resulted in a lot more work and few were successful in reaching the two
solutions.
Question 4
In part (a), candidates needed to use the sum to infinity for a geometric series, equate it to the sum of 8
terms for an arithmetic series then solve for a. Some sign and arithmetical errors were seen and it was
common for candidates to use the 8th term formula instead of the formula for the sum.
© 2021
In part (b), a method mark could be awarded for using the correct formula for the 8th term with their value of
a, if there had been errors in part (a).
Question 5
The majority of candidates attempted this question and some fully correct answers were seen. Some
candidates showed working but could not identify numerical values for a, b or c in part (a). The horizontal
stretch proved to be the most challenging of the three transformations.
In parts (b) and (c), few candidates produced correct answers. There was evidence that some candidates
6 6
had used the graph to help them, adding straight lines corresponding to   x and 6 –   x . However,
π π
many candidates did not do so, perhaps because they did not realise they needed to consider the
intersection of straight lines with the curve.
Question 6
π
In part (a), candidates needed to recognise that the angle in the sector/triangle was then calculate the
3
perimeter as the length of two arcs plus 6. A number of candidates were unable to find an angle from the
information given in the question, hence could not make progress.
Part (b) involved using the angle found in part (a) and devising a strategy for calculating the area. There
were several ways to do this: sector + segment, triangle + 2 segments or 2 sectors – triangle. A reasonable
number of fully correct responses were seen for each of these strategies, with candidates generally making it
clear what their chosen strategy was. Those who had no angle in part (a) often made little progress. A
common error was to calculate half the area of the triangle rather than the whole area. Some candidates
misquoted the area formula for the triangle.
Question 7
Many candidates calculated the radius correctly from the points given, then formed an equation for the circle.
Errors seen involved omitting to square the radius or including a negative sign rather than a positive one in
the circle equation. Some candidates found an equation of a straight line which was incorrect.
In part (b) it was necessary to eliminate one variable by substituting the equation of the line, then form and
solve a 3-term quadratic equation to find the coordinates of the two points. Some errors were seen in
manipulating and solving this equation. The final part of this question required candidates to calculate the
length of a line segment, a standard procedure which some of them could not carry out as they had no
coordinates.
Question 8
Many candidates were able to complete the square in part (a) and obtain correct values for a, though fewer
had b correct also. Although the majority of candidates had (x – 2) in their bracket, they did not all relate this
to the maximum value of k in part (b).
Correct answers to part (c) often used a sketch graph to establish that it was possible to calculate the range
by substituting x = –1 into the function, or realised it was one-to-one from the coordinates of the vertex.
Common errors included using the wrong inequality sign or incorrect notation to denote the range.
In part (d), finding an inverse function is a standard algebraic procedure and many candidates were able to
demonstrate this successfully. However, in this case, they needed to realise that the range of f–1(x) had a
maximum of 2 since it was equal to the domain of f(x), hence the positive square root should be discounted.
Few candidates had this expression fully correct.
In part (e), many correct responses were seen that replaced x with x + 3 in either the original form of f(x) or
its completed square form, and added 1 to y. A significant number of candidates either did not know how to
modify the equation or made algebraic errors and could not obtain the correct answer.
© 2021
Question 9
In part (a), many candidates successfully found the equation of the curve, including evaluating the constant
of integration. Those candidates who omitted the constant could not be awarded the final two marks. Weaker
responses contained sign errors.
Part (b) required candidates to form a quartic equation in x which was a hidden quadratic equation and could
therefore be solved using quadratic methods. Candidates who did not show how they solved the equation
could not be awarded a method mark. Successful candidates either factorised the quartic itself or substituted
another variable for x2. Surprisingly, it was common for arithmetic errors to creep in when candidates
calculated y-coordinates. In some cases this was because the candidate had omitted to calculate c in
part (a) while others used f'(x) instead of f(x).
In part (c) many candidates had a correct 2nd derivative, but some had an incorrect power in the second term
and others introduced errors when simplifying their expression.
‘Hence’ in part (d) was a clue for candidates to use the 2nd derivative test with the x-values they had found in
(b). It was necessary to evaluate the expression, give correct values and draw an appropriate conclusion
about whether each of the points was a maximum or minimum. Alternatively, candidates could have used the
first derivative to check a point either side of each of the stationary points; again values were necessary to
justify their conclusions.
Question 10
Some fully correct responses were seen but a number of candidates had difficulty working with the indices in
this question. In part (a), common errors were omitting the 3 in the denominator when integrating or
introducing errors when attempting to simplify their result. Some candidates could not rewrite the expression
for y in a form that they could integrate. Few candidates could complete the solution: using the limits to show
2
that the first term disappeared and so the value of the integral was in fact .
3
In part (b), most candidates omitted to square the expression for y, as required by the formula for integration
to find a volume of revolution. Those who did square the expression often continued to a correct solution or
one with just a coefficient error.
Some fully correct responses were seen in part (c). As in part (a), a number of candidates could not rewrite
5
the expression for y in a form that they could differentiate. Others could not reduce the power to – . Other
2
candidates progressed to finding the gradient when x = 1 but omitted to use the gradient of the normal to find
the y-intercept of this line.
© 2021
MATHEMATICS
Paper 9709/12
Key messages
The question paper contains a statement in the rubric on the front cover that ‘no marks will be given for
unsupported answers from a calculator.’ This means that clear working must be shown to justify solutions,
particularly in syllabus items such as quadratic equations and trigonometric equations. For quadratic
equations, for example, it would be necessary to show factorisation, use of the quadratic formula or
completing the square as stated in the syllabus. Using calculators to solve equations and writing down only
the solution is not sufficient for certain marks to be awarded. It is also insufficient to quote only the quadratic
formula, candidates need to show values substituted into it. When factorising, candidates should ensure that
the factors always expand to give the coefficients of the quadratic equation.
It is very important that a sufficient level of accuracy is used in all calculations. If a result is asked to be
shown to 3 significant figures, then any intermediate calculations need to be given to at least 4 significant
figures. For integer answers candidates should ensure that they simplify them as much as possible rather
18
than leaving them as for example.
3
Questions need to be read very carefully and the phrase ‘it is given instead that’ means that the previous
answers obtained cannot be used in the next part of the question.
It is important that the correct terminology is used when answering transformation questions. Stretch (not
squeeze or compression, for example) should always be given with the correct scale factor and with the
direction clearly indicated. Translation (not move) should be given with a column vector.
General comments
The paper was reasonably well received by candidates and a reasonable number of very good responses
seen. Some candidates though seemed to have struggled with the time available for this particular paper, it
is important that candidates plan their time to allow sufficient time on the later questions. Presentation of
work was mostly good, although some of the answers were written in pencil and then superimposed with ink,
which makes it difficult to read. Candidates are strongly advised not to do this.
Question 1
Most candidates were generally able to make some correct progress and obtain at least one of the two
correct answers for this question. Weaker responses often did not form a 3-term quadratic or did not obtain
the solution –60°.
Question 2
In part (a) many good descriptions of the transformations were given. Some responses showed confusion in
the transformation terms, as some used the word factor in describing a translation or gave the stretch with a
1
factor of 2 rather than , or in the wrong direction.
2
Part (b) was found to be more challenging for candidates with a significant number applying the
transformations to the point given rather than finding the original point on the curve. Many did not seem sure
how to proceed. Candidates should ensure they ready the question carefully and make sure their answer fits
the requirements.
© 2021
Question 3
Both parts of this question were generally well answered. Some candidates made part (a) more complicated
by working out and attempting to simplify ff(x), then substituting 5, rather than finding f(5) then f(2). The
standard technique for finding the inverse for the type of function in part (b) was generally well known.
Question 4
This question was well done by most candidates. The integration was usually good with most candidates
remembering to divide by 3 as well as –1 and the constant of integration was usually found successfully.
Weaker candidates sometimes ignored the word curve and used the equation of a straight line instead of
integrating.
Question 5
Many candidates found this question to be particularly challenging. The trigonometric form of the terms of the
arithmetic progression meant that some candidates were unable to make sufficient progress although the
vast majority did attempt to do so. Some candidates only considered two of the three terms and so did not
obtain the two different expressions or equations needed. Some considered the gaps between the terms as
d rather than 2d and although this was condoned within the method in part (a) it led to incorrect answers in
part (b). Part (b) was often omitted because no answers had been obtained in part (a) and occasionally the
25th term was found rather than the sum.
Question 6
Many fully correct answers were seen for this question with candidates realising the need to form two
equations from the given information, although several candidates did not show their method of solution of
the quadratic equation which was obtained. Weaker responses often confused by the statement that r was
1
greater than and tried to use this as a value in their expressions. The term ‘exact’ was not noticed by
2
some candidates, giving their answer as a decimal. 10 – 1 should have been evaluated to 9 if the final
answer was given in index form.
Question 7
A significant number of candidates omitted this question, in particular part (b) where the given answer from
part (a) could have been used. In part (a) a variety of approaches were taken involving finding different
lengths and angles, but often intermediate working was not given to at least 4 significant figures, and hence
did not sufficiently justify the given answer. In part (b) most candidates attempted to find the area of the
sector BCQ although weaker responses sometimes gave 9 or 15 as the radius. The area of the triangle,
which needed to be subtracted from the sector, could usually be found but some issues of premature
approximation occurred. A few candidates attempted the area of triangle PQC plus the segment.
Question 8
Part (a) was well attempted by many candidates, but part (b) proved more challenging with a significant
number omitting it. In part (a) most candidates realised that two terms from the expansion would give a
coefficient of x2 and were able to form a quadratic equation in a. Weaker responses often obtained only one
term. Sign errors in the expansion were common with negative signs sometimes discarded for no obvious
reason, but the subsequent method marks were still available. In part (b) the phrase ‘it is given instead that’
was misunderstood or ignored in weaker responses that attempted to use the answers from part (a) to
answer this part. Those who understood what was required were usually able to form a quadratic in a and
use the discriminant to find the required value of k and subsequently of a.
Question 9
Many candidates found this question challenging, in particular part (b) and a sizeable minority omitted it
completely. In part (a) most candidates realised the need to differentiate V and this was usually done
correctly although a few left the –1 on the end of their differential. Many realised that they needed to apply
the chain rule, but this was often done incorrectly with candidates not recognising the components correctly
from the given information. A few candidates appear confused confused by the reference to a ‘circular’
© 2021
mound and tried to use formulae for the volume of a sphere or hemisphere, even though the formula for the
volume of the mound was given.
A significant number of candidates tried to link their answers to part (a) with part (b). It is important for
candidates to realise that in this situation it is only the stem part of the question which refers to both parts.
Question 10
The first two parts of this question proved to be straightforward for most candidates. Nearly all realised the
need to differentiate the given function in part (a) and set it to 0, although some weaker responses set the
original function equal to 0 instead. Most managed to solve the resulting equation but there were a number
of errors. In part (b) the vast majority of those who attempted it tried to make use of the second differential. A
small number did not substitute x = 2 into their second differential. Part (c) proved more challenging with
many candidates not realising the significance of the given information. Some put the range equal to a single
value and others made it > , < or ⩽ f(2).
Question 11
Many fully correct solutions were seen to this question, especially part (a). In part (a) most realised the need
to differentiate, and this was often done correctly although some failed to realise the need to substitute x = 3.
In part (b) many candidates realised the need to integrate and this was often done well but the upper limit
was often incorrect with the answer to part (a) common. Many then ignored the triangular part of the area or
made extra work by integrating the normal. A few only integrated the normal.
Question 12
In part (a) many good solutions were seen with candidates realising the need to find the centre of the circle,
the gradient of the normal and then the gradient of the tangent. Some attempted implicit differentiation and
others completed the square and re-arranged to find y = then differentiated, with mixed success. Weaker
responses sometimes used the origin rather than the centre of the circle. Some stopped when they had
found A and B.
Many found part (b) very challenging with almost half of candidates omitting it. Of those who did attempt it,
many realised the need to find the radius but were unsure about how to proceed. Those who used the given
diagram to see which lengths were needed were often most successful, with a number of different
approaches possible.
© 2021
MATHEMATICS
Paper 9709/13
Key messages
Candidates should be encouraged to draw appropriate sketch diagrams, which can aid them in answering
certain types of questions, for example Question 9.
When a question asks candidates to “show” a given result all stages of the process leading to that result
need to be shown.
Candidates are expected to solve quadratic equations by factorisation, formula or completing the square and
their method of choice should be shown clearly in their solution.
General comments
For this paper, the knowledge and use of basic algebraic, including manipulating and form algebraic
expressions, and trigonometric methods from IGCSE or O Level is expected, as stated in the syllabus.
Question 1
In general, candidates used mathematical language well to describe the two transformations. Those who
viewed the sequence as a reflection or stretch by scale factor –1 followed by a translation often scored full
marks. Those who viewed the sequence as a translation followed by a reflection were rarely able to describe
the translation correctly. When describing a translation it should be noted that use of a vector is encouraged
as it avoids the ambiguities of verbal descriptions. Reflections should be accompanied by a clear statement
of the line in which the reflection takes place.
Question 2
(a) This part was well answered by most candidates. Occasionally the first term was omitted and the
power of a was omitted in the third term.
(b) The requirement to obtain two terms in x2 was seen by most candidates and the majority of these
went on to form a correct quadratic equation and find both answers correctly. It is a requirement of
the syllabus that solutions are fully justified so it is important that candidates showed a clear
method to obtain and solve the quadratic equation for this question.
Question 3
(a) Various techniques for completing the square for this type of quadratic were evident and most were
used successfully to find the two required values.
(b) Whilst most candidates appreciated the need for differentiation of f(x) and completed this correctly
only stronger responses showed why this must always be positive. It was expected that the result
from part (a) would be adapted for this purpose but complete methods using the discriminant of the
quadratic in x2 were acceptable although rarely seen. Candidates should realise that showing a
function is positive for a selection of x values does not prove it is positive for all values of x.
© 2021
Question 4
(a) This part was solved in a variety of ways including trial and improvement and use of the nth term
formula. Many correct answers were seen when the correct inequality was formed and solved.
(b) The use of the main two forms of the formula for the sum of n terms was nearly always seen. The
two sums were often formed and equated correctly leading to many correct answers.
Question 5
(a) Almost all candidates were able to correctly find the angle in radians, though a few found it
necessary to obtain the angle in degrees and then convert it to radians.
(b) This part was also often answered completely correctly. Use of rounded intermediate answers was
rarely seen and candidates were able to effectively deal with angles presented in radians both in
the arc length formula and their use of trigonometric functions.
Question 6
(a) This proved to be the most challenging question part for many candidates. Candidates appreciated
a reflection of the given graph was required and a lot of reflections of the graph were seen but few
were reflections in y = x. It was rare to see the line y = x on the candidates’ sketches.
(b) The need to change the subject of the function was appreciated by most but the algebraic skills
necessary to complete the procedure were not always evident. Those who cleared the fraction
before squaring tended to be more successful than those who squared the given function. Errors in
squaring the negative term were frequent enough to be mentioned as were the sign errors in the
gathering of terms. The final correct inverse with the negative alternative of 2x in the numerator
was very rarely seen.
(c) This was answered well by those candidates who appreciated a value of a rather than x needed to
be given.
(d) The correct order of functions for the composite was nearly always used. The initial substitution of
2x was usually carried out well but the factorisation of the denominator and subsequent
simplification was only seen in the better responses.
Question 7
(a) There were several equivalent acceptable variations. Working from the initial statement to the given
result or the reverse of this or showing that the given value of k could be substituted into the final
result to give zero. No matter which method was chosen candidates should be advised that in a
‘show that’ question it is necessary to show all steps taken to obtain the solution. The candidate is
being asked to show, step-by-step, the path from one equation to the other. Some candidates
showed the first three lines in the necessary working and then the final result, which was not
sin θ
enough. It was important that the use of both tan θ = and the standard trigonometric identity
cos θ
sin2θ + cos2θ = 1 and the gathering of terms prior to factorisation was present. Candidates are
advised to work from the given statement to the given result as it will usually be the easiest route.
Those who followed this usually obtained the given result correctly.
(b) Candidates generally followed the instruction of ‘hence’ and used k = 4 in the final equation from
part (a). Candidates are again advised to show a full method for solving the quadratic, in this case
using the formula or completing the square. Many complete solutions were seen with both answers
quoted to the required degree of accuracy.
Question 8
(a) Most candidates found this question to be relatively straightforward. Some combined the two
equations and used one integral whilst others calculated two integrals separately. The fractional
1
indices were mostly correctly integrated and there were few errors when substituting x = . Most
4
© 2021
candidates appreciated that the substitution of the limits should be made clear before the use of a
1
calculator. A very small minority attempted to find a volume of revolution or used limits of 2 and .
2
(b) This part was also well attempted by the majority of candidates. Most were able to differentiate the
fractional index, gaining the first two marks. A large proportion went on to find the gradient and
equation of the normal and the required value of p.
Question 9
(a) The method for finding the x coordinates of A and B was understood by most candidates and the
correct quadratic equation was usually reached. It should be clear that to obtain the roots in surd
form use of the quadratic formula or completion of the square is required. Completion of the square
provided the most direct route to answers in simplified surd form. The formula for the distance
between two points was well known and rarely misquoted leading to many correct answers in
suitable surd form.
(b) This was the most omitted question part. Very few diagrams were seen even though candidates
this would have greatly benefited candidates due to their usefulness in selecting suitable methods.
This was particularly true for this question which could be solved algebraically or through the use of
geometry or trigonometry. Most attempts involved solving the circle equation with the tangent
equation and setting the discriminant of the resulting equation to zero. Although this produced
some more difficult algebraic manipulation correct solutions were often seen. The quickest route to
the solution via the angle the tangent made with the x-axis was rarely seen.
Question 10
(a) The majority of the candidates saw the need for differentiation. The best responses showed that
the differentiation of –(1 + k)–2 with respect to x led to zero and not 2(1 + k)–3. The use of f''(x) >0
was often seen as was the resulting inequality solved by inspection or simple algebra.
(b) Most candidates answered this part very well. Those who did not integrate the constant term with
respect to k usually found the equation correctly. A few candidates chose to work with k rather than
substituting it, which made it slightly more challenging. It was expected that the complete equation
would be stated as the final answer.
(c) A significant number of candidates were able to set the first derivative from part (a) to zero and find
the two solutions. They often reached the correct value of x, found the corresponding y value and
then the nature of the point. A small number of successful candidates did not attempt to find the y-
coordinate and a few did not determine the nature of the point found, as requested by the question.
© 2021
Cambridge International Advanced Subsidiary Level
MATHEMATICS
Paper 9709/21
Key messages
Candidates should be reminded of the need to ensure that they have read the question carefully and met the
full demands of the question to the required level of accuracy. Candidates should ensure they understand
the requirement of an exact answer. Candidates should also recognise that in some questions, work they
have done in previous parts is meant to help them with a solution; for example, in Question 2 and Question
7.
General comments
Some candidates did not appear to be well prepared for the examination, with incomplete questions or
questions that were not attempted. There appeared to be no issues with timing and the majority of
candidates had sufficient room on the paper to answer the questions.
Question 1
Most candidates realised that the form of the integral involved k1e 2 x + k 2e −2 x . There were errors with the
values of k1 and k 2 . Most candidates who obtained an integral of this form were able to apply the limits
correctly although some used their calculators to give a rounded answer rather than the exact answer as
required.
Question 2
(a) Most candidates were able to sketch a V shaped graph together with a straight-line graph. The
positioning of the V shaped graph was occasionally incorrect. It was essential that the gradient of
the straight-line graph was such that there was only one point of intersection with the graph of the
modulus function, this being in the first quadrant. The gradient of the straight-line graph had to be
greater than the gradient of the graph of the modulus function for x > 3 . It was evident that many
candidates did not consider this when sketching the straight-line graph.
(b) Many correct solutions were seen, with candidates who chose to form two linear equations usually
having more success. Some candidates who opted for the method of squaring did not square the
3x term. For those that did, sometimes sign errors or poor algebra resulted in incorrect solutions. It
should also be noted that part (a) was meant to be a guide for part (b) together with the wording of
the question which stated that there was only one point of intersection. Many candidates obtained
two solutions, but some did not relate them to the graph they had sketched in part (a) and did not
go on to discard one of them. Some candidates omitted finding the y-coordinate.
(c) This part asked for candidates to deduce the solution of the inequality, however some candidates
chose to solve the inequality again rather than use their answer to part (b).
Question 3
Very few correct solutions were seen. The main difficulty for candidates was identifying the gradient of the
1
straight-line to , having written the given relationship in logarithmic form. Problems also arose when
ln a
attempting to substitute values in order to find ln k and hence k . It was also acceptable to use the
alternative method of using exponentials but very few responses using this method were seen.
© 2021
Question 4
(a) Provided candidates recognised the need to differentiate the term xe 2 x as a product, they were
usually able to obtain a method mark. Some candidates had difficulties differentiating the
exponential terms correctly. Most realised that they had to equate their derivative to zero and
attempt to rearrange the equation obtained to the given form. It was essential that the equation be
multiplied throughout by e x in order to obtain terms in e3 x . Few completely correct rearrangements
were seen.
(b) Most candidates were very familiar with solving equations using iterations and this question part
was usually well-answered. Candidates had to start with 0.35 and only a few iterations were
needed for convergence. Too many candidates obtained the correct number of iterations to the
correct level of accuracy and gave a final answer of 0.356, therefore did not gain the final accuracy
mark.
Question 5
dx
(a) Many candidates were unable to find correctly with the most common mistake being the
dt
omission of 2 in the numerator. Most candidates realised that the quotient rule had to be used in
dy dy
order to obtain . Most candidates used a correct approach in order to subsequently find .
dt dx
(b) Most candidates realised that they needed to find the value of t when x = 0 in order to find the
gradient of the curve at the point A.
(c) Most candidates realised that they needed to find the value of t when y = 0 in order to find the
gradient of the curve at the point B.
Question 6
(a) Very few incorrect approaches were seen, although some candidates chose to attempt algebraic
long division or synthetic division rather than a straight-forward application of the factor theorem.
Those who chose to use algebraic long division or synthetic division, were usually less successful.
Candidates are advised to check their working to ensure that their solutions they obtain are realistic
or possible.
(b) Very few incorrect approaches were seen to this part, but the best and most accurate solutions
were obtained when the remainder theorem was used rather than algebraic long division or
synthetic division.
(c) Few correct solutions were seen with candidates recognising that x + 3 is a factor of both f ( x ) and
( )
g ( x ) . If the common factor was identified, the final result was often given as ( x + 3 ) 4 x 2 − 1 rather
than as a product of linear factors. For those candidates who had incorrect solutions for part (a),
credit was given if they attempted to find a quadratic factor, using, x + 3 and their values from
part (a).
(d) It was essential that candidates make use of their answer from part (c), ensuring that cosecθ = k
1
where k < −1. Some candidates did not recall that cosecθ = . Few fully correct solutions were
sinθ
seen although some candidates were able to obtain one valid solution.
Question 7
(a) Most candidates were able to use the correct expansion for cos ( 2θ + θ ) although some had
incorrect signs in their expansion. Most made use of sin2θ = 2sinθ cos θ but did not use a correct
identity for cos 2θ . It was essential that sufficient detail be shown in order to gain full marks.
© 2021
(b) Very few correct solutions were seen. Candidates needed to make use of part (a) and write the
1  5π 
given expression as cos3   . This could then be evaluated but an exact answer was needed
2  18 
for full marks.
(c) Very few correct solutions were seen, with some candidates making no attempt at this part.
Candidates needed to make use of part (a) and write the given integral as  ( 9 cos x − cos 9 x ) dx
after simplification.
© 2021
MATHEMATICS
Paper 9709/22
Key messages
full demands of the question to the required level of accuracy. Candidates should also recognise that in
some questions, work they have done in previous parts is meant to help them with a solution; for example, in
Question 2 and Question 7.
General comments
It was evident that many candidates were well prepared for the examination and generally attempted most, if
not all the questions. There appeared to be no issues with timing and most candidates had sufficient room
on the paper to answer the questions.
Question 1
(a) Most candidates recognised the need to use both the factor theorem and remainder theorem to
form two equations. The majority of these equations were correct, but some candidates made
simple errors when solving their correct equations simultaneously. Candidates are advised to
check their working to ensure the solutions they obtain are realistic.
(b) Provided the solutions obtained in part (a) were correct, most candidates made a valid attempt to
obtain a quadratic factor making use of the linear factor given in part (a). In most cases this was
done either by inspection or by algebraic long division. Some candidates gave their final answer as
the product of the correct linear factor and the correct quadratic factor, not realising that the
quadratic factor could be factorised further. For those candidates who had incorrect solutions for
part (a), credit was given if they attempted to find a quadratic factor, using the given linear factor,
and their values from part (a).
Question 2
positioning of the V shaped graphs was occasionally incorrect. It was essential that the gradient of
the straight-line graph was such that there were two points of intersection with the graph of the
modulus function. It was evident that many candidates did not consider this when sketching the
straight-line graph. There should be a point of intersection in the second quadrant and a point of
intersection, or implied point of intersection in the first quadrant.
x + 3 term. For those that did, sometimes sign errors or poor algebra resulted in incorrect solutions.
It should also be noted that part (a) was meant to be a guide for part (b). Many candidates
obtained two solutions but did not relate them to the graph they had sketched in part (a). If this had
been done, then perhaps the sketch in part (a) could have been amended to give two points of
intersection in some cases.
© 2021
(c) Many candidates related this part of the question with the work they had done in part (b) and made
use of their positive answer to part (b). Some candidates chose to start again but invariably made
algebraic errors.
Question 3
A reasonable attempt was made by most candidates to differentiate the given equation with respect to x.
However, errors often occurred in the differentiation of 2 tan2x . These errors were usually with the
dy
coefficient of sec 2 2x or the coefficient of x itself. Although most realised that they had to equate their to
dx
zero and solve, some were unsure of how to deal with the trigonometric term correctly, not realising that
1
sec 2 2 x = . Occasionally the y-coordinate was omitted, hence underlining the importance of making
cos2 2 x
sure that the question has been answered fully. Many candidates did obtain a fully correct solution.
Question 4
The given form of the question did help most candidates realise that logarithms were involved in the
integration. Most errors occurred with the coefficient of the logarithmic term, with many candidates not having
1
the coefficient of . The rules involving logarithms were usually applied correctly but it should be noted that
3
some candidates obtained a fortuitously correct answer by stating that ln ( a + 14 ) − ln a was the same as
ln ( a + 14 )
, for which accuracy marks were not awarded.
ln a
Question 5
Candidates found this question to be quite demanding; candidates had to realise that not only implicit
differentiation was needed but also that a product was involved. Several errors were noted, often which
usually involved omissions of terms or coefficients in the differentiation of the product. Some candidates
equated their result to 6 rather than zero. Most candidates attempted to substitute the given x and y values
into their derivative and fine the gradient of the tangent to the curve. This question highlighted the need for
candidates to make sure that they had answered the question fully, checking to ensure they have fulfilled all
of the requirements. Having obtained the gradient of the tangent, many omitted to find the gradient of the
normal as required. Of the candidates that did find the gradient of the normal, most did leave it in an exact
form as required. When attempting implicit differentiation, candidates should be encouraged not to start their
dy
work with = ... as very often this extra derivative gets considered in the overall working and thus
dx
constitutes an incorrect method.
Question 6
(a) Very few correct sketches were seen. Many candidates did not attempt a sketch. It was intended
that a sketch of the graph of y = ln x together with a sketch of the graph of y = 2e − x be produced
and the single point of intersection be acknowledged as representing a single solution to the given
equation.
(b) Most candidates considered either ln x − 2e − x or 2e − x − ln x with substitutions of 1.5 and 1.6 to
show that a change of sign occurred and so the root lies between 1.5 and 1.6.
(c) Very few correct solutions were seen, with many candidates misunderstanding what was required.
−x
It was intended that the iterative formula given be reduced to x = e 2e followed by the application of
logarithms to obtain ln x = 2e − x .
(d) Most candidates seemed very familiar with solving equations using iterations and this question part
was usually well-answered. Those candidates who started with 1.55 required fewer iterations than
those who started with either 1.5 or 1.6. Many candidates obtained the correct number of iterations
to the correct level of accuracy and gave a final answer of 1.53, thus not gaining the final accuracy
mark.
© 2021
Question 7
(a) Many candidates realised that the use of the double angle identities was needed and obtained
correct solutions. It is essential in questions of this type that sufficient detail is shown. It was
intended that candidates deal with the left-hand side of the given identity and manipulate it correctly
to obtain the right-hand side of the identity. A correct solution starting with the right-hand side was
also acceptable but more difficult to show.
(b) It was essential that candidates realised that they needed to use the result from part (a) to re-write
the integrand before attempting integration. Of those candidates that did, many made reasonable
attempts at integration with errors usually involving signs or coefficients. Most candidates
attempted to give their final answer in exact form.
(c) For this part it was also essential that candidates realised that they needed to use the result from
part (a) to re-write the left-hand side of the equation. It was also important that candidates realised
that they were dealing with 4y not 2y in order to obtain a correct solution. Correct solutions in exact
form were seen.
© 2021
MATHEMATICS
Paper 9709/23
Key messages
full demands of the question to the required level of accuracy. Candidates should also recognise that in
some questions, work they have done in previous parts is meant to help them with a solution; for example, in
Question 2 and Question 7.
General comments
It was evident that many candidates were well prepared for the examination and generally attempted most, if
not all the questions. There appeared to be no issues with timing and most candidates had sufficient room
on the paper to answer the questions.
Question 1
(a) Most candidates recognised the need to use both the factor theorem and remainder theorem to
form two equations. The majority of these equations were correct, but some candidates made
simple errors when solving their correct equations simultaneously. Candidates are advised to
check their working to ensure the solutions they obtain are realistic.
(b) Provided the solutions obtained in part (a) were correct, most candidates made a valid attempt to
obtain a quadratic factor making use of the linear factor given in part (a). In most cases this was
done either by inspection or by algebraic long division. Some candidates gave their final answer as
the product of the correct linear factor and the correct quadratic factor, not realising that the
quadratic factor could be factorised further. For those candidates who had incorrect solutions for
part (a), credit was given if they attempted to find a quadratic factor, using the given linear factor,
and their values from part (a).
Question 2
positioning of the V shaped graphs was occasionally incorrect. It was essential that the gradient of
the straight-line graph was such that there were two points of intersection with the graph of the
modulus function. It was evident that many candidates did not consider this when sketching the
straight-line graph. There should be a point of intersection in the second quadrant and a point of
intersection, or implied point of intersection in the first quadrant.
x + 3 term. For those that did, sometimes sign errors or poor algebra resulted in incorrect solutions.
It should also be noted that part (a) was meant to be a guide for part (b). Many candidates
obtained two solutions but did not relate them to the graph they had sketched in part (a). If this had
been done, then perhaps the sketch in part (a) could have been amended to give two points of
intersection in some cases.
© 2021
(c) Many candidates related this part of the question with the work they had done in part (b) and made
use of their positive answer to part (b). Some candidates chose to start again but invariably made
algebraic errors.
Question 3
A reasonable attempt was made by most candidates to differentiate the given equation with respect to x.
However, errors often occurred in the differentiation of 2 tan2x . These errors were usually with the
dy
coefficient of sec 2 2x or the coefficient of x itself. Although most realised that they had to equate their to
dx
zero and solve, some were unsure of how to deal with the trigonometric term correctly, not realising that
1
sec 2 2 x = . Occasionally the y-coordinate was omitted, hence underlining the importance of making
cos2 2 x
sure that the question has been answered fully. Many candidates did obtain a fully correct solution.
Question 4
The given form of the question did help most candidates realise that logarithms were involved in the
integration. Most errors occurred with the coefficient of the logarithmic term, with many candidates not having
1
the coefficient of . The rules involving logarithms were usually applied correctly but it should be noted that
3
some candidates obtained a fortuitously correct answer by stating that ln ( a + 14 ) − ln a was the same as
ln ( a + 14 )
, for which accuracy marks were not awarded.
ln a
Question 5
Candidates found this question to be quite demanding; candidates had to realise that not only implicit
differentiation was needed but also that a product was involved. Several errors were noted, often which
usually involved omissions of terms or coefficients in the differentiation of the product. Some candidates
equated their result to 6 rather than zero. Most candidates attempted to substitute the given x and y values
into their derivative and fine the gradient of the tangent to the curve. This question highlighted the need for
candidates to make sure that they had answered the question fully, checking to ensure they have fulfilled all
of the requirements. Having obtained the gradient of the tangent, many omitted to find the gradient of the
normal as required. Of the candidates that did find the gradient of the normal, most did leave it in an exact
form as required. When attempting implicit differentiation, candidates should be encouraged not to start their
dy
work with = ... as very often this extra derivative gets considered in the overall working and thus
dx
constitutes an incorrect method.
Question 6
(a) Very few correct sketches were seen. Many candidates did not attempt a sketch. It was intended
that a sketch of the graph of y = ln x together with a sketch of the graph of y = 2e − x be produced
and the single point of intersection be acknowledged as representing a single solution to the given
equation.
(b) Most candidates considered either ln x − 2e − x or 2e − x − ln x with substitutions of 1.5 and 1.6 to
show that a change of sign occurred and so the root lies between 1.5 and 1.6.
(c) Very few correct solutions were seen, with many candidates misunderstanding what was required.
−x
It was intended that the iterative formula given be reduced to x = e 2e followed by the application of
logarithms to obtain ln x = 2e − x .
(d) Most candidates seemed very familiar with solving equations using iterations and this question part
was usually well-answered. Those candidates who started with 1.55 required fewer iterations than
those who started with either 1.5 or 1.6. Many candidates obtained the correct number of iterations
to the correct level of accuracy and gave a final answer of 1.53, thus not gaining the final accuracy
mark.
© 2021
Question 7
(a) Many candidates realised that the use of the double angle identities was needed and obtained
correct solutions. It is essential in questions of this type that sufficient detail is shown. It was
intended that candidates deal with the left-hand side of the given identity and manipulate it correctly
to obtain the right-hand side of the identity. A correct solution starting with the right-hand side was
also acceptable but more difficult to show.
(b) It was essential that candidates realised that they needed to use the result from part (a) to re-write
the integrand before attempting integration. Of those candidates that did, many made reasonable
attempts at integration with errors usually involving signs or coefficients. Most candidates
attempted to give their final answer in exact form.
(c) For this part it was also essential that candidates realised that they needed to use the result from
part (a) to re-write the left-hand side of the equation. It was also important that candidates realised
that they were dealing with 4y not 2y in order to obtain a correct solution. Correct solutions in exact
form were seen.
© 2021
Cambridge International Advanced Level
MATHEMATICS
Paper 9709/31
General comments
Some candidates showed a good understanding of the topics examined. The majority of candidates offered
no response to several items and showed only limited understanding of basic algebra and arithmetic.
Key messages for candidates
• Candidates are reminded to ensure they are confident with the basic rules of algebra and arithmetic and
use brackets when appropriate.
• Practise the basic methods in calculus to ensure that you are familiar with all the standard techniques
and patterns.
• Ensure that working out is displayed clearly in the answer space and that all the stages of the working is
shown.
• In the case where candidates wish to replace their solution, neatly cross out the solution and re-attempt
the question below this. Candidates should ensure that they do not overwrite one solution with another.
• If a question asks for an exact answer, then decimal working is not appropriate and exact numbers are
expected throughout the working.
• Candidates are reminded to read the question carefully and make sure that their response matches
what is being asked.
Question 1
The majority of candidates started their solution by squaring the equation and multiplying out the brackets.
Errors such as 4 × 5 x = 20 x and ( 5 x ) = 252 x were common. Candidates who got as far as 3 × 5 x = 4 did not
2
always obtain a correct value for 5 x.
Question 2
(a) Some candidates were familiar with the processes required here and there were some correct
solutions. Those candidates who found the value of α first and used that to find R were not often
able to state the exact value of R.
(b) This question was testing the candidates’ knowledge of the function R sin ( x − a ) . A few wrote
down the correct answers but the majority either did not respond or offered irrelevant working.
Question 3
(a) There were several correct attempts to use the product rule to differentiate this function. Obtaining
the correct coordinates for the turning point required both the correct derivative and correct
algebra.
(b) The question states that the curve has one stationary point, so the candidates had a wide choice of
methods here. They could consider the sign of the second derivative, which was a popular option,
or they could consider the value of y or the value of the gradient at another point on the curve.
Candidates who chose one of the latter two approaches often scored no marks because although
© 2021
they calculated values, they did not say whether these were values of y or values of the gradient,
and they did not draw clear conclusions.
Question 4
du
Some candidates were clearly familiar with the method required, they started with a statement of , or
dx
equivalent, and attempted to form an integral in terms of u. Some candidates made algebra slips in the
process and some stopped part way through. Very few obtained 2 tan−1 u.
Question 5
(a) The candidates who attempted this question usually made good progress in forming an equation in
tanθ and rearranging this to form a quadratic in tanθ . However, this was not always correct, often
due to slips in the algebra and arithmetic.
(b) Candidates with a correct quadratic equation usually obtained at least one correct angle. Some
candidates who obtained a negative angle did not go on to use the periodic nature of tanθ to
obtain the obtuse angle.
Question 6
The majority of candidates made no attempt to expand 1 + 4 x . Those candidates who did attempt the
expansion usually obtained a relevant pair of equations in a and b which they were able to solve.
Question 7
(a) The answer here was given, so candidates needed to present convincing evidence that they
dy 1 1 1
understood how to obtain the result. The minimum expected here was = × = .
dx x ln x x ln x
(b) Those candidates who started by substituting the boundary condition did not get as far as
appreciating the relevance of part (a). A small minority of candidates did start by separating the
variables correctly and making some progress with the integration.
(c) Most candidates offered no solution to part (b) or they had a solution of an incorrect form, therefore
found it difficult to answer this part as it required them to interpret their solution.
Question 8
(a) A small number of candidates recognised the need to use integration by parts. There were some
errors in the coefficients, but these candidates usually made good progress towards demonstrating
the given result.
(b) The most successful candidates compared the value of a with the value of exp ( 1
a )
+ 2 for
a = 9 and a = 11 and drew the correct conclusion. Some candidates mentioned making an
appropriate comparison but did not provide any values to support their answer.
(c) Many candidates did not respond to this part, which suggested that these candidates were not
familiar with using an iterative process. This particular formula worked equally quickly with 9, 10 or
11 as the starting point. A small number of candidates used a starting point of 10.12, which
suggested inappropriate use of calculator functions to solve the equation. Candidates who did not
work to the level of accuracy requested did not score the accuracy marks.
Question 9
(a) Those candidates who demonstrated correct use of the scalar product of the direction vectors of
the two lines usually scored both marks.
© 2021
(b) Some candidates were familiar with the method for finding the point of intersection of two lines. The
two common errors were to overlook the need to verify that the equations for all three components
were satisfied, or to show that the two lines do intersect but not state the point of intersection.
(c) To make any progress with this part of the question, a candidate needed to start by considering a
general point P on the line and using the scalar product to determine the position vector of P; very
few candidates used a strategy that did this.
Question 10
(a) The first three parts of this question are all inter-related. The approach taken by the minority of
candidates who attempted the question was to find the square and the cube of u and substitute
these into the equation p ( u ) = 0. With this approach it was then necessary to compare the real and
imaginary parts of the equation and solve the resulting simultaneous equations to obtain a and b.
(b) Some correct answers were seen to this part. The common incorrect answer was −1 − 2i.
(c) Very few candidates demonstrated knowledge of how to use the pair of complex roots to form a
quadratic factor with real coefficients.
(d) (i) Candidates who knew how to represent the two regions on an Argand diagram usually drew a
good, clear diagram.
(ii) In order to obtain the correct answer candidates needed to understand where to look on the
diagram, and that the answer would be negative. There were a small number of correct answers.
© 2021
MATHEMATICS
Paper 9709/32
General comments
Many candidates offered solutions to all eleven questions, however the standard of the responses seen was
very variable, with some responses demonstrating a clear lack of understanding of the content of the
specification.
Questions which are seen to be generally be more straightforward for candidates, such as solution of a
modular inequality (Question 2), partial fractions (Question 4) and use of an iterative formula (Question
11(c)) did not earn many marks. In questions where the first part gives an indication of how to solve the
second part, such as complex numbers (Question 3), integration of a trigonometric function (Question 6),
and solution of a trigonometric equation (Question 8), many responses did not suggest that the link between
the parts had been recognised.
Key messages for candidates
• Candidates are reminded to ensure they are confident with the basic rules of algebra and arithmetic and
use brackets when appropriate.
• Practise the basic methods in calculus to ensure that you are familiar with all the standard techniques
and patterns.
• Ensure that working out is displayed clearly in the answer space and that all the stages of the working is
shown.
• In the case where candidates wish to replace their solution, neatly cross out the solution and re-attempt
the question below this. Candidates should ensure that they do not overwrite one solution with another.
• If a question asks for an exact answer, then decimal working is not appropriate and exact numbers are
expected throughout the working.
• Candidates are reminded to read the question carefully and make sure that their response matches
what is being asked.
Question 1
The majority of candidates recognised the need to use logarithms, and many solutions contained some
correct use of the rules of logarithms. Several candidates reached the correct solution. A common error was
ln2 + ln3
to start by claiming that 3 ( 21− x ) = 61− x. Some candidates with correct working got as far as x = and
ln2 + ln7
stopped without completing the question. A few who reached this stage then ‘cancelled’ the ln 2 to obtain an
answer of the required form, which was not a correct step.
Question 2
The most common approach was to start by squaring both sides of the inequality to obtain an expression
free of modulus signs. Many candidates obtained the correct quadratic equation for the critical values. A few
went on to obtain the critical values of x in terms of a. Some did not reach the correct final answer as they
3
thought that since a is a positive constant then the critical value x = − a was not possible. The majority of
5
candidates who obtained the correct quadratic equation made no further progress because they tried to
solve b 2  4ac rather than to solve for x. The most common reasons for not reaching the correct quadratic
were to forget to square the 2, or to make errors in squaring the brackets and simplifying.
© 2021
A small number of candidates tried the alternative approach of considering two linear equations, some of
which were successful. The main errors in using the alternative method were slips in the arithmetic and the
algebra.
Question 3
(a) Candidates found this question to be particularly challenging, with many candidates not
communicating a clear proof. Some candidates did not understand the notation u * for the
conjugate of u. Some responses made no mention of the conjugate at all, and claimed incorrect
statements such as a + ib + c + id = a − ib + c − id . Many solutions appeared to assume the given
result and to be working on both sides of the equation at the same time, rather than working on the
two separately and concluding that they were equal.
(b) A minority of candidates realised that they needed to substitute z = x + iy and then use the result
from part (a). Those candidates who did this often went on to consider the real parts and the
imaginary parts of their equation and complete correctly. A few equated the real part to x and the
imaginary part to y, which was an incorrect approach. Many candidates did a lot of work trying,
unsuccessfully, to form a solvable equation in z.
Question 4
Candidates who recognised this as a top-heavy fraction often answered the question correctly. For some, the
most common errors were slips in the arithmetic.
4 x 2 − 13 x + 13 A B
The majority of candidates started with the incorrect form = + and only scored a
( 2 x − 1)( x − 3 ) 2 x − 1 x −3
A Bx + C
maximum of 2 marks. It is possible to solve the problem in two stages by starting with + . Some
2x − 1 x − 3
candidates completed the first stage correctly, but few of them went on to split the top-heavy fraction.
Question 5
(a) Several fully correct Argand diagrams were seen. Some candidates made the task more difficult by
using different scales on the two axes, whilst some candidates used inconsistent scales. The most
common errors were to draw a circle in the wrong quadrant, to draw a circle when the scale
required an ellipse, or to draw the wrong straight line. On correct diagrams the shading was usually
correct, if it was shown.
(b) Many candidates offered no response to this item. There were a few fully correct solutions, usually
following a clear diagram with a clearly marked tangent. Some candidates made errors in the
trigonometry, using tan−1 13
1
in place of sin−1 13
1
. The question asked for an answer in degrees, so
an answer in radians was not awarded the final mark.
Question 6
(a) Most candidates started with a correct expansion, usually for sin ( 2 x + 3 x ) . Several then followed
1
the expected route of substituting the two expansions in ( sin5 x + sin x ) and deriving the given
2
result within two or three lines of working. A significant minority started by trying to use
sin ( 2 x + 3 x ) = sin ( 2 x − 3 x ) .
(b) The result in part (a) and the use of ‘hence’ in the question were intended to help candidates with
this integral, however many did not make use of this. There were some errors with the coefficients
and some errors in using the limits, but several reached the given answer correctly. When the
answer is given, candidates are expected to show sufficient working to make it clear that they have
reached the answer correctly; some candidates went direct from the first stage of substitution to the
given answer with no interim working. A large number of candidates tried to use integration by
parts to evaluate the integral, but none of them succeeded in completing this.
© 2021
Question 7
The correct approach here is to separate the variables, integrate both sides of the resulting equation and
then use the boundary condition. A large minority of candidates started by trying to use the boundary
condition and were unable to make any progress. Several candidates recognised the correct form for
1 1
 y 2 dy. The common incorrect answers were ln y and 2y ln y . The majority of candidates recognised the
2 2
need to use integration by parts for the second integral. Many had the correct approach, but there were
errors in the coefficients, and e −2 x often became e 2 x . Some candidates had no constant of integration, and
there were several errors in taking the reciprocal to find the expression for y.
Question 8
(a) Candidates found this part to be particularly challenging. Many candidates equated
( cos θ + sin2 θ ) to zero. Only a small minority started by equating it to one, and a few more
2 2
concluded that it was 1 after expanding and rearranging the terms. Some candidates did succeed
1
in justifying 2sin2 θ cos2 θ = sin2 2θ , but many derived the result from a succession of incorrect
2
statements.
(b) The majority of candidates did try to use the result from part (a). The candidates found potential for
error at each stage of rearranging the resulting equation. 2θ often became θ during the working,
and some candidates thought that sin2 2θ = sin ( 2θ ) , so they obtained an angle and then took the
2
square root. It was common for candidates to give more than one decimal place in their answers,
often rounding incorrectly. Most candidates obtained at least one correct value, but it was unusual
to see all four values. Some candidates ignored the ‘hence’ in the question. They were often able
to reach a correct equation in cos θ or sinθ , but completely correct solutions were rare.
Question 9
(a) Most candidates used the correct method in answering this part. Candidates are reminded to be
careful to ensure the correct use of brackets and to pay careful attention to signs to ensure marks
can be gained. Candidates with errors in their working often claimed to have reached the given
answer without trying to find the error in their working. Several of the weaker responses
demonstrated that they were not familiar with implicit differentiation.
(b) Many candidates attempted to find the point where the gradient was equal to zero, and several
looked for the point where the gradient was equal to 1. Those candidates who interpreted the
question correctly often made good progress. Those who formed and solved an equation in y were
often more successful than those who found the value of x first. The unsimplified answer x = 31 ln8
was common and made the task of solving for y more complex than necessary.
Question 10
(a) The majority of candidates started correctly by forming a vector in the direction of the line. Several
reached an incorrect k component because of the error 1 − ( −1) = 0. Most candidates attempted to
use the correct structure for the equation of the line, but many responses concluded ‘line = ...’ or
‘equation = ...’ or ‘ l = ...’ rather than the correct form ‘r = ...’.

(b) Many candidates formed a correct expression for AC but a lot did not go on to use this to find the

position vector OC.
(c) but did not have a correct strategy to find P.

Many candidates found the lengths of various vectors
Some candidates who gave a correct expression for OP in component form did not go on to use
the length 14. Most of the candidates who formed a correct equation in λ did complete the
question correctly, but there were several errors in the algebra and arithmetic between the initial
equation and the final answers.
© 2021
Question 11
tan x as ( tan x ) 2 often then used the chain rule to

1
(a) Those candidates who started by writing
obtain a correct form of the derivative. Some used sec 2 x = 1 + tan2 x to obtain an answer in tan x,
as asked for in the question. The final mark, for verifying that the derivative was equal to 1 when
x = π4 was available from any correct form of the derivative. Many candidates did not score this
mark because, despite showing their expression with x = π
4
substituted, they did not show any
evidence of working to obtain the given answer.
(b) Candidates found this part to be particularly challenging. Rather than starting with the equation
dy
= 1 and squaring this to form an equation in t, most candidates spent a lot of time trying to
dx
manipulate the given cubic equation. Obtaining the given answer to part (b) does depend on
starting with a correct expression from part (a), but most candidates did not adopt a strategy where
this became relevant.
(c) The question does not specify a starting value for the iteration. Although the graph given in the
question clearly indicates that a lies between 0 and π4 , many candidates used a starting value
outside this interval. For most of the values chosen, the iteration does eventually converge. Some
candidates were not working in radians, and many candidates keyed in the formula incorrectly.
Candidates with a correct sequence of iterations often rounded the final answer incorrectly.
© 2021
MATHEMATICS
Paper 9709/33
General comments
The standard of work on this paper was extremely high. A considerable number of candidates performed well
on many of the questions.
It was clear that most candidates had taken notice and understood the instructions in the rubric and including
sufficient detail in their solutions to demonstrate that they had not used a calculator.
Candidates should be reminded to set out their work clearly to ensure it is legible and presented in a logical
manner.
Key messages
• Candidates should be aware of how to use a sketch of an inequality graph to aid them in checking their
solutions when solving inequalities, see Question 2(a) and (b),
• Candidates should ensure they are familiar with basic techniques, specifically using the chain rule which
was particularly relevant for Question 7(a) and 9(a),
• Candidates should be aware of the meaning of the word ‘hence’, this is used to indicate that a solution
or information from previous part should be used to answer the question.
• Candidates should ensure that they show adequate detail when solving equations, in particular when
solving the equation ax = b,
• It is recommended that candidates use the same scale on both axes when sketching an Argand
diagram.
Question 1
Most candidates opted to solve by long division as opposed to inspection using

2x4 + 1 = (x2 − x + 2)(Ax2 + Bx + C) + Dx + E. Many excellent solutions were seen, but a considerable
number of candidates either did not correctly obtain the constant term in the quotient or made an arithmetical
error in one of the terms in the remainder.
Question 2
(a) The sketch required symmetry, both arms of approximately the same length and the points (0, 3)
3 
and  , 0  shown. A large number of candidates believed that since it was |2x − 3| then x < 0 was
2 
not possible and stopped their sketch at the y-axis.
(b) The sketch in part (a) could have been used in part (b) to determine which of the two lines to solve
with y = 3x + 2, and hence to establish the correct region. Instead, most candidates either created
1
a quadratic equation or two pairs of linear equations. As this led to having two critical points x =
5
and x = −5, this often resulted in candidates making the wrong choice.
© 2021
Question 3
Some candidates found this question to be particularly challenging. Candidates who scored zero did not
4x
express 4x−2 as and obtain an expression for 4x. These candidates often tried to apply log laws as
16
162 162
opposed to laws of indices. Having 4 x = candidates needed to proceed with xln4 = ln( ) and then
15 15
obtain x to 3 d.p.
Question 4
This question required integration by parts and was generally well answered. However, many candidates did
1 
not integrate sin  x  correctly, with the 2 in the denominator instead of the numerator or made a sign error
2 
1 
in integrating cos  x  . Substitution of limits into the trigonometrical functions needed sufficient detail to
2 
1 3
establish that the correct trigonometrical values such as zero, , and 1 had been used. Only a few
2 2
candidates gave a decimal solution instead of the requested exact answer.
Question 5
Most candidates gained either 4 or 5 marks on this question. Those who gained 4 marks often either forgot
the solution from sin θ = −1 or did not realise that the solution from the other sin θ value also had an answer
in the second quadrant: 180° − 56.4°.
Question 6
(a) Many candidates produced fully correct solutions to this trigonometrical problem. It was rare to see
an incorrect expansion of cos(x − 60°) and values of sin(60°) and cos(60°) were almost always
exact and correct. The main error seen was expressing R as a decimal rather than an exact value.
(b) Candidates found this part to be particularly challenging with many incorrect responses seen. Many
candidates believed that cos(x − α) was least when the argument was 0°, 90°or 270°, instead of at
180°. Candidates are recommended to use a sketch of the cosine function, which could have
helped to avoid these errors.
Question 7
(a) Most candidates found using the chain rule within the implicit differentiation of the original
 1 + dy 
 dx 
equation,   , to be challenging. Some candidates removed the ln by exponentiating so that
(x + y )
the chain rule was only required on e2y, and this proved slightly easier. In addition, many
candidates manipulated the original equation into other alternative forms before differentiating.
However, this approach often led to errors which could have been avoided had they differentiated
the equation in its original form.
(b) Candidates generally found this part to be straightforward. Nearly every candidate successfully set
the given answer from part (a) to zero and substituted it in the original equation of the curve.
Realising ln(1) was zero meant that most candidates scored full marks for the coordinates.
Question 8
(a) Several candidates found this question to be challenging. Many candidates thought that the vector
line equation for MN was in fact only the vector MN. Candidates should also be reminded that the
left side of the vector line equation of MN should be a vector and symbolised by the vector r.
© 2021
(b) The majority of candidates realised that setting the scalar product of r from part (a) and the vector
MN to zero and solving the resulting equation produced the required point on the line MN. Most
candidates then found the length of this vector by Pythagoras to obtain the given answer.
Alternative approaches were either to use a scalar product to find the projection of OM or ON onto
MN, followed by Pythagoras, or to find either cos PMO or cos PNO then determine OM sin PMO or
ON sin PNO.
Question 9
−1
 1 3
 1
(a) Solutions were possible via the chain rule using  9x 2 − x 3  , the quotient rule using
   2 1 3

 9x − x 
2
 
1
−
or the product rule using (9 − x )−1 x 2 . All these were methods regularly seen. However, many
attempts using the product or quotient rules saw errors by having incorrect signs in the middle. In
the case of the quotient rule, some candidates did not square the denominator of the formula when
substituting. The three different approaches then required different amounts of algebra to
determine the x-coordinates of the stationary point. Some solutions needed only a couple of lines,
others considerably more. When undertaking the extra algebra, care must be taken if squaring
since this often introduces values that are not solutions of the original equation and need to be
rejected.
du
(b) Most candidates obtained correctly, but many then made errors in substituting and hence did
dx
not obtain the correct integrand in terms of u. Of those that were successful about half realised this
was an integral in the formulae book and so they did not need to show all the partial fraction
details. However, if quoting from the formula book care is needed, especially in this case. Instead
 1   (3 + u )   1   (3 + x ) 
of seeing   ln   with limits u = 0 and u = 2, it was common to see   ln   with
3
   (3 − u )   3   (3 − x ) 
 1   (u + 3 ) 
limits x = 0 and x = 4 or limits x = 0 and x = 2, or   ln   . The latter led to the appearance
 3   (u − 3) 
of ln of negative numbers which needed to be handled carefully using modulus signs.
Question 10
(a) Many candidates were unable to change the proportionality information into an equation involving
an arbitrary parameter, hence were unable to generate the given equation. Those that could do so
had no difficulty using the two pieces of information.
(b) Most candidates were able to separate variables correctly and to integrate with respect to t. Many
(20 − x ) (20 − x )
candidates divided throughout by 19 and hence finished with as opposed to . The
19x x
19 appearing in two terms led to considerable confusion and many candidates were unable to
(20 − x )
integrate correctly with respect to x. Several other candidates did not see that could be
x
20
expressed as , producing the easiest of integrals, instead using integration by parts. This
x −1
meant that they needed to know how to integrate lnx and ensure that their work was error free
otherwise the two xlnx terms did not cancel out. Some candidates who did integrate correctly forgot
to introduce a constant of integration. Candidates needed to reach a correct solution to their
differential equation and then manipulate that solution into the given form to gain the final mark,
however this was not often seen. Candidates needed to realise that being given x = e0.9 + 0.05x
meant that it was necessary to write lnx = f(x) and then exponentiate.
(c) This iteration was almost always error free, with convergence clearly shown. As the initial value
was given in the question, this needed to be used by candidates for all of available marks to be
awarded.
© 2021
(d) Of the candidates that had obtained a correct solution of their differential equation in part (b), many
did not set x = 20 and solve for t.
Question 11
(a) Whilst most candidates were able to obtain the modulus correctly, this was not the case with the
π 5π
argument, with far more common than . Most candidates’ solutions resulted in a choice of
6 6
two answers, so they needed to sketch an Argand diagram and plot the given point u in order to
select the correct value.
(b) This part started with the word ‘Hence’, this means that the candidate need to continue with the
6
 i5π 
result they found in part (a). The question is requesting that  2e 6  becomes 26ei5π = −64.
 
Candidates should not restart with the binomial expansion of (− 3 + i)6, as this was not answering
the question being asked.
(c) (i) To represent complex numbers satisfying equations and inequalities on an Argand diagram it is
essential that the scales on the Real and the Imaginary axes are identical. For the locus Re z  2,
the line x = 2 needed to be shown and the point (2, 0) clearly labelled. For the locus
π π
0  arg(z − u)  , half lines needed to be drawn from point u at arg z = 0 and arg z = , with
4 4
π
the coordinates of u and the angle clearly indicated.
4
(ii) Of the candidates who obtained the correct shaded region in part (c)(i) many, despite realising that
the shaded triangle was isosceles, omitted to add unity to the imaginary coordinate of z. Hence,
gave the greatest |z| as √ (22 + ( 3 + 2)2) instead of √ (22 + ( 3 + 3)2).
© 2021
MATHEMATICS
Paper 9709/41
Paper 4 Mechanics
Key messages
• Non-exact numerical answers are required correct to 3 significant figures or angles correct to 1 decimal
place as stated on the front of the question paper and cases where this was commonly not adhered to
were seen in Questions 3, 4, 5 and 7. Candidates are strongly advised to carry out all working to at
least 4 significant figures if a final answer is required to 3 significant figures.
• When answering questions involving an inclined plane, a force diagram could help candidates to include
all relevant terms when forming a Newton’s Law equation or a work/energy equation. This was
particularly noticeable in Questions 4 and 7.
• In questions such as Question 6 in this paper, where velocity is given as a quadratic function of time,
then calculus must be used and it is not possible to apply the equations of constant acceleration.
3
• In cases such as Question 3, where an angle is given in the form sin α = then there is no need to
5
4
determine the angle explicitly since it is simple in this case to show that cos α = and these values
5
can be used in any calculations. If the angle is evaluated this may lead to rounding errors. The angle
given in Question 5 can also be treated similarly.
General Comments
There were some excellent responses seen for each of the questions on paper. Overall, a wide range of
performance was seen but the paper was generally well answered.
Some candidates did not give answers to 3 significant figures as requested on the front of the question
paper, mainly due to prematurely approximating within their calculations leading to the final answer. In
Question 3 the sine of an angle is given and in Question 5 the angle is given in terms of inverse sine. As
mentioned in the Key Messages, in such questions it was not necessary to determine the actual angle as this
often leads to premature approximation and to a loss of accuracy.
On the front cover of the question paper it tells candidates take g = 10 and most candidates are correctly
following this instruction.
Comments on Specific Questions
Question 1
(a) The most straightforward approach to this problem is to draw and use a velocity-time graph. This
consists of a trapezium shaped graph. Since the total distance travelled corresponds to the area
under the graph, the given distance of 585 m can be equated to the area of the trapezium and this
information can be used to set up an equation in V, the constant speed. This method was adopted
by most candidates. Solving this equation gives the required speed directly. This question was well
answered.
(b) The required deceleration can be found as the gradient of the final stage of the velocity-time graph.
Since the magnitude of the deceleration is required, the value stated must be positive.
© 2021
Question 2
(a) This part involves the use of the principle of conservation of momentum. The two particles are
moving in opposite directions and so the momentum of the particles before collision will have
opposite signs. In the resulting collision the particles coalesce and so their velocity after collision
will be the same. Most candidates were able to write down the required momentum equation and
solve this to find k. A common error was to use the momentum of each particle before collision with
the same sign.
(b) This question required candidates to evaluate the kinetic energy of the particles before and after
1
collision. Use of the formula KE = mv 2 will enable the required kinetic energies to be found. Most
2
candidates made a successful attempt at this part of the question.
Question 3
In this question it is necessary to express the the forces are in equilibrium by resolving forces horizontally,
along the direction of the 20 N force, and vertically, perpendicular to the 20 N force. This will produce two
equations in the unknown variables P and θ. The value of sinα is given in the question and so the actual
angle is not required to be found. If the equations are rearranged to make Psinθ and Pcosθ the subjects of
the equations, then the value of θ can be found by dividing these two expressions and using trigonometry.
P can then be determined by using the angle found for θ. Most candidates used this method. Common errors
were to mix sine and cosine of the angles and also some sign errors were seen. Accuracy was sometimes
lost if the actual angle α was found rather than to use the given value of sin α and the corresponding value of
cos α.
Question 4
(a) This part of the question asked for a force diagram. All forces and any angles must be shown in
such a question. Here this means showing the weight as a vertically downwards force of magnitude
12 g, the force P N acting parallel to the plane, the friction force, F, also acting parallel to the plane
and the normal reaction, R, acting perpendicular to the plane. The 25° angle must also be shown.
All forces should be shown with arrows indicating their direction of action. The friction force, F,
could be shown either parallel to the plane acting upwards or downwards as this depends on the
direction of motion of the particle. Many candidates did not show enough detail in terms of the
directions and in some cases forces were omitted.
(b) The question asked for the least possible value of P. This will be when P is applied up the plane
with the aim of stopping the particle from slipping down the plane and so if the motion began it
would be down the plane. This means that the friction force, F, will be acting up the plane in the
same direction as the force P. Many candidates incorrectly thought that F acted down the plane but
this leads to the other limiting case of the particle being about to slip up the plane. It is necessary to
resolve forces along and perpendicular to the plane. As the particle is in limiting equilibrium the
friction force is given by F = 0.35 R. Most candidates made a good attempt at this question with
common errors being a mix of sine and cosine and the direction of friction taken to be in the wrong
sense.
Question 5
(a) This question required candidates to firstly determine how far the car travels up the plane in 30
seconds. Once this distance, d, has been determined then the potential energy gained in this time
can be found as PE = 1600 × g × d × 0.12. Most candidates used the correct definition and made a
good attempt at this part of the question.
(b) In this part of the question the work-energy conservation property is required using the PE gain
found in part (a). The equation takes the form work done (WD) by the engine = WD against friction
+ PE gain. This enables the WD against friction to be found and then the constant frictional force
can be found by dividing this WD by the distance, d, travelled in 30 seconds. Some good attempts
were seen but errors in sign or forgetting to divide by d were most common.
© 2021
(c) In this part of the question the power can be found in a variety of ways although the most
straightforward method is to use the formula Power = WD by the engine ÷ Time. Most candidates
made a good attempt at this question.
(d) In this part of the question Newton’s second law must be used along the direction of the road. Once
the new reduced power is found the driving force, DF, can be found by using DF = P ÷ v. Other
forces acting are the resistance force and the component of the weight along the direction of the
incline. The effect of these three forces causes the deceleration. Some errors seen here included
missing forces and incorrect evaluation of the new power.
Question 6
(a) In this question the velocity is given as two different functions of t over the two given time ranges. A
condition is given that the acceleration at t = 2 is zero. In order to determine the values of the
unknowns p and q it is necessary to set up two simultaneous equations. The first of these comes
from using the condition on the acceleration. Differentiating the velocity in the range 0 < t < 6 and
using the condition that a = 0 at t = 2 gives the first equation. The second equation comes from
matching the velocities at t = 6 since it is given that there is no instantaneous change in velocity.
These two equations can then be solved for p and q. Most candidates performed well on this
question. Errors were sometimes seen in the manipulation and the solving of the simultaneous
equations.
(b) In this part of the question a velocity-time graph is required. Once p and q are found it is clear that
the shape of the two graphs are quadratic over the first six seconds and linear from t = 6 to t = 14.
It is necessary to determine the time at which v = 0 which shows when the quadratic crosses the
t-axis. Most candidates correctly sketched the linear part, but the quadratic was often incorrectly
positioned.
(c) Here the total distance is required and it is important to use the velocity time graph found in part
(b). Use can be made of the fact that the area under the v-t graph represents the distance travelled
provided that care is taken with areas above and below the t-axis. The quadratic graph is seen to
be below the t-axis from t = 0 to t = 4 and above the t-axis from t = 4 to t = 6. The linear graph is
above the t-axis from t = 6 to t = 14. The distance travelled from t = 0 to t = 6 must be split into two
parts. Integration of v gives the displacement and this will be negative from t = 0 to t = 4 and
positive from times t = 4 to t = 6. Provided that the signs of these two values are considered the
distance travelled from t = 0 to t = 6 can be found. The area under the graph from t = 6 to t = 14 is
readily found by evaluating the area under the triangle formed by that part of the v-t graph. The
error most often seen was to integrate the quadratic from t = 0 to t = 6 without considering the sign
of v.
Question 7
This question involves two connected particles, one on an inclined plane and the other hanging freely and
there is no friction involved. One method of approach is to apply Newton’s second law parallel to the plane
for particle B and in the vertical for particle A. For both particles there are two forces acting. For particle A it
is the weight of the particle and the tension in the string. For B it is the component of the weight along the
direction of the plane and the tension in the string. The tension is the same throughout the string but some
wrongly used different tensions acting on each particle. When the two equations of motion are stated, they
produce two equations each involving the tension T and the acceleration a. These equations can now be
solved simultaneously to find the value of a. Alternatively, since only a was needed, the equation for the
system can be found and many candidates took this approach. Once the value of this acceleration is
determined, the constant acceleration equations can be used to find the speed of the particles as A reaches
the ground. At this point A remains at rest but B continues to move up the plane until it comes to
instantaneous rest. The only force acting on B during this second phase of motion is its weight. Applying
Newton’s second law enables the deceleration to be found. Once this is known, the equations for constant
acceleration can again be applied and the distance travelled up the plane until B comes to instantaneous
rest can be found. The main cause of error was candidates using the wrong component of weight parallel to
the plane namely 3gcos18 for particle B or not using a component but using 3g parallel the plane when
finding a in the first part of the motion. For the motion after A reaches the ground, an error seen was to use
the deceleration as g rather than a component of g. Most candidates used this method.
© 2021
An alternative approach to this problem is to use energy methods for either or both parts of the motion. Some
candidates attempted this successfully. Some tried to solve the problem by considering only particle B but if
this was attempted then the work done by the tension also needed to be considered.
© 2021
MATHEMATICS
Paper 9709/42
Paper 4 Mechanics
Key messages
• Non-exact numerical answers are required correct to 3 significant figures or angles correct to 1 decimal
place as stated on the front of the question paper and cases where this was not adhered to were seen
in Questions 5, 6 and 7. Candidates are strongly advised to carry out all working to at least 4 significant
figures if a final answer is required to 3 significant figures.
• When answering questions involving an inclined plane, a force diagram could help candidates to include
all relevant terms when forming a Newton’s Law equation or a work/energy equation. This was
particularly noticeable in Questions 5 and 7.
• In questions such as Question 4 in this paper, where displacement is given as a cubic function of time,
then calculus must be used and it is not possible to apply the equations of constant acceleration.
General comments
There were some excellent candidates who produced very good answers on this paper.
Some candidates did not give answers to 3 significant figures as requested on the front of the question
paper, mainly due to prematurely approximating within their calculations leading to the final answer. This was
often seen in Questions 5, 6 and 7. In Questions 5 and 7 the sine of an angle is given in the question. In
such questions it was not necessary to determine the actual angle to 1 decimal place as this often leads to
premature approximation and a loss of accuracy.
On the front cover of the question paper it tells candidates take g = 10 and most candidates are correctly
following this instruction. In some cases it is impossible to achieve a correct given answer, such as in this
paper in Question 3, unless this value is used.
Question 1
(a) The majority of candidates gained full marks for this part of the question. It was necessary to use
the fact that the acceleration is given by the gradient in a velocity-time graph. The accelerations in
the two stages from t = 0 to t = 5 and the from t = T to t = 50 must be equated to form an equation
in T. Some candidates set up the equation but then made errors with their algebra so did not obtain
the correct value for T.
(b) Almost all candidates correctly attempted to find the area beneath the velocity-time graph with
many different approaches taken on this part. The most common approach was to find the area of
the six different stages and add. Some tried to combine areas by finding areas of trapeziums. An
error seen was to wrongly consider the area under the curve from t = 0 to t = 30 as a trapezium.
A very small number of candidates calculated the area of the large trapezium, from t = 0 to t = 50,
and subtracted the area of the smaller trapezium, from t = 25 to t = 50; this was the best approach.
Question 2
(a) In this question most candidates attempted to write down the equation of motion either for the van
or the trailer or for the system. Two of these equations were needed in order to solve for the
required tension. Some errors seen included using the resistance for the trailer into the equation for
© 2021
the van or vice versa. Another error seen was to include a tension into the equation for the whole
system.
(b) In this part of the question many candidates did not use the fact that that T = 0 and hence could not
make any progress. In fact, many used values obtained in part (a) but this was no longer
applicable.
Question 3
(a) In this question many candidates incorrectly used constant acceleration formulae which cannot be
used for motion which does not take place on a straight line. Conservation of energy methods must
be used here and those candidates who realised this generally achieved full marks.
(b) In this part of the question energy methods must be used. There is an initial kinetic energy and
during the motion this is lost to friction and to a gain in potential energy. Most candidates who used
energy methods included these three effects but in several cases these were combined with
incorrect signs.
Question 4
(a) This part of the question was well done by the majority of candidates who differentiated the given
expression for s in order to find the velocity. It was necessary to set this expression to zero to find
when the cyclist comes to rest. Particular care must be taken in questions such as this where the
answer is given, making sure that full working is shown. Most candidates found the time at which
the cyclist came to rest and then used this in the given expression for s in order to find the distance
AB. A few errors were seen in the differentiation but overall most candidates obtained the given
answer without error.
(b) Most candidates realised that the maximum velocity took place when the acceleration is zero and
that this required further differentiation of their expression for velocity. Some used the symmetry
property of the quadratic expression for velocity in order to find the time at which maximum velocity
occurred. Again, this part was well answered by the majority of candidates.
Question 5
(a) From the given information many candidates correctly used the formula F = P ÷ v to find the
constant driving force acting on the engine. The sine of the angle of inclination is given and so the
angle need not be found. However, some wrongly interpreted the angle and used sin0.01 in their
working. Some wrongly attempted to include an acceleration in their working. Most candidates
correctly used the equilibrium of the three forces, namely, driving force, the effect of the component
of the weight and the required resistance force.
(b) In this part of the question many candidates did not identify that an energy approach must be used.
The work done by the engine in 60 seconds can be found using WD = Power × Time. This work
done by the engine has two effects. It has to overcome the given WD against resistance and also
to increase the kinetic energy. The work energy equation is used to express this fact. Some
candidates incorrectly attempted to find a resistance force and use Newton’s second law but it is
given that the resistance is not constant and so this method cannot be used.
Question 6
(a) Many good responses were seen to this part of the question. Using the equilibrium of forces
horizontally and vertically was a common approach. This led to two simultaneous equations
involving the two tensions and these can be solved to find the required tension. A commonly seen
error was to assume that the two tensions were equal. By considering the equilibrium in the vertical
it is clear that this cannot be the case. Some candidates attempted to use Lami’s equations but
there are four forces acting here and so it is not possible unless two of the forces are combined.
Since only the tension in the lower string is required then a very good method for solving this
problem is to resolve forces perpendicular to the upper string which gives the required result in one
equation. A few candidates did adopt this approach.
© 2021
(b) Candidates found this part of the question to be particularly challenging. This was mainly due to not
realising that the lower tension must be set to zero in this case. Very few candidates successfully
solved this problem. The solution can be found by resolving vertically and horizontally. However,
since the lower tension is zero this means that in this case it is possible to solve using Lami’s
equations.
Question 7
(a) This question involves application of Newton’s second law to particle Q and including the friction
effect. Care must be taken here to show sufficient detail within the working as the answer is given
in the question. One of the common errors made here was to mix the m and 2m masses of the
particles within the same equation. The specific angle of the plane need not be found as sinα = 0.8
is given and cosα can easily be shown to be cosα = 0.6. Most candidates made a reasonable
attempt at this part, including the two force terms, the weight component and friction and equating
to mass × acceleration.
(b) In this part almost all candidates did not realise that particle Q comes to rest and then moves down
the plane before the two particles meet. Particle P moves down the plane from rest and it can be
shown that it has acceleration 4.4 m s–2 and this can be used to find the distance sP travelled by
particle P until collision. It is necessary to find the distance sQ travelled by particle Q until collision.
Once this has been done use of the equation sP + sQ = 6.4 enables the required time to be found.
Since many candidates misinterpreted this question, a special case mark scheme was used to
ensure that these candidates were not unduly disadvantaged by this error.
(c) In this part it is necessary to determine the velocities of the two particles at the instant at which the
collision takes place. Once these have been found then use of the conservation of momentum
equation in the form muP + 2muQ = 3mv enables the required speed, v, of the combined particle to
be found. When candidates correctly found uP and uQ, the majority were able to correctly apply
conservation of momentum. A number of candidates incorrectly used the initial velocities of zero
and 10 as their values of uP and uQ.
© 2021
MATHEMATICS
Paper 9709/43
Paper 4 Mechanics
Key messages
• Non-exact numerical answers are required correct to 3 significant figures as stated on the front of the
question paper rather than correct to 2 decimal places as sometimes seen for example in Question
4(a)(iii).
• When resolving forces in equilibrium or when forming an equation of motion, a clear and complete force
diagram can be helpful in ensuring that all relevant forces have been considered for example in
Question 1(b), Question 4(a)(iii) and 4(b).
• When two separate strings are attached to a particle, the two tensions should be assumed to be
different unless the situation suggests otherwise, e.g. Question 2 and Question 6(a) and (b).
General comments
Many responses of a high standard and clearly presented solutions were seen. There were also several
candidates who had difficulty attempting some questions. Question 1(a) and Question 4(a)(i) and (a)(ii)
were found to be the most straightforward questions for candidates whilst Question 6(b) and (c) were the
most challenging.
Question 1
(a) Nearly all candidates applied ‘conservation of momentum’ correctly to solve this problem. Very few
omitted the momentum of the object after impact, 8 × 120 = 40v, or mistakenly attempted to apply
‘conservation of energy’.
(b) A suitable method was usually attempted, either applying Newton’s second law and a ‘suvat’
formula or forming and solving a work/energy equation. A common error was to omit the weight
160g when forming the equation of motion, –4800 = 160 a. In some solutions, either the mass of
the object or the mass of the post was used instead of the combined mass,
e.g. 120 g – 4800 = 120 a or 120 g – 4800 = 40 a. When applying v2 = u2 + 2as to calculate the
distance required, the initial velocity was sometimes taken to be 8 ms–1, which was the speed
before rather than after the impact. For the work/energy method the main errors were to omit a
1
term from the equation for example 4800 s = × 160 × 62 (no potential energy), or to include an
2
1 1
extra term or a sign error, e.g. × 160 × 62 − × 160 × 82 = 1600 s − 4800 s .
2 2
Question 2
(a) Whilst many complete force diagrams were seen, there were also a variety of incorrect or
incomplete diagrams. The weight was sometimes either omitted or labelled as mass 8 kg. A normal
reaction was occasionally seen even though the particle was suspended. Some diagrams indicated
the same tension in each string. Others omitted to show the direction if each force.
(b) The majority of candidates resolved the forces in equilibrium to find the two different tensions
accurately even following incomplete diagrams in part (a). A few tried to resolve assuming that the
© 2021
tension in each string was the same leading to Tsin45 + Tsin60 = 80 in this oversimplified situation.
The use of a 3 significant figure approximation in working does not ensure 3 significant figure
accuracy in a final answer, candidates are advised to work to at least 4 significant figures
throughout their intermediate working. In this case the use of 41.4 N for the tension in one string led
to 58.5 N rather than 58.6 N for the other tension.
Question 3
(a) The most direct method in answering this question was to equate the initial potential energy minus
8 J to the final potential energy. Many candidates misread or misinterpreted the loss of kinetic
energy on hitting the ground and equated 8 J to the final potential energy to obtain h = 0.5 m. A
correct maximum height could still be obtained from 5 – 0.5 = 4.5 m, but this complete method was
rarely seen. Those who mistakenly increased the energy by 8 J on impact with the ground found an
increased maximum height of 5.5 m. Less direct methods were also regularly seen, for example,
using an energy equation to calculate the velocity after impact with the ground and a constant
acceleration formula, usually v2 = u2 + 2as, with acceleration –g m s–2 to find the height.
1 2 1
(b) Candidates who used s = gt to find the time taken to reach the ground and s = − ( −g ) t 2 to
2 2
find the time from the ground to the maximum height were generally more successful than those
who involved the velocities at the ground. Whilst the time taken for the drop was frequently correct,
the time for the rise was less straightforward. The initial velocity for the ball rising was sometimes
used as 10 m s–1, the velocity on impact with the ground rather than the changed velocity following
the energy loss. It was also common to see the use of v = 10 m s −1 suggesting that the kinetic
1 
energy at the ground  × 1.6 × v 2  was 8 J rather than 80 J before and 72 J after impact.
2 
Question 4
(a) (i) The work done against the resistance was usually calculated successfully as WD = DF × v × t. This
was also seen correctly as P × t. Occasional errors included the use of an incorrect formula such
v P
as WD = DF ´ or WD = .
t t
(ii) The solution for this part was usually fully correct with only a few candidates answering in watts
instead of the required kW.
(iii) Most candidates adjusted the driving force for the increased power as necessary, although a few
45 000 - 12 000 45 000 + 12 000
decreased using instead of . When forming the equation of
36 36
57 000
motion, the resisting force 1250 N was sometimes missing e.g. = 1400a . An alternative
36
method seen was to consider only the increased power causing the acceleration and to solve
12 000
= 1400a . The final answer was sometimes seen as 0.24 m s–2 correct to 2 instead of 3
36
significant figures.
(b) The inclined plane required a component of weight down the plane whilst the constant speed
implied no acceleration. When resolving the forces along the plane, errors seen included the use of
1400sinθ instead of 1400gsinθ leading to θ = 32.4, or the omission of the resistance 1250 N
leading to θ = 8.2. A few candidates overlooked ‘constant speed’ and included ‘ma’ in their
equation using the acceleration from part (a)(iii).
Question 5
(a) This question was well answered by the majority of candidates who understood that integration
was needed and fully correct solutions were often seen. Only a few candidates attempted to use
1 1
( )
constant acceleration formulae, e.g. s = ut + at 2 = kt 2 16 − t 2 . Most integrated and then
2 2
© 2021
calculated k successfully following the substitution of t = 4 and v = 8. Those who included a

constant of integration C without recognising that C = 0, were unable to evaluate k. Others
integrated twice and then inappropriately used the given answer to find the value of k rather than
finding k in order to show the answer given.
(b) Candidates were expected to solve s(t) = 0 in order to find the time taken and hence the speed on
return to O. Some assumed that a = 0 rather than s = 0 on return. The most common error was to
find the velocity correctly as –29.4 m s–1 but omit to state the speed 29.4 m s–1, as was required by
the question.
(c) Maximum displacement required the solution of v(t) = 0 to find the time taken. As in part (b) some
candidates mistakenly used a = 0 leading to t = 4 and s = 20.
Question 6
Part (a) was generally well answered but parts (b) and (c) proved to be particularly challenging for many
candidates.
(a) This was a straightforward question for most candidates who were able to apply F = μR
appropriately to obtain the given answer. When the answer is given in the question, sufficient detail
is expected. Reference to the 40 N and 60 N weights was expected before reaching an equation
such as 20 = 50μ. A few candidates considered mass instead of weight showing 6 – 4 = 5μ. Others
incorrectly included a tension, e.g. 60 – T + T – 40 = 50μ.
(b) The solution was expected to involve three simultaneous equations for the three particles. The
situation was frequently oversimplified by the assumption that the tensions in the two strings were
equal. This usually resulted in the solution of 8g – T = 8a and T – 4g = 4a with no consideration of
the 5 kg mass and the friction between the table and this particle. Some erroneously included the
frictional force in the equation of motion for the 4 kg or 8 kg mass, e.g. T – 4g – 0.4 × 5g = 4a. The
acceleration could be found directly from the equation of motion for the whole system
8g – 0.4 × 5g – 4g = 17a with the two tensions then found from T1 – 4g = 4a and 8g – T2 = 8a.
(c) This solution depended on recognising that the acceleration after the 8 kg particle hit the ground
needed to be calculated either from a system equation for the two particles in motion or from
solving simultaneous equations of motion for the 4 kg and 5 kg particles. Many candidates
assumed either that the acceleration was –g ms–2 or the same as found in part (a). Those who
recalculated the acceleration, sometimes used a faulty equation, e.g. –4g – 20 = 5a instead of
–4g – 20 = 9a or –T – 20 = 5a. Some calculated different accelerations for each of the two particles
despite their connection and then found and combined two different times taken. It was common to
see a calculation for the time taken for the 8 kg mass to reach the ground or for the velocity at this
time with no further progress.
© 2021
MATHEMATICS
Paper 9709/51
Paper 5 Probability & Statistics 1
Key messages
Candidates need to have an understanding of all the topics in the syllabus for this component.
Candidates should be aware of the need to communicate their method clearly. Simply stating values often
does not provide sufficient evidence of the calculation undertaken, especially when there are errors earlier in
the solution.
Candidates should state only non-exact answers to 3 significant figures, exact answers should be stated
exactly. It is important that candidates realised the need to work to at least 4 significant figures throughout to
justify a 3 significant figure final value. The only exception is if a value is stated within the question. It is an
inefficient use of time to convert an exact fractional value to an inexact decimal equivalent, there is no
requirement for probabilities to be stated as a decimal.
General comments
Candidates appeared to have sufficient time to complete all the paper. Many candidates did not appear to
have sufficient knowledge of the topics for this component, with some not attempting any questions.
The use of simple sketches and diagrams can help clarify both context and conditions. These were
frequently present in good solutions.
Candidates would be well advised to enhance their interpretation of success criteria, such as ‘fewer than 2
days’, as this is essential when answering probability questions.
It is good practice to read the question again after completing a solution to ensure that all the requirements
have been fulfilled. As this is an applied mathematics component, comments about data should be related to
the context of the question.
Question 1
Although many candidates did identify that the criteria for a geometric approximation was fulfilled, there were
a significant number of solutions which used the binomial approximation. This was sometimes successful in
part (b) when all the possible scenarios that did not fulfil the success criteria were identified.
Many candidates did not interpret the success criteria accurately, which is a fundamental expectation of this
component. Candidates would be well advised to practice this process in preparation for the examination.
Many answers were stated to 2 significant figures or 3 decimal places, rather than 3 significant figures as
required.
(a) The best solutions identified the possible outcomes for the two coins to find the probability of
obtaining two tails. The geometric approximation was then used and evaluated accurately with the
probability stated as a fraction. Weaker solutions misinterpreted the criteria and found the
probability of obtaining two tails on the 8th throw. The weakest attempts used the binomial
approximation and often found the probability of having 1 success in 7, or 8, throws.
© 2021
1 1
A number of candidates stated that the probability of throwing two tails was , or occasionally ,
2 3
and used this value appropriately.
(b) Better solutions recognised that the question used a standard property of the geometric
approximation and used this accurately to answer efficiently. Weaker solutions used a binomial
approximation approach and calculated the probability of obtaining two tails for each number of
throws up to 9 throws and subtracting from 1. Errors in interpreting the success criteria resulted in
either the omission of the 9th throw, or the inclusion of the 10th throw.
Question 2
This question involving coding was found challenging by many candidates. The standard formulae were not
seen in many of the solutions. Work was often presented without clear structure, and candidates did not
always communicate their logic within their solution.
(a) The best solutions used the relationship ‘the mean of coded values = the mean of the original
 (x − k )  x 
values subtract the coding value’,  = − k  , with the given values substituted followed
 40 40 
by a clear algebraic solution. Many candidates were successful using a simple problem solving
logic approach, although there was often little communication about the steps that were being
undertaken or the reasoning behind them.
(b) A significant number of candidates made no attempt at this part. This part simply required the data
provided to be substituted into the standard coded data variance formula. Good solutions did state
the formula before substituting. A number of solutions either did not square the coded mean or
simply used the original mean in their calculation.
Question 3
The majority of solutions identified that a conditional probability was being calculated and made some
progress towards the expected value. Solutions with a tree diagram were often more successful, as this
enabled the information provided to be clearly displayed within the required context. The best solutions
stated the required conditional probability formula before substituting the unsimplified probability calculations.
Weaker solutions often had difficulties in using the information ‘when Suki has milk, she never has a biscuit’
within the context, and it was often omitted from the expected denominator.
Question 4
Many good solutions to this probability question were noted. Candidates did show appropriate supporting
work for the variance, often calculating E(X) initially before substituting in the variance calculation.
Candidates should be aware that there is no expectation to convert probabilities stated accurately as
fractions to inaccurate decimal values.
(a) Most successful solutions contained an outcome space to identify the possible outcomes for the
two spinners. It was noted that working with negative numbers was an occasional problem, and
some unexpected values were obtained as possible outcomes. Weaker solutions often identified
the possible outcomes and then stated each had a probability of 0.2.
(b) Candidates who had produced a probability distribution table in part (a) were usually able to
complete this part successfully. Good solutions included the unsimplified arithmetic expressions for
both E(X) and Var(X), with all values left as fractions. Weaker solutions often did not square E(X),
or square the probability rather than the outcome in the variance formula.
© 2021
Question 5
Many candidates found this question on permutations and combinations challenging. The use of simple
diagrams to help clarify the given scenario is often helpful and seen in more successful solutions. As the
number of arrangements and selections is always exact, candidates should be aware that the final answer
should not be rounded to 3 significant figures.
(a) Many good solutions to this question were seen. The best included a simple diagram to illustrate
the scenario, which can aid understanding of the process needed. Almost all attempts included the
unevaluated factorial term as well as stating the accurate value. A few solutions provided an
answer rounded to 3 significant figures rather than the exact value that is expected. A common
error was not excluding Raman from the group when calculating the arrangements. A small number
of solutions calculated 8! + 1, presumably to account for the one way that Raman can appear in the
arrangements. Some better responses considered the arrangements as being two groups of four,
which is true, but needed to be treated as 8P4 × 4P4 rather than 4P4 + 4P4.
(b) Some excellent solutions were presented for this question. These usually included a simple
diagram to clarify the scenario. The most successful approach was to consider the total number of
arrangements that the team could make and subtract the number of arrangements where Raman
and Sanjay were together. A number of candidates provided explanations of what was being
calculated being included in the solution. The alternative approach was often less successful as
many did not allow Raman and Sanjay to swap places and used 8C2 rather than 8P2 when
calculating how many ways they could be inserted into the line of other seven team members.
Again, some candidates stated only a rounded final answer rather than the exact number of
arrangements.
(c) Although many solutions identified that a group of five and a group of four was required, the
calculation presented by many assumed that the groups were picked independently, and all team
members were available for the second group that was being selected. The best solutions realised
that once the first group was selected, all the remaining team members had to be in the second
group – and so could be ignored in the calculation. A few solutions used permutations rather than
combinations, so implying that the order of the team members was to be considered.
(d) Many candidates found this question challenging, and a significant proportion of candidates
presented no attempt. The best solutions realised that part (c) had calculated the value for the
denominator of the probability and used a similar approach to calculate the number of ways that
Raman and Sanjay were in the same group. Common errors were to consider the number of ways
they could be in just one of the groups, usually the group of five, not reducing the number of team
members the group could be picked from if they had already been selected or finding the product
rather than the sum of their answers. Many solutions simply calculated the number of ways the
groups could be selected and made no attempt to use this information to find the probability.
Question 6
The overall quality of the statistical diagrams presented was good, although a number of candidates did not
use a ruler when drawing the box-and-whisker plot, which does make interpreting the information presented
more difficult.
(a) Many excellent back-to-back stem-and-leaf diagrams were seen. The best solutions used a ruler to
align the stem initially, included units for both teams in the single key and ensured that the
individual data values were inserted into the leaves equally space, so that the diagram was a good
visual representation of the data spread and skew. Weaker solutions included errors where the
corrections made the vertical alignment inaccurate. Candidates should appreciate that as a
diagram, it is recommended to be completed in pencil, so that errors can be erased and replaced. If
it is necessary to cross-through, then any replacement value should ideally be place above the
error.
(b) Most candidates were able to find the median accurately. The most common error when finding the
interquartile range was not finding the mid-values between the median and the maximum/minimum
n
values, with the use of being seen regularly.
4
© 2021
(c) Many good box-and-whisker plots were seen. A small number did have the whiskers incorrectly
passing through the box. Some candidates had values from part (b) where the median was not
between the quartiles, which they attempted to plot here. Candidates should be aware that a ruler
should be used to construct these plots to ensure that the anticipated degree of accuracy can be
achieved.
(d) Many candidates simply made comparisons of individual values, for example the greatest weights
rather than interpreting the statistical information displayed in parts (c) and (a). At AS and A level,
comparisons are normally related to the range, central tendency or skew and must be made in
context. Good solutions often identified that the players from Rebels tended to be heavier than the
Sharks.
Question 7
This question was found challenging by a large number of candidates, and some parts were not attempted in
a significant number of scripts.
Although the question identifies that the original data has a normal distribution, candidates should be aware
that other approximations may be required because of the given context. A number of solutions attempted to
use the normal distribution in part (a)(ii), when the success criteria had evolved to discrete data and a
binomial approximation was appropriate.
(a) (i) Almost all solutions started by using the normal standardisation formula to find a probability. The
best solutions often had a sketch of the normal distribution curve and identified the probability area
which fulfilled the given criteria. These solutions also identified that the original data for the times
was continuous and so did not require a continuity correction. A probability to at least 4 significant
figures was then used to calculate the expected number of days when Karli would spend more than
142 minutes on social media, with a final conclusion of an integer value of days, with no indication
of rounding. Weaker solutions either showed a rounding process for the number of days, or simply
omitted this process. A significant number of solutions misinterpreted the success criteria and
found the probability of spending less than 142 minutes. Candidates are well advised to spend time
in ensure that they are able to interpret this type of success criteria accurately.
(ii) To be successful, candidates needed to use a result that they had found in part (a)(i) in a context
which required a binomial approximation. Although the actual mathematics required was quite
standard, few solutions were fully correct. Common errors were not using the probability calculated
in part (a)(i) or misinterpreting the success criteria by including two days in their solution. A
significant number of candidates submitted no attempt at this question.
(b) Although a significant number of candidates made no attempt at this question, many good solutions
were seen. The most successful often used a simple diagram of the normal distribution curve to
help identify the required probability area. As 90 per cent is a critical value, candidates needed to
use the stated critical z-value from the normal table rather than using the main tables or a
calculator to find a z-value. Most solutions realised that the normal standardisation formula needed
to be used to form an equation. Common errors were to equate to the probability of z = 0.9, or
z = 1.282 which would find the time of 10 per cent of the days spending more than t minutes on
social media. The majority of equations formed were solved appropriately with sufficient evidence
to justify the stated answer. It was noted that some solutions had accuracy errors from either using
the tables or rounding from the calculator inaccurately.
© 2021
MATHEMATICS
Paper 9709/52
Key messages
Candidates should be aware of the need to communicate their method clearly. Simply stating values often
does not provide sufficient evidence of the calculation undertaken, especially when there are errors earlier in
the solution. When errors are corrected, candidates would be well advised to cross through and replace the
term, rather than overwrite it.
Candidates should state only non-exact answers to 3 significant figures, exact answers should be stated
exactly. It is important that candidates work to at least 4 significant figures throughout to justify a 3 significant
figure value. The only exception is if a value is stated within the question. It is an inefficient use of time to
convert an exact fractional value to an inexact decimal equivalent, there is no requirement for probabilities to
be stated as a decimal.
The interpretation of success criteria is an essential skill for this component. Candidates would be well
advised to include this within their preparation.
General comments
Although many well-structured responses were seen, candidates would be advised to use the response
space in a logical and clear manner. Better solutions often included some simple notation to clarify the
process that was being used.
The use of simple sketches and diagrams can help to clarify both context and information provided. These
were often seen in successful solutions. Candidates should be aware that cumulative frequency graphs are
constructed with a curve, and that this needs to be reasonably accurately drawn. It was noted that many of
the statistical diagram were well labelled.
Sufficient time seems to have been available for candidates to complete all the work they were able to,
although some candidates did not complete the very last question. A few candidates found some topics and
skills required particularly challenging, including being able to differentiate between the binomial and
geometric approximation requirements. Many good solutions were seen for Questions 3 and 5. The context
in Questions 2, 4 and 6 was found to be challenging for many.
Question 1
(a) Almost all candidates simply stated the required probability, although good solutions included the
unsimplified calculation. The most common error was to truncate rather than round the decimal
probability, which if not supported by an acceptable fraction could gain no credit.
(b) Most candidates recognised that the conditional probability was required. The most efficient
method was to identify the values required in the data table and state the probability directly. The
P (M ∩ D )
most common method was to use the standard formula P (M|D ) = , and find the required
P (D )
probabilities from the data table. A common error was to assume that M, male, and D, drums, were
independent and use P(M) × P(D) to calculate P(M ∩ D).
© 2021
(c) Some good solutions to this part were seen; these stated the appropriate test in context, clearly
identified the required probabilities and made a numerical comparison before stating the
conclusion. The majority of solutions omitted one or more of these elements, and so not fully
justifying their conclusion. Almost all checked the identity P(F ∩ G) = P(F) × P(G), but a small
number of candidates used the alternative approach involving conditional probability.
Question 2
This combinations question was found challenging by many candidates. The best solutions often contained
simple ‘diagrams’ which clarified the conditions and possible outcomes. A few solutions assumed that the
condition applied in part (a) was also required in part (b). Candidates should be aware that when the
question is structured as here, then the conditions are only applied within the single part.
(a) Most candidates deduced that the group required five women to be chosen from the group of 11
and used 11C5 as expected. A significant number of solutions stated this as the final answer, while
better solutions multiplied by 4 with no justification. The best solutions included a statement to
show that one man was being picked from the group of four men. A small number of solutions
incorrectly added rather than multiplied these values.
(b) A variety of approaches were seen. The most successful was to calculate the total number of ways
that a group of six people could be chosen from the full group and subtract the number of groups
which contained Jane and Kate. Candidates who used the more common approach of considering
the different scenarios which fulfilled the condition often omitted the scenario when neither Jane
nor Kate were present. Better solutions identified the value that was being calculated at each
stage. It is expected that the unevaluated combination calculation is stated within the working to
support the solution.
Question 3
Many good solutions to this probability question were seen. Candidates should be reminded that there is no
expectation for probabilities to be stated as decimal values. Although not required by the question, a tree
diagram was often present in more successful solutions, as it enabled the success criteria to be identified
easily and supported the necessary calculations.
(a) As this part required a statement to be shown to be true, candidates should be aware that full
justification for any calculations used need to be provided. Where a tree diagram had been
generated, simply identifying the required branches would be sufficient to show that there were
three possible outcomes that fulfilled the success criteria. Better responses often used the
combination method for calculating probabilities successfully. Weaker solutions simply had a
calculation that provided the required value, with no justification and often not related to the
context.
(b) A high number of candidates made no attempt at the probability distribution table. As one value
had been provided in part (a), it was anticipated that the initial table would include at least one
correct probability. Solutions where a tree diagram had been used in part (a) often utilised this to
support the work here. Poorer solutions did not seem to relate the work attempted in part (a) as
guidance to complete the table. A common error was to omit the outcome 0 yellow marbles
selected, and so the probabilities in the table did not sum to 1. A few solutions included 4 or 5
yellow marbles picked as outcomes, which did not apply the condition given in the initial question
information.
At this level, individual probabilities greater than 1 should highlight to the candidate that there is an
error in their work that should be checked.
(c) Where a probability distribution table had been completed, most were able to calculate the value.
Candidates should be aware that answers need to be supported by an unevaluated calculation,
which here enabled E(X) found from incorrect probability distribution tables to gain credit.
© 2021
Question 4
This question was found to be challenging by many. Good solutions often included simple diagrams which
clarified the conditions that were being applied. Weaker solutions frequently had poor communication of the
logic that was being applied to determine the number of arrangements that fulfilled the criteria.
(a) Almost all candidates were successful in determining the correct number of arrangements and
providing sufficient evidence of the calculation performed. Weaker solutions often did not remove
the effect of the repeated letter E in the word.
(b) The arrangement criteria appeared to be challenging for many. A large number of solutions had
little indication why calculations were being attempted, and how they linked to the question. The
most successful solutions had a simple diagram such as ‘^ ^ ^ ^ T ^ ^ C ^’ initially and then stated
that there were seven remaining letters when the T and C had been removed, that the T and C
could be located in six different positions (sometimes these were illustrated) and that the order of T
and C could be reversed. The calculation was then shown and evaluated. There were a number of
different approaches based on this principle that were clearly communicated and successfully
used, However, many solutions presented only a calculation with little justification, and credit would
be awarded based on the approach it was most similar to. A common error was not realising that
the T and C could be reversed and so not multiplying by 2. A number of solutions attempted to use
a similar method to Question 2(b) but made little progress with identifying all the scenarios which
needed to be removed.
Question 5
This question was successfully identified by most candidates as involving discrete random variables and so
linked to the binomial and geometric approximations. However, there was some confusion as to the
conditions required for the geometric approximation to be applied.
(a) Most solutions recognised that this was a fairly straightforward binomial approximation question.
Good solutions clearly stated all the unsimplified terms required and then evaluated without
including intermediate steps. Where candidates stated intermediate values, accuracy was often lost
with premature approximation. The most common error was to misinterpret the success criteria and
calculate the probability that three or more days will be wet. A less common, but still frequent, error
was to include the probability that three days will be wet.
The interpretation of this type of success criteria is an important skill in this component, and
candidates should be encouraged to focus on achieving a good understanding of the differences
that occur.
(b) Most candidates identified that the geometric approximation was appropriate. There was some
inconsistency with the interpretation of the context with a common error being having eight dry
days and so actually finding the probability that the first wet day is 9 October. Less successful
solutions found the eight term of a binomial approximation, normally out of 31 for the days in
October. A significant number of solutions were not presented to 3 significant figures with the
probability stated to 3 decimal places. At this level, candidates should be able to differentiate
between decimal places and significant figures for accuracy.
(c) Many found this question very challenging. The best solutions realised that the probability from
part (b) was required, recognised that a binomial approximation was appropriate and evaluated
accurately. Weaker solutions seemed to continue with the geometric approximation and found the
probability that year 4 was the first to fulfil the success criteria. Many solutions used the binomial
approximation with the original probabilities, and so found the probability that 8 October was wet in
exactly 1 of the 4 years.
© 2021
Question 6
Most solutions recognised that normal approximation was appropriate throughout the question. Many
answers were presented in a clear, logical order with good communication of the processes and
mathematical techniques used. The context was generally identified as continuous, with few attempts to use
the continuity correction. The most successful solutions often included a sketch of the normal distribution
curve to help identify the required probability area.
(a) The majority of candidates successfully constructed an appropriate standardisation calculation and
found the anticipated z-value accurately. Many were able to use normal tables effectively to find the
probability, but many did not interpret the success criteria accurately and found the complement of
the required answer. The use of a simple sketch of the normal distribution curve was seen in many
successful solutions.
(b) The best solutions initially found the anticipated z-value and formed an equation using the
standardisation formula with the mean and standard deviation values substituted. There was a
clear algebraic process presented to find the value of t. Weaker solutions used the z-value for the
incorrect probability area. It was noted that candidates who did not use the tables often truncated
rather than round their answers, resulting in accuracy errors. A number of solutions simply equated
the standardisation formula to a probability which was not sufficient.
(c) This question was found challenging by many and was omitted by some candidates. The most
efficient solutions correctly interpreted the 15 minutes as the numerator of the standardisation
formula, had a simple sketch which identified that both tails of the normal distribution were
excluded and found the required probability area using 2Φ(1 5625) – 1. Less efficient approaches
involved determining the times that were 15.0 minutes from the mean and substituting into the
normal standardisation formula with the mean. Many of these solutions were incomplete and only
found the probability that the task was completed in less than 47 2 minutes. The weakest solutions
misinterpreted the criteria and substituted 15 as the time into the normal standardisation formula.
Again, simple diagrams were often seen in successful solutions.
Question 7
(a) Most graphs used scales which enabled points to be plotted accurately. The labelling of axes was
also much improved, with units being included on the distance axis and the cumulative frequency
axes also labelled. Almost all candidates correctly interpreted the data as being continuous and did
not use a continuity correction when plotting points. Not starting the graph at (0, 0) was the main
error. Candidates should be aware that a curve is required for a cumulative frequency graph, as
line segments creates a cumulative frequency polygon. Plotting the graph at mid-points was
uncommon.
(b) Where there was an appropriate graph in part (a), most candidates provided an attempt at the
interquartile range. As the question required the use of the graph, there needed to be a clear
indication that values had been obtained from part (a), with the best solutions having lines drawn
from the quartiles to the curve.
(c) Although some excellent solutions were seen for this question, many candidates appear to have
found the task challenging. The most efficient approach used was to create a data table which
included the class mid-values (m) and frequencies (f) which was then used to generate mf and mf2.
These were then used to calculate the mean and variance, and then the standard deviation. Most
solutions simply substituted values into the mean and variance formula, which was both repetitive
and time-consuming. Common errors were the use of the upper boundary or class size rather than
the mid value and using the cumulative frequency instead of the frequency. A small number of
these solutions then summed these cumulative frequency values to find the total number of
children, even though the question stated that there were 140 children in the data.
Candidates should be aware that class details can be presented in a variety of ways, and would be
advised to practice interpreting the different formats accurately.
© 2021
MATHEMATICS
Paper 9709/53
Key messages
• Candidates should be aware of the need to communicate their method clearly. Simply stating values
often does not provide sufficient evidence of the calculation undertaken, especially when there are
errors earlier in the solution. Where formulae are used, as in Questions 3b, 6c and 7c, the unsimplified
formula with the correct substitutions should be shown before the final answer is presented. In
Question 5b where different scenarios were being considered, the scenarios need to be clearly
identified.
• Candidates should state only non-exact answers to 3 significant figures, exact answers should be stated
exactly. It is important that candidates work to at least 4 significant figures throughout to justify a 3
significant figure value. Candidates are reminded that if an answer is exact, as in Questions 1 and 6b,
they should not round the answer to 3 significant figures.
• Candidates are expected to be able to use the normal distribution tables and they are expected to know
and use the critical values quoted in the formula booklet where appropriate, as in Question 4c.
General comments
Better responses within this paper demonstrated that many candidates were careful to read the questions
carefully, made sure that they showed all their working, labelled their diagrams and ensured that their
answers make sense within the context of the question. It is important that the final answer fits the context of
761
the question, for example in Question 3b it was not sensible to give an answer of when the context is
48
the time taken to travel to work. Similarly, those who did not draw a diagram in Question 4c and worked with
a positive z-value often did not query the answer that put 95% of the data above a time greater than the
mean.
Question 1
This question was answered well by the majority of candidates, most appreciating that a combination was
required and that they needed to subtract 3 from both 26 and 20 to find 23C17. Many candidates also identified
that since the answer obtained was exact it should not be rounded.
Question 2
(a) Most candidates appeared to know what was required when drawing a back-to-back stem and leaf
diagram and followed the request for Lakeview to be represented on the left-hand side. Better
responses showed the schools labelled correctly and took care to line up the leaves in order from
the centre with no commas in between. The stem needed to be a column of single digits starting
with the lowest number, 1, at the top. With so little data, there was no need to split the leaves for
any part of the stem.
Most candidates realised the importance of including a key for both schools with the units, metres,
clearly stated. It is good practice to use three different digits in the key to avoid confusion. Many
used 0|1|8 from the top row but those who used the second or third row digits, 2|2|0 or 0|3|0,
introduced some ambiguity.
© 2021
(b) This question was well answered with most candidates correctly identifying the positions of the
quartiles as 3rd and 9th, naming the lower quartile as 19 and the upper quartile as 32 and
subtracting. The most frequent error was to think that the lower quartile was the arithmetic mean of
the 3rd and 4th items and, similarly, that the upper quartile was the mean of the 9th and 10th.
Question 3
(a) A significant minority did not appear to understand that in a histogram the area of the bars
represents the frequency and plotted bars of the required widths with the frequencies for heights.
However, most candidates knew that a histogram required them to calculate frequency densities,
most of those correctly dividing the given frequencies by the correct class widths, 5, 5, 10, 10, 20.
A few worked with incorrect class widths, usually either too small by 0.5 or too large by 1, and a
few others incorrectly divided by the midpoint or the upper bound.
Better responses used a sensible scale and plotted the tops of the bars in the correct. The most
common scales used on the vertical axis were 4 or 5 units to 2 cm. Candidates are encouraged to
clearly write down their calculated frequency densities to ensure they are plotting the correct
values. Candidates are advised to use a sharp pencil to ensure their lines are clearly defined; some
candidates used a thick or blunt pencil which often obscured the height at which they wished to
draw the top of the bar.
Most candidates appreciated that the bar ends should be at 0, 5, 10, 20, 30 and 50 but those who
incorrectly tried to plot them at 4.5, 9.5 etc. frequently ended up with a horizontal axis that was not
to scale from 0 to 4.5. Most also knew to label their vertical axis with ‘frequency density’ and ‘time
in minutes’ on the horizontal axis. Some candidates omitted axes labels, which are essential to
given meaning to their graph.
(b) Most candidates used the formula for an estimated mean appropriately and so long as they were
using the correct class boundaries to find the mid-points were able to produce the correct answer.
The mid-point attempts needed to be clearly substituted into the correct formula, with many
candidates giving the final quotient only which was not sufficient. Presenting the final answer as
761
was not appropriate in the context of the question, since it asked to find an estimate of the
48
mean time taken by an employee to travel to work, with time conventionally presented as a decimal
or a mixed number.
Question 4
(a) This question was answered well with most candidates correctly standardising 43.2 to find a z-
value of 0.5 . The most frequent error was to truncate the value to 0.555 rather than rounding,
giving these candidates the final probability of 1 – 0.7105 rather than the required 1– 0.7108.
Although very few drew a diagram to find the correct tail, most did find an appropriate area less
than 0.5.
(b) Most candidates multiplied a probability by 365 and found the probability by subtracting their
answer to part (a) from 1. However, many used a probability rounded to 3 significant figures,
forgetting that to obtain an answer correct to 3 significant figures they need to use input numbers
correct to 4 or more significant figures. Others either gave a decimal answer, used an
approximation sign, or stated the answer correct to 3 significant figures and did not follow the
instruction to find an estimate for the number of days, which required an integer as the final
answer, which could have been either side of the decimal result.
(c) The majority of candidates used the inverse normal distribution to find thfe required z-value with
only a small number using the tables the wrong way round. Candidates are required to be familiar
with the tables provided and to use the critical values for the normal distribution where appropriate.
Better responses, and especially those who used a diagram, found a negative z-value and usually
obtained the correct final answer. Those who used a positive z-value did not seem to identify that
their final answer, usually 47.1, was greater than the mean which should have flagged that an error
had occurred.
© 2021
Question 5
(a) Those who understood that the order of the letters and digits in the code mattered, knew to use
permutations and usually obtained the correct answer. A significant number of candidates,
however, used combinations and gave the answer 5C2 × 7C4 = 350.
(b) Candidates found this to be one of the most challenging questions on the paper. The most efficient
method was to subtract the number of codes that had neither A nor 5 from the total found in part
(a). This method was seen in only a minority of scripts.
The most frequently attempted method was to attempt identifying the scenarios and to add the
three subtotals for the three identified scenarios, A and no 5; 5 and no A; A and 5. However, many
candidates seemed to confuse the scenarios, and these were generally poorly described; most not
being clear about whether ‘With A’ meant 5 to be included or not. This confusion meant that some
obtained the correct totals for Method 3 in the mark scheme but added them all instead of
subtracting the number with A and 5. Others did not appear to realise that the order mattered and
hence all the sub-totals were incorrect with some candidates not describing the scenarios or
explaining what they were trying to do, providing only values.
(c) Candidates also found this question to be particularly challenging. The most frequently seen
method was to divide the number of successful codes by the total found in part (a). However, many
found it difficult to find the correct number of ways, often not realising that the hundreds digit could
be 5, 6 or 7. Some found the correct number of successful codes but did not divide and make it into
a probability.
Those who chose the probability method were often less successful. While a number of candidates
1
realised that the probability of DE was they had difficulty finding the probability of a number
20
between 4500 and 5 000, often forgetting the three possibilities for the hundreds digit again and
1 1 3
multiplying by instead of .
7 6 6
Question 6
(a) Almost all candidates were able to find the two required equations from summing the probabilities
to 1 and correctly using the formula for E(X). A few candidates misunderstood how to calculate the
E(X), sometimes dividing by 6, while others did not sum the probabilities to 1. Candidates should
ensure they show their method when solving the two equations with the minimum requirement
being to rearrange the equations into the same linear form, preferably with the unknowns on one
side of the = sign and the number on the other side.
(b) This question was answered well with most candidates knowing how to use the variance formula
with only a small number not subtracting the square of the mean.
However, some did not realise that their solution here was exact, the variance in this question
259
being exactly equal to 0.6475 or . Many candidates rounded their answer to 3 significant
400
figures without showing this exact value at all; it is good practice to initially write an answer to more
than 3 significant figures before rounding.
(c) This question was generally answered well. Most candidates recognised that it required use of the
binomial distribution with the value of p from part (a) and n = 12 although a number omitted the nCr
element in the terms. Some candidates interpreted the wording of ‘at least 3’ to mean ‘less than 3’,
‘more than 3’ or ‘exactly 3’. Others expressed the intention to subtract the P(fewer than 3) from 1
but then did not actually do this. Candidates should ensure they show the unsimplified binomial
terms in their working.
(d) Candidates found this question to be particularly challenging. Of the candidates who recognised
the need for the geometric distribution many often did not use the correct probability, choosing to
use the value of p again rather than P(3) from the table in part (a). Others did the correct
calculation but rounded the initial answer to 3 decimal places instead of 3 significant figures.
© 2021
Question 7
(a) Those who understood what was required in this question generally obtained the correct
probabilities. Some ignored the requirement ‘in terms of x’ and filled in numerical answers after
calculating them in part (c). Others did not appear to have read all of the information, mainly
omitting that the ball chosen from box A was ‘placed in box B’ and used the denominator x + 9 in all
four of their probabilities.
(b) For those who had found the correct probabilities in part (a) this was relatively straightforward.
Many of those who had not answered part (a) correctly tried to work backwards to find the
P(blue in box B) that made the probability of both balls being blue correct. However, it was
important that they showed a product that was consistent with their tree diagram.
(c) This question was generally answered well. Almost all candidates equated the answer to part (c) to
1
and worked out that x equalled 14. Some stopped at this point, but most went on to attempt use
6
of the conditional probability formula using x = 14. Some candidates showed little or no working in
obtaining their final answer, it is important that clear method is shown throughout. The product
used in the numerator and the sum of products in the denominator needed to be shown before the
final answer. Candidates who had made errors in their tree diagram but who showed their working
clearly were able to gain the relevant method and follow through marks.
© 2021
MATHEMATICS
Paper 9709/61
Key messages
For questions that require several probability terms it is necessary to show all of the terms in full. In
Question 5(b) it was necessary to write down all of the required binomial terms as well as the probability
sum.
Final answers are generally required to be given to 3 significant figures, unless otherwise specified in a
question. Candidates are reminded to ensure they round their final answer appropriately, ensuring that
significant figure accuracy is used, rather than decimal places. To ensure the accuracy of the final answer it
is important that candidates maintain a 4 significant accuracy throughout their working.
General comments
There were many clear and well explained answers to all questions throughout the question paper.
Question 1
(a) The question required the distribution of the sample mean of the heights. This was a normal
distribution with mean 12.5 and variance 0.4096.
(b) The requested probability was for a region symmetrically placed about the mean. Candidates who
did not notice this could still find the probability by standardising twice, using the parameters found
for the distribution of the sample mean. It was necessary to deal with the Φ values. A diagram
could help candidates with this.
Question 2
A suitable approximating distribution to the given Poisson distribution Po(45.2) was the normal distribution
N(45.2, 45.2). To find the probability that the customer service desk could not deal with all the enquiries in a
day required the use of a continuity correction, 60.5, in the standardisation. Many candidates applied this
correctly. Some candidates incorrectly used 60 or 59.5. Candidates then needed to subtract the
corresponding z-value from 1 since the smaller tail probability was needed for the final answer.
Question 3
In answering this question candidates initially needed to find the z-value, 1.754, corresponding to the given
0.2 × 0.8
confidence interval width by using the variance of proportions, . Secondly candidates needed to find
75
the value of α corresponding to this z-value by using Φ – (1 –Φ) or by using 2Φ –1 or equivalent. In this
question an answer given to 2 significant figures was requested. Some candidates found only the z-value
and the Φ value, 0.9603, and incorrectly gave 96 as their answer for α. Other candidates incorrectly included
p in their calculation with the given width.
© 2021
Question 4
(a) To find the probability it was necessary to integrate f(x) between the limits 0 and 1.2. It was
essential to clearly write down the integration result. Many candidates did all of this correctly. A few
candidates used limits 1.2 to 3 but omitted to subtract their result from 1, which would have
obtained the correct answer.
(b) To find E(X) it was necessary to integrate xf(x) between the limits 0 and 3. It was essential to
clearly write down the integration result. Some candidates omitted the x or incorrectly multiplied by
x2. Other candidates integrated incorrectly.
(c) It was necessary to integrate f(x) between the limits 0 and m and to equate the result to 0.5. It was
essential to clearly write down the integration result and then to show sufficient steps in reaching
the given cubic equation. Many candidates did all of this correctly.
Question 5
(a) (i) The suitable approximating distribution for the given binomial distribution was the Poisson
distribution with parameter 0.025 which could be expressed as Po(0.025). The justification involved
the value of the sample size, n = 2500, and the value of np, 0.025. These two values had to be
related to the critical values 50 and 5. Many candidates did not complete this justification fully.
Some candidates incorrectly gave their answer as the binomial distribution. Other candidates
incorrectly suggested a normal distribution.
(ii) Many candidates answered this question correctly by finding 1 – P(0) using the Poisson
distribution. The answer required 3 significant figures and not just 3 decimal places which was
often seen.
(b) The significance test required the hypotheses to be stated, the relevant probability to be found, the
comparison to be shown and the conclusion to be stated. A one-tailed test was appropriate and the
tail required the probability P(x  4) to be found using the binomial distribution B(28, 0.3). It was
necessary to write down the binomial terms as well as the sum. Many candidates found this sum of
probabilities correctly as 0.0474. Some candidates instead found P(x  3), though some follow
through marks were available. A few candidates found only P(x = 4) which was not a valid test. It
was necessary to show the comparison of the probability with 0.02 for the 2% significance level.
Finally, the conclusion needed to be stated in the context of the question in a non-definite form with
no contradictions. From the wording of the question the context could refer to the researcher’s
suspicion or to the percentage of the people having the medical condition being less than 30%.
Many candidates answered this question very efficiently and accurately.
Question 6
(a) Most candidates found the unbiased estimates correctly, showing the substitution of the given
sample values into the formulae accurately.
(b) It was necessary to create a new variable such as S – T and to compare this with 0.1 seconds. For
this variable the expectation was 0.2 and the variance was 0.5564. After standardising the given
value 0.1 the required probability could be found. A diagram could be helpful in deciding which was
the relevant area for the probability. Other new variables such as S – T – 0.1 could be used. Some
candidates made errors in finding the variance, such as subtracting Var(S) and Var(T) or including
the 0.1 in the variance calculation. Other candidates did not attempt to find a new variable and so
could make little relevant progress. Some candidates incorrectly gave the smaller area as their
probability.
Question 7
(a) To carry out the significance test it was necessary to use the normal distribution of means of
æ 5.22 ö÷
samples Nçç64.6, ÷÷ . The mean mass, 63.5, of the sample of the 100 apples could then be
ççè 100 ø÷
standardised, the comparison with the significance level carried out and the conclusion made. The
hypotheses needed to be for a one-tailed test. For the 2.5% significance level the critical z-value
© 2021
was 1.96 and the comparison using z-values was – 2.115 < –1.96. Using the corresponding
probabilities gave 0.0172 < 0.025. These values indicated a significant result. Many candidates
followed these steps accurately and produced the correct conclusion. The conclusion needed to be
stated in the context of the question in a non-definite form with no contradictions. Some candidates
lost some accuracy in their final answer by rounding prematurely, particularly when changing to the
probability approach. Other candidates used an incorrect critical z-value. Alternatively, some
candidates used the critical value method leading to 63.5 < 63.58. This work could then be used in
part (b) as well.
(b) To find the probability of a Type II error it was necessary to find the acceptance region for H0 using
æ 5.22 ö÷
the normal distribution Nçç64.6, ÷÷ . This value could then be used in the normal distribution
ççè 100 ÷ø
æ 5.22 ö÷
Nçç62.7, ÷÷ . A few candidates completed this work successfully. Many candidates did not find
ççè 100 ø÷
the acceptance region and attempted only the second part of the method, for which some credit
could be gained. Some candidates tried to follow the steps for a significance test in their attempt
here; this was not appropriate. Two diagrams or a combined diagram could be helpful when
answering this question.
© 2021
MATHEMATICS
Paper 9709/62
Key messages
• Candidates are advised to always read the question carefully.

• All answers must be fully supported with the relevant working.
• Candidates should round, rather than truncate, figures during working. Candidates should maintain a 4
significant figure accuracy throughout their working, and only round the final answer to 3 significant
figures, or the degree of accuracy specified in the question.
• When writing conclusions to hypotheses tests the answer must be given in context and there must be a
level of uncertainty in the language used.
• When carrying out a significance test the comparison between the test value and critical value (or
equivalent comparison) must be clearly shown in order to justify the conclusion drawn.
General comments
Many candidates found some questions on this paper particularly challenging. This was specifically true for
Questions 3b, 4a, 7b and much of Question 6. Questions that candidates found to be more straightforward
and generally answered well were Questions 1a, 2a, 5a and 7a.
This paper, therefore, generated a complete spread of marks; there were some good marks awarded, but
also there were many cases where candidates appeared unprepared for the demands of the paper.
In general, timing did not appear to be a problem for candidates.
The comments below indicate common errors and misconceptions, however, there were also some full and
correct solutions presented too.
Question 1
(a) Many candidates successfully found the mean, though attempts at the unbiased estimate of the
variance were varied; the most common error was the confusion between the two alternative
formulae for the unbiased estimate of the variance. Calculation of the biased estimate was
occasionally seen, and also the answer of 0.0056859 was often truncated to 0.00568, rather than
correctly rounded to 0.00569.
(b) Many candidates successfully found the mean but attempts at the variance were not done as well.
Errors included calculating 11 × their variance from part (a) and sometimes adding 0.5, or correctly
using 112 × their variance but then incorrectly adding 0.5.
Question 2
(a) This question was well answered by the majority of candidates. Many candidates made a good
attempt to explain why the given method to choose the sample was unsatisfactory. Many correctly
identified that either the location or time led to the exclusion of many candidates, or some correctly
identified that the 6 might have similar preferences due to the restrictive method. Some candidates
made irrelevant comments about the size of the sample and some merely suggested it was
‘unrepresentative’ without explaining why.
© 2021
(b) This part was not as well answered as part (a). Some candidates did not appear to know what was
required and, of those who did, many forgot to discard any repeated candidate numbers (for
example 204) or to discard candidate numbers greater than 256. There were many partially correct
rather than fully correct attempts.
Question 3
(a) This was a reasonably well attempted question. The confidence interval was correctly found by
many candidates. Errors included use of an incorrect z value and attempting to centre the interval
35
on 35 rather than .
140
(b) It is important that candidates, as well as being able to find a confidence interval, also understand
what a confidence interval is. Few recognised the fact that, for example, a 95 per cent confidence
interval has a 0.95 chance of containing the true value of p. Thus, the probabilities required were
0.9, 0.95 and 0.99 which then needed to be combined correctly so that 2 out of the 3 intervals
contained the true value of p. Many candidates used a probability of 0.25 or attempted to use z
values or did irrelevant calculations finding further confidence intervals.
Question 4
(a) It was important for candidates to read the question carefully here to fully understand the scenario
given. Many candidates gave answers relating to cost or time. Many did not identify that testing
destroys the fireworks, and therefore it is necessary to take a sample rather than using all the
fireworks, as testing the whole population would leave no fireworks left.
(b) This part was reasonably well attempted. Hypotheses were usually given, though not always fully
correctly, and standardising was done well. It is important that the comparison is clearly shown and
a conclusion is drawn that is in context and not definite. Errors here included use of an incorrect z
value and giving a conclusion that did not have the correct level of uncertainty in the language
used; for example, ‘the mean time lasted is not less than 30 seconds’ would not be acceptable as it
is a definite statement.
(c) Few candidates explained the need for the Central Limit Theorem (CLT) fully correctly. Many did
not make it clear that it was the population distribution that was unknown, stating that the
distribution (or ‘it’) was unknown (or not known to be normal) was not enough. Some candidates
thought the CLT was not needed and others gave statements relating to the size of the sample.
Question 5
(a) In general, this part was well answered. Many candidates correctly used a Poisson distribution
here, though not all used the correct value of lambda; the value of 0.2 rather than 2 was often
used. Another error seen was to calculate less than or equal to 3, or just equal to 3, rather than
fewer than 3 as requested. Some candidates used an incorrect binomial distribution.
(b) Not all candidates used the correct approximating distribution N(40,40). Other normal distributions
were used, and some credit could be gained in this instance for standardising. Many did not use, or
used an incorrect, continuity correction.
(c) Many candidates used a Poisson distribution but with a mean of 6 rather than 10 (i.e. only
considering the formatting errors rather than the formatting and typing errors as requested); it is
important for candidates to read the question carefully in order to correctly interpret what is
required. Other errors included calculating just 9 and 10, omitting to note the requirement 8 to 11
‘inclusive’, and other distributions e.g. normal and binomial) were incorrectly used. It was important
that all supporting working was clearly shown.
Question 6
Candidates found this question, in general, to be particularly challenging and generally was not well
answered. It is important that candidates know how to find a rejection region and understand the meaning of
Type I and Type II errors.
© 2021
(a) Many candidates correctly defined their null and alternative hypotheses in answering this part.
(b) To find the rejection region P(0) and P(0,1) using B(30,0.1) needed to be calculated. Candidates
should have noted that P(0) was less than 0.1 and P(0,1) was greater than 0.1 thus concluding that
the rejection region was 0. Some candidates used B(30,0.1) but calculated P(0) and only P(1),
others calculated different point probabilities rather than cumulative probabilities, and some did not
show any comparison with 0.1. A significant number of candidates gave unsupported answers and
a common error was to give the answer (a rejection region) as a probability. Many candidates used
incorrect distributions.
(c) A common incorrect answer of 0.1 was often seen here, though a few candidates, mainly those
who had been successful in part (b), stated the correct answer.
(d) This part was generally not well attempted. The Type II error was found by calculating 1 – P(0)
1
using the distribution B(30, ). A large number of candidates used an incorrect binomial
40
1 1
distribution, often B(40, ) or Bin(40, ), and others incorrectly used a normal or Poisson
40 10
distribution.
(e) It was important that this was answered in the context of the question, many candidates did not do
this. Generic definitions of a Type II error were not acceptable, and those that gave statements
1
such as ‘saying the probability is less than when it is not’ were not fully correct, the contextual
10
1
requirement was to say ‘… when it is actually less than ’.
10
Question 7
(a) (i) This was well attempted with the majority of candidates giving a full and credible solution. It is
important in a ‘show that’ question that steps in the working are all fully shown.
(ii) This part was also well attempted with most candidates attempting to integrate xf(x) using the
correct limits. Errors included omitting ‘k’ and calculation errors were made when substituting the
limits.
(b) Candidates generally found this part to be very challenging. Although there were candidates who
attempted to draw a diagram of a suitable probability density function, few candidates labelled their
diagrams fully. Some candidates attempted to find a value for a (a = 1 being common), others
worked with ɸ(0.2) trying incorrectly to fit a normal distribution to the information given, and some
attempted to integrate various functions, for example y = x or the function given in part (a). Many
did not seem sure of what method to use, not realising that use of a diagram, as directed, and
symmetry led to the solution.
© 2021
MATHEMATICS
Paper 9709/63
Key messages
For questions that require several probability terms it is necessary to show all of the terms in full. In
Question 5(b) it was necessary to write down all of the required binomial terms as well as the probability
sum.
Final answers are generally required to be given to 3 significant figures, unless otherwise specified in a
question. Candidates are reminded to ensure they round their final answer appropriately, ensuring that
significant figure accuracy is used, rather than decimal places. To ensure the accuracy of the final answer it
is important that candidates maintain a 4 significant accuracy throughout their working.
General comments
There were many clear and well explained answers to all questions throughout the question paper.
Question 1
(a) The question required the distribution of the sample mean of the heights. This was a normal
distribution with mean 12.5 and variance 0.4096.
(b) The requested probability was for a region symmetrically placed about the mean. Candidates who
did not notice this could still find the probability by standardising twice, using the parameters found
for the distribution of the sample mean. It was necessary to deal with the Φ values. A diagram
could help candidates with this.
Question 2
A suitable approximating distribution to the given Poisson distribution Po(45.2) was the normal distribution
N(45.2, 45.2). To find the probability that the customer service desk could not deal with all the enquiries in a
day required the use of a continuity correction, 60.5, in the standardisation. Many candidates applied this
correctly. Some candidates incorrectly used 60 or 59.5. Candidates then needed to subtract the
corresponding z-value from 1 since the smaller tail probability was needed for the final answer.
Question 3
In answering this question candidates initially needed to find the z-value, 1.754, corresponding to the given
0.2 × 0.8
confidence interval width by using the variance of proportions, . Secondly candidates needed to find
75
the value of α corresponding to this z-value by using Φ – (1 –Φ) or by using 2Φ –1 or equivalent. In this
question an answer given to 2 significant figures was requested. Some candidates found only the z-value
and the Φ value, 0.9603, and incorrectly gave 96 as their answer for α. Other candidates incorrectly included
p in their calculation with the given width.
© 2021
Question 4
(a) To find the probability it was necessary to integrate f(x) between the limits 0 and 1.2. It was
essential to clearly write down the integration result. Many candidates did all of this correctly. A few
candidates used limits 1.2 to 3 but omitted to subtract their result from 1, which would have
obtained the correct answer.
(b) To find E(X) it was necessary to integrate xf(x) between the limits 0 and 3. It was essential to
clearly write down the integration result. Some candidates omitted the x or incorrectly multiplied by
x2. Other candidates integrated incorrectly.
(c) It was necessary to integrate f(x) between the limits 0 and m and to equate the result to 0.5. It was
essential to clearly write down the integration result and then to show sufficient steps in reaching
the given cubic equation. Many candidates did all of this correctly.
Question 5
(a) (i) The suitable approximating distribution for the given binomial distribution was the Poisson
distribution with parameter 0.025 which could be expressed as Po(0.025). The justification involved
the value of the sample size, n = 2500, and the value of np, 0.025. These two values had to be
related to the critical values 50 and 5. Many candidates did not complete this justification fully.
Some candidates incorrectly gave their answer as the binomial distribution. Other candidates
incorrectly suggested a normal distribution.
(ii) Many candidates answered this question correctly by finding 1 – P(0) using the Poisson
distribution. The answer required 3 significant figures and not just 3 decimal places which was
often seen.
(b) The significance test required the hypotheses to be stated, the relevant probability to be found, the
comparison to be shown and the conclusion to be stated. A one-tailed test was appropriate and the
tail required the probability P(x  4) to be found using the binomial distribution B(28, 0.3). It was
necessary to write down the binomial terms as well as the sum. Many candidates found this sum of
probabilities correctly as 0.0474. Some candidates instead found P(x  3), though some follow
through marks were available. A few candidates found only P(x = 4) which was not a valid test. It
was necessary to show the comparison of the probability with 0.02 for the 2% significance level.
Finally, the conclusion needed to be stated in the context of the question in a non-definite form with
no contradictions. From the wording of the question the context could refer to the researcher’s
suspicion or to the percentage of the people having the medical condition being less than 30%.
Many candidates answered this question very efficiently and accurately.
Question 6
(a) Most candidates found the unbiased estimates correctly, showing the substitution of the given
sample values into the formulae accurately.
(b) It was necessary to create a new variable such as S – T and to compare this with 0.1 seconds. For
this variable the expectation was 0.2 and the variance was 0.5564. After standardising the given
value 0.1 the required probability could be found. A diagram could be helpful in deciding which was
the relevant area for the probability. Other new variables such as S – T – 0.1 could be used. Some
candidates made errors in finding the variance, such as subtracting Var(S) and Var(T) or including
the 0.1 in the variance calculation. Other candidates did not attempt to find a new variable and so
could make little relevant progress. Some candidates incorrectly gave the smaller area as their
probability.
Question 7
(a) To carry out the significance test it was necessary to use the normal distribution of means of
æ 5.22 ö÷
samples Nçç64.6, ÷÷ . The mean mass, 63.5, of the sample of the 100 apples could then be
ççè 100 ø÷
standardised, the comparison with the significance level carried out and the conclusion made. The
hypotheses needed to be for a one-tailed test. For the 2.5% significance level the critical z-value
© 2021
was 1.96 and the comparison using z-values was – 2.115 < –1.96. Using the corresponding
probabilities gave 0.0172 < 0.025. These values indicated a significant result. Many candidates
followed these steps accurately and produced the correct conclusion. The conclusion needed to be
stated in the context of the question in a non-definite form with no contradictions. Some candidates
lost some accuracy in their final answer by rounding prematurely, particularly when changing to the
probability approach. Other candidates used an incorrect critical z-value. Alternatively, some
candidates used the critical value method leading to 63.5 < 63.58. This work could then be used in
part (b) as well.
(b) To find the probability of a Type II error it was necessary to find the acceptance region for H0 using
æ 5.22 ö÷
the normal distribution Nçç64.6, ÷÷ . This value could then be used in the normal distribution
ççè 100 ÷ø
æ 5.22 ö÷
Nçç62.7, ÷÷ . A few candidates completed this work successfully. Many candidates did not find
ççè 100 ø÷
the acceptance region and attempted only the second part of the method, for which some credit
could be gained. Some candidates tried to follow the steps for a significance test in their attempt
here; this was not appropriate. Two diagrams or a combined diagram could be helpful when
answering this question.
© 2021

Mathematics: Paper 9709/11 Paper 1 Pure Mathematics 1

Uploaded by

Copyright:

Available Formats

Mathematics: Paper 9709/11 Paper 1 Pure Mathematics 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mathematics: Paper 9709/11 Paper 1 Pure Mathematics 1

Uploaded by

Copyright:

Available Formats

Cambridge International Advanced Subsidiary and Advanced Level

9709 Mathematics November 2021

Comments on specific questions

Comments on specific questions

Comments on specific questions

Comments on specific questions

Comments on specific questions

Key messages for candidates

Comments on specific questions

always obtain a correct value for 5 x.

Key messages for candidates

Comments on specific questions

(c) but did not have a correct strategy to find P.

tan x as ( tan x ) 2 often then used the chain rule to

Comments on specific questions

Most candidates opted to solve by long division as opposed to inspection using

Comments on Specific Questions

Comments on specific questions

Comments on specific questions

calculated k successfully following the substitution of t = 4 and v = 8. Those who included a

Comments on specific questions

Comments on specific questions

Comments on specific questions

Comments on specific questions

• Candidates are advised to always read the question carefully.

In general, timing did not appear to be a problem for candidates.

Comments on specific questions

Comments on specific questions

You might also like