How to perform in the new G2 SEC Mathematics examination at an advanced level includes knowing when a line of algebra still describes exactly the same problem. Many symbolic errors come from treating any familiar operation as automatically reversible. A transformation can preserve all solutions, lose valid solutions, or create extra candidates that must later be checked.
This hundred-and-second Learner’s Guide focuses on solution-set preservation. The central rule is: every algebraic move should preserve the meaning of the original condition unless you deliberately create candidate solutions and verify them later. Equivalent equations have the same solution set. Non-equivalent transformations require additional care.
For 2027, SEAB lists G2 Mathematics as K210. Use the official G2 syllabus directory for current subject documents. The examples below are original eduKateSengkang learning material rather than official specimen questions.
Equivalent transformation
An equivalent transformation changes the appearance of an equation or expression without changing which values satisfy it, within the stated domain.
This is why algebra can simplify a problem while preserving its solution.
Add the same quantity to both sides
If a = b, then a + c = b + c.
Adding the same quantity to both sides preserves equality.
Subtract the same quantity from both sides
If a = b, then a − c = b − c.
Again, equality is preserved.
Multiply both sides by the same non-zero constant
If a = b and c ≠ 0, then ac = bc.
The non-zero condition matters because multiplying by zero would turn every equation into 0 = 0 and destroy information about the original solution set.
Divide both sides by the same non-zero constant
If a = b and c ≠ 0, then a/c = b/c.
Division by zero is undefined, so the condition must be visible.
Expand and factorise
Expansion and factorisation rewrite an expression into equivalent forms.
For example, 3(x + 2) and 3x + 6 have the same value for every allowed x.
Collect like terms
Combining 2x + 3x into 5x preserves the expression because the terms are of the same algebraic type.
Combining 2x + 3 into 5x does not preserve meaning because unlike terms were merged incorrectly.
Equivalent expressions versus equivalent equations
Equivalent expressions have the same value for all values in the common domain.
Equivalent equations have the same solution set. These ideas are related but not identical.
The solution-set question
After each important transformation, ask: Does every solution of the old line satisfy the new line, and does every solution of the new line satisfy the old line?
If both directions are true, the equations are equivalent.
Squaring both sides
If a = b, then a² = b². But the reverse is not always true because a² = b² allows a = b or a = -b.
Squaring can therefore introduce extra candidates when used to reverse an equation.
Example: squaring creates a candidate
Equation x = 3 has the single solution x = 3. Squaring both sides gives x² = 9, whose solutions are x = 3 and x = -3.
The new equation has a larger solution set. It is not equivalent to the original equation.
Square-rooting
Taking square roots requires attention to domain and sign.
From x² = 9, the real solutions are x = ±3, not only +3. From a length context, the negative candidate may later be rejected because of the domain.
Multiplying by an expression containing a variable
Multiplying both sides by an expression involving x can be safe only if you know when that expression is zero and how the domain is affected.
If the factor can be zero, the transformation may introduce or hide cases.
Dividing by an expression containing a variable
Dividing by x assumes x ≠ 0.
If x = 0 could be a valid solution of the original equation, dividing by x would remove it without checking.
Example: losing a solution by division
Suppose x(x − 2) = 0. The solutions are x = 0 and x = 2.
Dividing both sides by x gives x − 2 = 0 and leaves only x = 2. The valid solution x = 0 has been lost because division by x excluded that case.
Factor first, then use zero-product reasoning
For a product equal to zero, identify each factor that can be zero.
This preserves all solution branches without illegal division.
Clearing fractions
Multiplying an equation by a common denominator can simplify it, but values that make the original denominator zero remain excluded from the domain.
A simplified later line cannot restore values that were never allowed in the original expression.
Original-domain memory
Keep restrictions from the original equation visible even after denominators cancel.
The final solution set belongs to the original problem.
Cross multiplication
Cross multiplication is a shortcut for multiplying both sides by denominators.
It is valid only when the original denominators are defined and non-zero. Domain restrictions should not disappear because the shortcut looks familiar.
Taking reciprocals
If a = b and both are non-zero, then 1/a = 1/b.
If either side can be zero, taking reciprocals may be undefined.
Absolute value
Equations involving absolute value can produce multiple branches.
The learner should preserve both valid cases where the syllabus and context require them rather than treating |x| as simply x.
The implication direction
Some transformations guarantee that every original solution remains a solution of the new equation without guaranteeing the reverse.
That means the new equation may contain extra candidates. Verification in the original equation becomes essential.
Candidate solution
A candidate is a value produced by the transformed equation that still needs to be checked against the original condition.
Candidate does not mean wrong; it means not yet fully verified.
Extraneous solution
An extraneous solution satisfies a transformed equation but not the original equation.
Substitution into the original equation exposes it.
The substitution-back rule
Whenever a transformation could have expanded the solution set, test final candidates in the original equation.
Do not substitute only into the simplified final line; that line is where the candidate came from.
The domain-first rule
Before manipulating a rational or root expression, identify excluded values or required sign conditions where relevant.
This prevents later confusion when a candidate appears outside the original domain.
The no-cancel-across-addition rule
Cancellation applies to common factors, not terms separated by addition.
For example, (x + 2)/x cannot be simplified by “cancelling x” from the numerator because x is not a factor of the entire numerator.
Factor before cancelling
If an expression can be factorised, identify the common factor explicitly.
Only then can factor cancellation be considered, with domain restrictions preserved.
Equality versus identity
An equation may be true only for certain values. An identity is true for every value in its domain.
Understanding the difference helps the learner know whether they are solving for x or rewriting a relationship that already holds generally.
Equivalent form checking
Pick a simple allowed value and evaluate both expressions. If the values differ, the forms are definitely not equivalent.
One matching test value does not prove full equivalence, but one mismatch refutes it.
The transformation audit
- What operation was performed?
- Was it applied to both sides where required?
- Is any divisor guaranteed non-zero?
- Did the domain change?
- Could the move introduce extra candidates?
- Could it remove a valid solution?
This audit is especially useful after a suspicious line in a long solution.
The one-change-per-line rule
Perform one clear algebraic change per line during difficult work.
This makes equivalence easier to inspect and errors easier to contain.
The annotation rule
During practice, annotate risky steps: “÷(x−2), so x≠2”, “square both sides—check candidates”.
The annotations can disappear once the reasoning becomes automatic.
Solution-set preservation in word problems
An algebraic candidate must also satisfy the context.
A negative length or non-integer number of buses may satisfy the equation but fail the problem domain.
Practice clinic one: legal move
Equation: 4x + 7 = 31. Subtract 7 from both sides, then divide both sides by 4. Each step is reversible because the same valid operation is applied to both sides and the divisor is a non-zero constant.
The solution x = 6 can be checked in the original equation: 24 + 7 = 31.
Practice clinic two: illegal division
Equation: x(x − 5) = 0. Dividing both sides by x produces x − 5 = 0 and loses x = 0.
The correct route preserves both zero-product branches: x = 0 or x = 5.
Practice clinic three: squaring
Original equation: x = 4. Squared equation: x² = 16.
The squared equation has x = ±4, so squaring expanded the solution set. If squaring is used during a solution, candidates must return to the original equation.
Practice clinic four: denominator restriction
Expression (x² − 4)/(x − 2) simplifies to x + 2 for x ≠ 2.
The simplified expression has an apparent value at x = 2, but the original expression is undefined there. The cancelled form must keep the original restriction.
Practice clinic five: cross multiplication
Equation 1/(x−1) = 2/(x+3) requires x ≠ 1 and x ≠ -3 before any cross multiplication.
After cross multiplication, solve the resulting equation, then reject any candidate violating the original restrictions.
Practice clinic six: reciprocal
Equation x = 2 can be inverted to 1/x = 1/2 because x is non-zero for the solution.
But taking reciprocals of an equation without knowing either side is non-zero can introduce undefined expressions.
Practice clinic seven: absolute value
Equation |x| = 5 gives x = 5 or x = -5.
The absolute-value expression represents distance from zero, so both signs satisfy the condition.
Practice clinic eight: contextual domain
Equation x² = 25 may produce x = ±5 algebraically.
If x is a physical length, the context keeps x = 5 and rejects x = -5. Algebra first identifies candidates; context selects feasible solutions.
Practice clinic nine: simplifying a rational expression
Expression x(x+3)/x can simplify to x+3 only for x ≠ 0 because the original denominator excludes zero.
The simplified appearance does not erase the original domain.
Practice clinic ten: multiplying by zero
Equation x = 7 multiplied by zero becomes 0 = 0.
The new equation is true for every x, so all information about the original solution has been destroyed. This shows why non-zero conditions matter in reversible transformations.
The two-direction equivalence test
To decide whether two equations are equivalent, ask both directions:
- Does every solution of the first satisfy the second?
- Does every solution of the second satisfy the first?
One-way implication is not enough for equivalence.
The one-way implication example
x = 3 implies x² = 9.
But x² = 9 does not imply x = 3 because x = -3 also works. Therefore the implication is one-way.
The candidate-expansion warning
Operations such as squaring can make the set of candidates larger.
Whenever this happens, the final answer must be filtered through the original equation.
The candidate-loss warning
Operations such as dividing by a variable expression can make the set smaller by excluding values that may have been valid.
Before dividing, handle the zero case separately.
The domain-shift warning
Simplification can produce an expression whose visible domain is larger than the original.
Keep the original exclusions as part of the final statement.
The zero-case habit
Whenever dividing by an expression that could equal zero, pause and ask what happens in the zero case.
This simple habit prevents many lost solutions.
The sign-branch habit
Whenever reversing an even power or an absolute value, ask whether more than one sign is possible.
Do not automatically choose the positive branch unless the domain or context justifies it.
The original-equation check
Substitute final candidates into the original equation, not only the last transformed form.
This catches extraneous solutions and forgotten domain restrictions.
The expression-equivalence check
For expressions, compare values at simple allowed inputs.
One mismatch proves the forms are not equivalent. Several matches support plausibility but do not replace algebraic justification.
The solution-set table
During practice, create a two-column table: transformation / effect on solution set.
- add same expression to both sides — preserved;
- multiply by non-zero constant — preserved;
- square both sides — may add candidates;
- divide by variable expression — may lose zero case;
- cancel factor — preserves values only where original denominator is non-zero;
This helps the learner classify risky moves before using them automatically.
The reversible-move drill
Give ten algebraic transitions and ask whether each move is reversible under the stated domain.
The learner must explain the condition, not just label safe/unsafe.
The find-the-lost-solution drill
Provide a solution where a valid root disappeared after division.
The learner identifies the exact line and restores the missing case.
The find-the-extra-solution drill
Provide a solution where squaring introduced an extra candidate.
The learner checks candidates in the original equation and removes the invalid one.
The domain-memory drill
Give rational expressions that simplify after factor cancellation.
The learner writes the simplified form and original excluded values together.
The branch drill
Use absolute-value or factor equations that naturally create multiple cases.
The learner writes each branch separately and verifies the final set.
The equivalent-expression drill
Give pairs such as 2(x+3) and 2x+6, or x²−9 and (x−3)(x+3).
Ask whether they are equivalent expressions and why.
The non-equivalent-pair drill
Include tempting false simplifications, such as (x+2)/x = 2.
Test a simple allowed value to refute the false equivalence quickly.
Solution-set preservation and invariant hunting
Use Vol 0074. The invariant here is the set of values that satisfy the condition.
Solution-set preservation and assumption audit
Use Vol 0064. Dividing by a variable expression silently assumes it is non-zero unless the zero case is handled.
Solution-set preservation and case splitting
Use Vol 0098. Risky algebra often becomes safe when separate cases are made explicit.
Solution-set preservation and backsolving
Use Vol 0047. Reverse operations are reliable only when the inverse step preserves the relevant domain and branches.
The advanced audit
- Original domain recorded;
- each transformation justified;
- zero divisors handled;
- sign branches preserved;
- candidate-expanding steps checked;
- candidate-losing steps avoided or split into cases;
- final candidates tested in original problem.
This audit is especially useful in long algebraic work.
The four-week solution-set build
Week 1 — safe equivalence
Addition, subtraction, non-zero constant multiplication/division, expansion and factorisation.
Week 2 — risky reversals
Squaring, square roots, absolute values and variable division.
Week 3 — domains
Rational expressions, exclusions and original-domain memory.
Week 4 — mixed timed algebra
Use problems where the learner must decide whether a transformation is equivalent or only produces candidates.
Use the Mathematics index
For underlying algebra teaching, continue through the Complete Mathematics Index.
The PSLE bridge
The PSLE habit Represent Before You Calculate remains useful.
At G2, algebra adds another demand: change the representation without changing the solutions unless the change is deliberate and checked.
Use Examination Craft
For checking and error containment, continue through the Examination Craft hub.
Final rule
Algebra is not permission to change symbols freely.
Every transformation must preserve the original solution set or be followed by a verification step that restores it. Watch zero cases, domain restrictions, sign branches and candidate-expanding operations. The safest algebra is algebra whose meaning survives every line.
Transformation clinic: preserve both directions
Equation 2x + 3 = 11 becomes 2x = 8 after subtracting 3 from both sides. Every solution of the first satisfies the second, and every solution of the second satisfies the first.
This is a true equivalent transformation.
Transformation clinic: multiplying by a variable
Suppose x = 2. Multiplying both sides by x gives x² = 2x. The new equation has solutions x = 0 and x = 2.
The new line contains an extra solution. Multiplying by a variable expression that can be zero may enlarge the solution set.
Transformation clinic: dividing by a variable
Equation x² = 2x can be written x(x−2)=0. Dividing by x yields x=2 and loses x=0.
The zero case must be handled separately before division.
Transformation clinic: cancelling a factor
Expression (x−3)(x+4)/(x−3) simplifies to x+4 only for x≠3.
The simplified expression is not a licence to substitute x=3 into the original.
Transformation clinic: clearing a denominator
Equation (x+1)/(x−2)=3 has original restriction x≠2.
Multiplying both sides by x−2 gives x+1=3(x−2). Solve the equation, but keep x≠2 as part of the final logic.
Transformation clinic: squaring after a square root
If √x = 4, squaring gives x=16 and remains valid within the square-root domain.
But if a more complicated equation is squared, extra candidates can appear. The original equation remains the final judge.
Transformation clinic: expanding
Equation 2(x+5)=18 can become 2x+10=18.
Expansion preserves the equation because the expressions are identical for every x.
Transformation clinic: factorisation
Equation x²−5x=0 becomes x(x−5)=0.
Factorisation preserves the expression and exposes solution branches without losing either one.
Transformation clinic: moving terms
Informally saying “move 5 to the other side and change sign” can hide the real operation.
The valid algebra is subtract 5 from both sides. Understanding the operation makes equivalence easier to audit.
Transformation clinic: multiplying an inequality
For inequalities, multiplying or dividing by a negative quantity reverses the inequality direction.
The transformation remains logically equivalent only if that reversal is made.
Transformation clinic: multiplying by an unknown-sign expression
Multiplying an inequality by an expression whose sign is unknown is risky because the required direction may depend on the sign.
Case splitting may be needed. Do not use equation habits mechanically on inequalities.
The solution-set snapshot
After a risky move, pause and write what candidates are currently possible.
This prevents branches from disappearing silently as algebra continues.
The original-domain snapshot
Write excluded values before simplification and carry them to the end.
A short note such as x≠2 can protect the entire solution.
The reversible-operation map
- +c / −c — mutually reversible;
- ×c / ÷c for c≠0 — mutually reversible;
- expand / factorise — equivalent expression forms;
- square / square root — not automatically one-to-one;
- multiply/divide by variable expression — depends on zero/sign cases.
The map helps the learner decide which steps are routine and which require a check.
The transformation-risk ranking
Low-risk moves preserve equivalence transparently. Higher-risk moves alter domain, branches or sign conditions.
Spend checking attention where risk is structurally higher rather than rechecking every line equally.
The proof-by-substitution limit
Substituting one test value can refute a false expression identity if the outputs differ.
One matching test value cannot prove two expressions are equivalent for all values.
The candidate-check discipline
When extra candidates may have been introduced, substitute every final candidate into the original equation.
Do not check only the candidate you expect to survive.
The lost-solution discipline
When division by a variable expression occurs, recover the excluded zero case and test it separately.
A missing case can be a real solution.
The branch-label discipline
Write each branch on a separate line when signs or factors create multiple cases.
This reduces accidental merging and makes verification easier.
The final-answer set
If more than one solution remains, present the full valid set clearly.
Do not write only one because it appeared first or because it is positive unless the domain requires positivity.
The context-filter stage
After algebra finds valid mathematical solutions, apply context.
Counts may need integers, lengths may need positive values, times may need non-negative values, and other problems may allow negative values legitimately.
The context-is-not-algebra rule
A context can remove a mathematically valid candidate without making the algebra wrong.
Keep the two stages conceptually separate: solve, then interpret.
The transformation error ledger
- operation applied to only one side;
- division by possible zero;
- inequality sign not reversed;
- domain restriction lost;
- extra candidate not checked;
- valid zero case lost;
- factor cancelled across addition;
- positive root chosen without justification;
These categories direct targeted repair.
The one-risky-step drill
Use a solution containing one risky transformation among several safe ones.
The learner identifies which line deserves additional checking and why.
The domain-retention drill
Simplify rational expressions and require the original excluded values to be written beside the final form.
This trains domain memory as part of algebra rather than an afterthought.
The branch-preservation drill
Use equations producing two branches and require both to remain visible until checked.
The learner should not prune a branch based on appearance alone.
The equivalence-pair drill
Present pairs of equations and ask whether they have exactly the same solution set.
If not, identify whether the second has gained or lost candidates.
The solution-set explanation drill
Ask the learner to explain in words why a transformation is safe.
“Because I moved it across” is not enough. “I subtracted the same quantity from both sides” shows the preserving operation.
The advanced standard
An advanced learner sees algebraic manipulation as controlled meaning-preservation.
They know when a line is equivalent, when a line only implies another, when a zero case needs separate handling, and when final candidates must return to the original equation.
Final perspective
The safest algebra keeps track of what values are allowed and which values satisfy the original condition.
Symbols may change appearance many times. The solution set should change only when the mathematics genuinely requires branching, restriction or later verification. Preserve the meaning, not just the pattern of symbols.
Final solution-set clinic: track what each move does to the set
Take a worked algebra solution and write one of three labels beside each transformation: preserved, expanded, or restricted. “Preserved” means the new equation has exactly the same solutions. “Expanded” means extra candidates may have entered. “Restricted” means some cases may have been removed. The exercise forces the learner to think about meaning rather than symbol motion.
For example, subtracting the same number from both sides is preserved. Squaring both sides can expand the candidate set. Dividing by an expression containing x can restrict the set if that expression might be zero. The label tells the learner what kind of checking is required next.
Original equation as the final authority
Whenever the solution path uses a non-equivalent or condition-dependent step, the original equation remains the final authority. A candidate produced later must be checked there. This is especially important when square roots, absolute values, denominator restrictions or variable factors are involved.
The final simplified line can be easier to solve, but it cannot redefine the original domain. A value excluded at the beginning does not become valid simply because a denominator has disappeared after cancellation.
Solution sets and graphical thinking
An equation can also be understood as asking where two expressions have equal values. Graphically, solutions correspond to intersections. If a transformation changes the number of intersections, it has changed the solution set. This perspective can make extra or lost solutions easier to visualise.
For instance, squaring can make two different signed values produce the same square, which helps explain why an additional branch can appear. Division by a factor can remove the point where that factor equals zero, which explains how a valid branch can disappear.
The no-automatic-cancellation rule
Cancellation is safe only when a common non-zero factor has been identified. Terms joined by addition are not separate factors. Before cancelling, factorise the entire relevant expression and record any values that make the original denominator zero.
This one habit protects both algebraic validity and domain memory.
The final solution-set checklist
- What is the original domain?
- Is this transformation genuinely reversible?
- Could it add candidates?
- Could it remove a zero or sign case?
- Have all branches been kept?
- Have final candidates been checked in the original condition?
If the answers are explicit, long algebra becomes far easier to audit.
The advanced standard
An advanced learner does not merely obtain x. They know why each line still describes the same solution problem, where equivalence may have been weakened to one-way implication, and why verification is needed after risky transformations.
Solution-set preservation is the discipline that keeps algebra honest from the first line to the final answer.
One last solution-set practice block
Take a completed algebra solution and reconstruct the solution set after every line. At a routine equivalent step, the set should remain unchanged. At a branch-producing step, note the additional cases. At a domain-restricting step, note what values are excluded. This makes the logical effect of each transformation visible rather than implicit.
Then reverse the audit. Start from the final candidates and trace them back through the original problem. A candidate that survives only the transformed equation but fails the original condition is extraneous. A value that disappeared only because a zero factor was divided away is a lost solution.
Use this practice especially on long solutions where several safe steps make one risky step easy to overlook. The learner should become able to point to the exact line where equivalence stopped being guaranteed.
Solution-set control also improves checking efficiency. Instead of redoing every line, inspect steps with known structural risk: variable division, squaring, cancellation, sign branching and domain restrictions. The rest of the algebra can be checked more lightly.
The final habit is simple: every candidate belongs to the original equation, not to the convenience of the last line. Keep the original domain and original condition visible until the answer set is fully verified.
Final calibration: solution sets under time
Under examination pressure, do not run the full audit after every routine line. Reserve deliberate checking for transformations with structural risk. Safe addition, subtraction, expansion and non-zero constant operations can move quickly. Slow down around variable division, denominator cancellation, squaring, absolute-value branches and domain restrictions.
This risk-based approach protects time while preserving mathematical validity. The learner is not checking less carefully; they are checking where the solution set is most likely to change.
A useful final cue is “same solutions?” Ask it immediately after a risky transformation. If the answer is uncertain, mark the step and verify candidates in the original equation. If a value was excluded because a factor could be zero, test that case separately before finalising the set.
Good algebra under time is therefore selective, not casual: routine equivalence flows, risky equivalence gets inspected, and the original condition remains available as the final check.
One final practice habit is to mark risky algebra with a small mental warning: “check domain”, “check zero case”, or “check candidate”. These cues keep attention on transformations that can alter the solution set without slowing routine algebra.
At the end, read the final answer set against the original equation and context. Every included value must be allowed and must satisfy the original condition; every excluded value should be excluded for a defensible mathematical or contextual reason.
The final algebra habit is to treat solution-set preservation as a live question, not an afterthought. Every risky transformation should trigger one short check: what values were allowed before, what values are allowed now, and did this move add or remove any candidates?
Before boxing the final solution set, read it once against the original domain and once against the original equation. This two-part check catches both invalid candidates and valid cases lost during algebra. The answer is complete only when every surviving value is allowed and every allowed solution has been preserved.
That final verification protects every valid branch before the answer is submitted.
Preserve every solution.
