A parent-friendly introduction to partial-fraction decomposition, polynomial identities, denominator restrictions, verification and transfer.
Contents and worked questions
- The short answer: partial fractions uncover simpler pieces
- First check which values are forbidden
- Why two fractions have the right structure
- Recombine the proposed partial fractions
- Find A and B by matching coefficients
- An alternative coefficient method: special substitution
- How can x = −1 be used when it is excluded?
- Verify the answer by recombining
- Two numerical spot checks
- What the graph would tell us
- The signs can tell a story
- Why A and B need not be positive
- Why the numerator degree matters
- Decompose the proper remainder
- How do we decide which denominator pieces to use?
- A small repeated-factor warning
- Why an irreducible quadratic is different
- What does an algebraic identity actually claim?
- A deceptively easy cancellation example
- Three ways to verify one decomposition
- A comparison table for method selection
- Six errors that look like the same problem but are not
- Two contrasting student solutions
- A five-minute transfer challenge
- How a parent can test comprehension without teaching the chapter
- How a small-group tutorial can build independent method choice
- Why partial fractions belongs to a wider algebra system
- Where the 2027 SEC syllabus places the topic
- Frequently asked questions
- What to read before and after this guide
- Official reference and interpretation limit
What is Additional Mathematics when a single fraction can be rewritten as the sum of two simpler fractions? This is the idea behind partial fractions. A complicated-looking rational expression may contain smaller pieces that are easier to inspect, manipulate and verify. The task is not to change its value. It is to uncover an equivalent structure.
Imagine receiving one algebraic expression with a denominator made from two factors. You are told that it was assembled from two simpler fractions, but the numbers in their numerators have been hidden. Additional Mathematics gives you a systematic way to recover those numbers. Once they are recovered, you can reconstruct the original expression and check that you have not changed its meaning.
This original teaching guide focuses on the two-distinct-linear-factor case and introduces related extensions without replacing the official syllabus or the more specialised guides on repeated factors. It is written for Singapore students and parents who want to understand why the method works, not merely which letters to put on the page.
The short answer: partial fractions uncover simpler pieces
Consider the rational expression:
F(x) = (5x + 7)/[(x + 1)(x + 2)].
The denominator contains two distinct linear factors, x + 1 and x + 2.
We will show that:
F(x) = 2/(x + 1) + 3/(x + 2),
provided x is not −1 or −2.
The expression on the right has two separate fractions, each with only one linear factor in its denominator. Those are partial fractions of the expression on the left.
The equals sign is important. We are not making an approximation or guessing a numerical pattern. We are claiming an exact algebraic identity on the common permitted domain.
A student should be able to explain what the equality means before performing any coefficient calculations.
First check which values are forbidden
The original denominator is:
(x + 1)(x + 2).
It becomes zero at x = −1 and x = −2.
Division by zero is undefined, so these two values are excluded from the original rational expression.
The domain is:
x ∈ ℝ, x ≠ −1 and x ≠ −2.
This restriction is not optional decoration. It remains relevant even if a transformation creates a formula that looks simpler.
A common mistake is to expand and cancel expressions until the original denominator is no longer visible, then forget the values at which the original question was not defined.
Good A-Math working records the excluded values before the algebra changes shape.
Why two fractions have the right structure
Because the denominator is a product of two different linear factors, try the form:
(5x + 7)/[(x + 1)(x + 2)] = A/(x + 1) + B/(x + 2).
The letters A and B stand for constants to be determined. They are not extra variables whose values vary with x.
At this stage, A and B are unknown coefficients. The purpose of the calculation is to find which constants make the equality true for every permitted x.
The order of the denominators matters because the coefficient attached to x + 1 is not necessarily the same as the coefficient attached to x + 2.
There are two pieces because there are two distinct linear factors. That is the structural clue in the denominator.
Recombine the proposed partial fractions
We can verify the proposed form using the familiar skill of adding algebraic fractions.
Start with:
A/(x + 1) + B/(x + 2).
The common denominator is:
(x + 1)(x + 2).
The combined numerator is:
A(x + 2) + B(x + 1).
Therefore:
A/(x + 1) + B/(x + 2) = [A(x + 2) + B(x + 1)]/[(x + 1)(x + 2)].
For this to equal F(x), we need the numerators to match:
5x + 7 = A(x + 2) + B(x + 1).
This is now a polynomial identity. We have reduced a rational-expression problem to matching coefficients in a linear expression.
That reduction is the hidden doorway into the solution.
Find A and B by matching coefficients
Expand the right-hand side:
A(x + 2) + B(x + 1) = Ax + 2A + Bx + B = (A + B)x + (2A + B).
Thus the identity becomes:
5x + 7 = (A + B)x + (2A + B).
For the expressions to be identical for all x, their coefficients of x must agree and their constant terms must agree.
Therefore:
A + B = 5
and
2A + B = 7.
Subtract the first equation from the second:
A = 2.
Substitute into A + B = 5:
B = 3.
So the partial fractions are:
F(x) = 2/(x + 1) + 3/(x + 2),
with the original exclusions x ≠ −1, −2.
This is not a new approximation. It is the same rational function expressed through simpler components.
An alternative coefficient method: special substitution
There is a second useful way to find the coefficients, sometimes called substitution or the cover-up approach when used in a suitable setting.
Return to the cleared-denominator identity:
5x + 7 = A(x + 2) + B(x + 1).
Set x = −1 in this polynomial identity.
The left side becomes −5 + 7 = 2. The right side becomes A(1) + B(0) = A.
Therefore A = 2.
Next set x = −2.
The left side becomes −10 + 7 = −3. The right side becomes A(0) + B(−1) = −B.
Therefore B = 3.
The answer agrees with coefficient comparison.
This method often saves time because choosing a zero of one factor makes one coefficient disappear.
But it also raises a subtle question that students should understand rather than memorise mechanically.
How can x = −1 be used when it is excluded?
We said the original rational expression is undefined at x = −1 and x = −2. Yet we just substituted those values into an equation. Is that a contradiction?
No. We did not substitute them into the original rational expression.
We first multiplied by the denominator on the permitted domain and obtained a polynomial identity:
5x + 7 = A(x + 2) + B(x + 1).
Once the identity is established for all permitted real values, the two polynomial expressions must have identical coefficients, and therefore represent the same polynomial for every real x. We can then evaluate that polynomial identity at x = −1 and x = −2 to solve for the constants.
Those substitutions do not make the original rational fraction defined at excluded values. They are tools for identifying the coefficients of a polynomial identity.
This distinction is essential. Good algebra keeps track of which expression we are evaluating and what its domain is.
Verify the answer by recombining
Take the proposed answer:
2/(x + 1) + 3/(x + 2).
Combine it over the common denominator:
[2(x + 2) + 3(x + 1)]/[(x + 1)(x + 2)].
The numerator becomes:
2x + 4 + 3x + 3 = 5x + 7.
Therefore:
2/(x + 1) + 3/(x + 2) = (5x + 7)/[(x + 1)(x + 2)].
The original expression has been reconstructed exactly.
This recombination is a stronger verification than merely checking one numerical value. It establishes the algebraic equality throughout the permitted domain.
A student should habitually perform a reconstruction check after decomposition. It is quick and often reveals a missing sign or a misplaced coefficient.
Two numerical spot checks
Numerical checks add a different kind of protection against mistakes.
At x = 0, the original expression is:
F(0) = 7/[(1)(2)] = 7/2 = 3.5.
The partial fractions give:
2/(0 + 1) + 3/(0 + 2) = 2 + 1.5 = 3.5.
At x = 1, the original expression is:
F(1) = 12/[(2)(3)] = 2.
The partial fractions give:
2/2 + 3/3 = 1 + 1 = 2.
The agreement is reassuring. However, a few numerical values cannot by themselves prove an identity for all x.
The reconstruction argument supplies the proof; the spot checks supply a fast error-detection layer.
What the graph would tell us
The two excluded values correspond to places where the original denominator is zero. In this example, the factors do not cancel with the numerator.
The graph has vertical asymptotes at x = −1 and x = −2.
In the partial-fraction form, the reason becomes easier to read. Near x = −1, the term 2/(x + 1) can become very large in magnitude, while 3/(x + 2) stays finite. Near x = −2, the roles reverse.
The decomposition makes the influence of each denominator factor more visible.
This is a conceptual extension, not an instruction to assume every factor in every rational expression creates a vertical asymptote. If a common factor cancels, the graph may instead have a removable hole. The original domain still needs attention.
A useful graph check is to choose values just to either side of an excluded number and compare the signs of the dominant term.
The signs can tell a story
Look near x = −1.
For x just greater than −1, x + 1 is a small positive number, so 2/(x + 1) is a large positive quantity.
For x just less than −1, x + 1 is a small negative number, so 2/(x + 1) is a large negative quantity.
The term 3/(x + 2) remains finite near x = −1 because x + 2 is close to one there.
Similarly, near x = −2, the term 3/(x + 2) dominates and changes sign across that excluded value.
This is one reason decomposed expressions are useful: a complicated whole can reveal local behaviour through simpler parts.
We are not conducting formal limit calculations here. The sign reasoning simply shows how algebraic structure can illuminate the graph.
Why A and B need not be positive
Students sometimes interpret A and B as amounts of physical material, as if each fraction were a positive contribution to a total. That interpretation is not part of the algebra.
Consider another expression:
H(x) = (2x + 5)/[(x + 1)(x + 2)].
Try:
H(x) = A/(x + 1) + B/(x + 2).
Clear the denominator:
2x + 5 = A(x + 2) + B(x + 1).
Matching coefficients gives:
A + B = 2,
2A + B = 5.
Subtracting yields A = 3 and then B = −1.
Therefore:
H(x) = 3/(x + 1) − 1/(x + 2).
The negative coefficient is entirely valid. Partial fractions are algebraic components, not probabilities or nonnegative shares.
For x = 0, the original gives 5/2 = 2.5, and the decomposed form gives 3 − 1/2 = 2.5.
Why the numerator degree matters
The direct decomposition method assumes the rational expression is proper: the numerator has lower degree than the denominator.
Our original numerator 5x + 7 has degree one. The denominator (x + 1)(x + 2) has degree two. So the expression is proper.
Now consider:
G(x) = (x² + 6x + 7)/[(x + 1)(x + 2)].
The numerator and denominator both have degree two. We should perform polynomial division before decomposing the remainder.
Expand the denominator:
(x + 1)(x + 2) = x² + 3x + 2.
Subtract that denominator from the numerator:
(x² + 6x + 7) − (x² + 3x + 2) = 3x + 5.
Therefore:
G(x) = 1 + (3x + 5)/[(x + 1)(x + 2)].
Only the proper fractional part needs partial-fraction decomposition.
Decompose the proper remainder
For:
(3x + 5)/[(x + 1)(x + 2)],
write:
3x + 5 = A(x + 2) + B(x + 1).
Matching coefficients:
A + B = 3
and
2A + B = 5.
So A = 2 and B = 1.
Thus:
G(x) = 1 + 2/(x + 1) + 1/(x + 2).
The original restrictions x ≠ −1 and x ≠ −2 remain in place.
At x = 0, the original expression gives 7/2 = 3.5.
The decomposed expression gives 1 + 2 + 1/2 = 3.5.
The polynomial-division step was not an arbitrary extra burden. It separated the whole-number-like polynomial component from the remaining proper rational expression.
How do we decide which denominator pieces to use?
The denominator controls the candidate structure.
For two distinct linear factors:
(x + 1)(x + 2),
the appropriate direct form is:
A/(x + 1) + B/(x + 2).
If the denominator contains a repeated linear factor, such as (x + 2)², the structure must account for the repeated power. A suitable form for a denominator (x + 1)(x + 2)² contains terms of the types:
A/(x + 1) + B/(x + 2) + C/(x + 2)².
If the denominator contains an irreducible quadratic factor, its partial-fraction numerator may need to be linear rather than constant.
These are not stylistic preferences. They follow from the factor structure. The official G3 Additional Mathematics syllabus specifies bounded categories of denominators, so students should use the factor patterns in their own syllabus, not assume every conceivable rational-function decomposition is examinable.
A small repeated-factor warning
Suppose a student is given a rational expression whose denominator includes (x + 2)².
Writing only B/(x + 2) may fail to provide enough freedom to match the original numerator. The repeated factor requires a separate term with the squared denominator.
That does not mean the coefficients are always nonzero. It means the decomposition form must be general enough to represent the intended class.
A careful tutor teaches the template from the denominator’s factorisation, then uses coefficient comparison or substitution to find the actual values.
For a dedicated discussion of repeated factors, see the related eduKateSG article How to Improve Additional Mathematics | Partial Fractions with Repeated Linear Factors. This guide owns the conceptual two-denominator introduction rather than duplicating that more specialised repair lesson.
Why an irreducible quadratic is different
A factor such as x² + 4 cannot be broken into real linear factors because x² + 4 is positive for all real x.
If it appears as a quadratic denominator factor in the permitted syllabus pattern, we may need a term with a linear numerator, such as:
(Bx + C)/(x² + 4).
That is because a constant numerator alone may not provide enough coefficients to reconstruct the general original numerator.
This is a structural observation, not a claim that a student must carry out every possible quadratic-factor decomposition in one exercise.
The official K341 syllabus limits the permitted denominator forms. It is better to master the allowed templates and their reasons than to memorise a long catalogue of unrelated rational expressions.
What does an algebraic identity actually claim?
An equation such as:
A/(x + 1) + B/(x + 2) = 4
could be true for particular values of x, depending on A and B.
But the decomposition identity:
(5x + 7)/[(x + 1)(x + 2)] = 2/(x + 1) + 3/(x + 2)
is claimed to hold for every x in the common permitted domain.
That is a much stronger statement than a numerical equation.
When students match coefficients, they are exploiting this stronger requirement. The x coefficient and the constant term must agree universally, not just at one chosen test value.
That is why the identity can be reconstructed through expansion. Its validity is not based on a single successful substitution.
A deceptively easy cancellation example
Consider:
J(x) = (x + 1)/[(x + 1)(x + 2)].
For x ≠ −1 and x ≠ −2, we can simplify it to:
J(x) = 1/(x + 2).
But the original expression is undefined at x = −1 as well as x = −2.
The simplified formula 1/(x + 2) has a numerical value when x = −1, yet that does not mean the original function suddenly gains that point. The two expressions agree only on the original common domain.
This is a helpful counterexample to the belief that cancelling a factor removes its domain history.
For this J(x), the graph of the original rational expression has a hole at x = −1, not a vertical asymptote there. That contrasts with our initial F(x), where the numerator does not cancel either denominator factor.
The distinction is an optional conceptual extension that makes the value of domain tracking easier to remember.
Three ways to verify one decomposition
A strong student does not have to repeat an entire solution. They can choose complementary checks.
First, recombine the fractions and verify the numerator algebraically. This proves the identity.
Second, substitute one or two convenient permitted values, such as x = 0 and x = 1. These numerical checks can expose sign errors quickly.
Third, compare the domain and basic graph behaviour. Do the same excluded values and dominant denominator factors appear?
Each check has a different purpose. A numerical check cannot replace an identity proof, while an identity proof that silently ignores the original domain can still lead to an incomplete contextual statement.
The goal is a solution that is both algebraically correct and logically precise.
A comparison table for method selection
| Expression or condition | First useful action | Reason |
|---|---|---|
| (5x + 7)/[(x + 1)(x + 2)] | Split into two linear-factor fractions | Proper expression with distinct linear factors |
| (x² + 6x + 7)/[(x + 1)(x + 2)] | Divide polynomials first | Numerator and denominator have equal degree |
| Denominator includes (x + 2)² | Include a squared-denominator term | Repeated factor needs a complete template |
| Expression with x² + 4 as a factor | Consider a linear numerator for that factor | Quadratic factor differs from a linear factor |
| Cancellation of x + 1 | Retain x ≠ −1 | Original domain is not erased by simplification |
The table is a decision guide, not a substitute for working. It helps students recognise the mathematical structure before choosing a method.
Six errors that look like the same problem but are not
Error 1: forgetting an excluded value
A student simplifies correctly yet writes that the identity holds for every real number. The repair is domain checking, not more coefficient practice.
Error 2: writing the wrong denominator template
A student uses only one fraction or omits the squared-denominator component for a repeated factor. The repair is structural recognition.
Error 3: mixing up A and B
A student attaches A to x + 2 but later treats it as the coefficient of x + 1. The repair is disciplined notation and reconstructing the expression.
Error 4: losing a minus sign during coefficient comparison
A student gets A = 3 but writes B = +1 rather than −1 in the second example. The repair is signed-number fluency and verification.
Error 5: decomposing an improper fraction without division
The student attempts to express a degree-two-over-degree-two fraction using only a pair of simple reciprocal terms. The repair is recognising the polynomial component first.
Error 6: treating a special substitution as evaluation of an undefined original fraction
The student substitutes x = −1 into the original F(x). The repair is distinguishing a cleared-denominator polynomial identity from the rational function’s domain.
These errors belong to different reasoning layers. Targeted teaching should diagnose which one actually occurred.
Two contrasting student solutions
Student A starts with the correct template, clears the denominator, matches coefficients and gets A = 2 and B = 3. The student stops immediately.
The algebra is correct, but the solution is stronger if it also states the excluded values and checks the reconstruction.
Student B starts with the same template but writes A = 3 and B = 2. If they recombine, the numerator becomes:
3(x + 2) + 2(x + 1) = 5x + 8.
That is not the required 5x + 7.
One quick reconstruction reveals the swap without needing to redo every step. Student B can then identify precisely which coefficient was attached to which denominator.
This is the value of an inspectable solution: an error can be localised and repaired rather than treated as a complete failure of understanding.
A five-minute transfer challenge
Try these without copying the first worked example.
1. Decompose (2x + 5)/[(x + 1)(x + 2)]. 2. State the excluded real values for the expression in Question 1. 3. Explain why a numerator of degree two over a denominator of degree two may require polynomial division. 4. What is the remainder when x² + 6x + 7 is divided by x² + 3x + 2? 5. Can you substitute x = −1 into the original F(x) to test its numerical value? 6. Explain why a cancelled denominator factor can still affect the function’s domain.
Answers: 3/(x + 1) − 1/(x + 2); x ≠ −1, −2; the polynomial component must be separated; remainder 3x + 5; no; and cancellation gives an equivalent formula only on the original domain.
If the learner can provide the reasoning for these answers, the method is beginning to transfer beyond a single matched-coefficients exercise.
How a parent can test comprehension without teaching the chapter
Ask what the denominator factors tell the student before any calculations begin. A good answer identifies the likely partial-fraction structure and forbidden values.
Ask why A and B are constants and how the student knows the identity is true. A good answer points to coefficient comparison or exact recombination.
Ask why x = −1 may be used in the cleared polynomial identity but cannot be inserted into the original fraction. This exposes whether the learner tracks the logical status of each expression.
Ask whether every coefficient in a partial fraction must be positive. The negative-coefficient counterexample should make the answer clear.
Finally, ask the student to explain which type of practice would help if the only mistake was swapping A and B. A good diagnostic response is different from “do another fifty questions”.
How a small-group tutorial can build independent method choice
The tutor’s first question should not automatically be “What is A?” It should be “What is the factor structure, and what decomposition would be general enough?”
A learner who chooses the correct template can then show the cleared identity and calculate coefficients. Another learner may need to revisit expansion of brackets or simultaneous linear equations before the new method becomes manageable.
After a worked demonstration, change the numerator and ask the student to check the coefficients without being shown the solution. Then change the degree and ask whether polynomial division is required.
The point is to move from understanding to retrieval, from retrieval to fluent execution, and from execution to selecting the correct route when the chapter name is no longer visible.
Why partial fractions belongs to a wider algebra system
Partial fractions is a good example of why Additional Mathematics is not just an accumulation of harder procedures.
The method depends on earlier ideas: factorisation, common denominators, polynomial identities, expansion, simultaneous equations and domain restrictions.
It can also prepare students for later algebraic manipulation in calculus and higher mathematics, where simpler terms may be easier to differentiate, integrate or analyse. Such later uses should not be confused with the exact requirements of a particular SEC examination paper.
The transferable mathematical insight is decomposition. A complex expression may become easier to reason about when its simpler constituent structures are exposed without changing its value.
Where the 2027 SEC syllabus places the topic
SEAB’s 2027 G3 Additional Mathematics K341 syllabus lists polynomials and partial fractions under Algebra A4.
The official content includes multiplication and division of polynomials, the remainder and factor theorems, and partial fractions with denominators no more complicated than the specified distinct-linear, repeated-linear and linear-times-irreducible-quadratic forms.
This article deliberately owns the introductory distinct-linear-factor case. It notes the other permitted templates to help students recognise their place without reproducing a specialist lesson on repeated factors.
The 2026 GCE O-Level Additional Mathematics code is 4049. From the 2027 Singapore–Cambridge Secondary Education Certificate, the G3 code is K341 and G2 Additional Mathematics has its separate K232 syllabus. The levels should not be assumed to assess identical content.
The equations in this guide are original teaching examples, not predictions or quotations from a specimen examination.
Frequently asked questions
Are partial fractions the same as splitting any fraction randomly?
No. The pieces must recombine exactly to the original rational expression over its permitted domain. The factor structure determines a suitable general decomposition.
Why use A and B rather than numbers immediately?
The coefficients are unknown at first. Writing them as constants allows us to form an identity and solve systematically.
Is matching coefficients better than substituting special values?
Both are valid for an appropriate polynomial identity. Matching coefficients works generally and makes the identity structure visible. Special substitution can be faster when factors make terms vanish.
Can a partial fraction have a negative numerator coefficient?
Yes. Algebraic components can be negative. The signs must be determined by the identity, not by an assumption about physical quantities.
Why does the original domain still matter after simplification?
An equivalent formula can look defined at a value where the original fraction had a zero denominator. Cancellation does not retroactively define the original expression at that value.
Does every rational expression need partial fractions?
No. The method is used when a decomposition is useful and structurally appropriate. Some expressions should be simplified or divided first; others are already as simple as necessary.
Do I need to study repeated factors?
The 2027 G3 K341 syllabus includes a bounded repeated-linear-factor case. A separate guide can practise the full template without confusing it with this basic two-denominator example.
Is this an algebra or calculus topic?
It appears explicitly in the Algebra A4 syllabus content. Its techniques also have connections to later mathematical analysis, but those connections do not change its official classification here.
What should my child revise if the method seems impossible?
Start with adding algebraic fractions, factorisation, expansion of brackets, simultaneous linear equations and recognising undefined denominators. Then practise constructing and checking identities.
What to read before and after this guide
The packaging-box article explains how factors reveal roots and why context can reject an algebraic solution. It provides a natural background to the factor structure used here.
For a more advanced decomposition, read Partial Fractions with Repeated Linear Factors.
The broader starting points are What Is Additional Mathematics? and the How Additional Mathematics Works hub.
For the role of careful foundational teaching and close review of a student’s mathematical moves, see the immutable Clementi Secondary 1 Mathematics small-group tutorial reference.
The core question connecting these pages is: can the learner explain why each algebraic transformation is valid, and can they detect when a new-looking expression requires a different route?
Official reference and interpretation limit
The principal source is SEAB’s G3 Additional Mathematics K341 syllabus for 2027 (PDF), especially Algebra A4. The official SEAB G3 listing identifies the current examination code and syllabus access.
All algebraic examples and student tasks here are constructed for explanation. The exclusions and identities are exact mathematics; the parent and tutorial suggestions are teaching approaches, not claims about individual examination outcomes.
Partial fractions is ultimately a lesson in reversible structure: you may take a complicated object apart, but you must always be able to put it back together and show that nothing essential was lost.