Find your next learning step
Choose the question closest to what you need today, or read the guide from the beginning.
If two real numbers are equal, their squares are equal. But equal squares do not always mean the original numbers were equal: 3² = (−3)² even though 3 ≠ −3. Squaring loses information about sign. That is why squaring both sides of an equation can create extra candidate solutions.
For example, √(x + 6) = x leads, after squaring, to x + 6 = x². The squared equation has candidates x = 3 and x = −2. Only three satisfies the original equation. At negative two, the original left side is √4 = 2 while the right side is −2.
This guide begins with that simple sign distinction and develops a careful equation-solving habit: generate candidates, retain domain restrictions and check the original equation. The basic reasoning is shared across Mathematics pathways. Radical equations and repeated squaring are extensions to use only where they match the learner’s school course; the title does not imply every example is assessed at every G1, G2 or G3 level.
Contents
Sections 1–5
Sections 6–10
Sections 11–15
Sections 16–20
Sections 21–25
Squaring an equality is a valid forward implication for real expressions that are defined: A = B implies A² = B². The reverse requires care because A² = B² allows A = B or A = −B. A squared equation may therefore have more solutions than the original equation.
When both original sides are known to be nonnegative, equal squares do imply equal original sides. The sign information makes squaring reversible on that restricted range. If you have not established such a condition, treat solutions of the squared equation as candidates until they are checked.
For √(x + 6) = x, the square root is the principal nonnegative root, so x must be nonnegative. Squaring gives x² − x − 6 = 0, with candidates three and negative two. The negative candidate violates the sign condition and fails substitution. State the original solution as x = 3, not as both roots of the transformed quadratic.
If A and B are equal real numbers, applying the same defined operation to both gives equal results. Adding the same number, multiplying by the same number and squaring both values all preserve the truth of that particular equality in the forward direction.
However, preserving truth forward is not the same as preserving exactly the same set of solutions in both directions. Multiplying both sides by zero gives a dramatic example: every original equality becomes 0 = 0, but that new statement no longer identifies which original values worked.
Squaring is less extreme but has the same logical issue. It maps both a positive number and its negative to the same square. The new equality can no longer distinguish between those two signs unless the original context supplies the missing information.
When solving equations, the goal is not merely to create another true statement from a true one. The goal is to find all and only the inputs satisfying the original statement. That requires attention to whether each transformation is reversible or whether a later check is needed to filter additional candidates.
Compare the false statement −4 = 4 with the true statement (−4)² = 4². Squaring both sides has turned a false equality into a true one because both squares are sixteen. This numerical example shows why the reverse implication cannot hold for arbitrary real numbers.
If you begin from a true statement, such as 4 = 4, squaring remains safe in the forward direction. The issue is not that squaring makes true equations false. It is that a false original equality can also produce equal squares, creating additional inputs when the operation is applied to variable expressions.
For A² = B², subtracting gives A² − B² = 0. Factorising yields (A − B)(A + B) = 0. Over the real numbers, this means A = B or A = −B. The squared equation includes both sign possibilities.
This factorisation is the general reason behind the numerical example. It explains the extra branch rather than treating extraneous solutions as mysterious mistakes that happen only in complicated square-root problems. The mechanism is already visible with four and negative four.
A candidate solution is a value produced by some stage of the solving process that still needs to meet the original equation and its domain conditions. Calling it a candidate is not a criticism of the algebra. It records that the transformation used may have widened the set of possibilities.
For example, if squaring leads to x² = 9, the transformed equation has roots x = 3 and x = −3. Whether both solve the original equation depends on what the original equation was. For x = 3, only positive three works. For |x| = 3, both work.
The same transformed equation can therefore arise from originals with different solution sets. You cannot reconstruct the correct answer by examining the final quadratic alone. You must retain the original problem or the conditions that make the transformations equivalent.
In clear working, use a sentence such as “The squared equation gives candidates…” before listing values. Then show which survive. This wording helps separate solving an intermediate algebraic equation from finishing the original problem, especially when the quadratic arithmetic is the most visually prominent part of the page.
For a nonnegative real number a, √a denotes the principal nonnegative square root. Thus √9 = 3, not “plus or minus three”. The equation y² = 9 has two real solutions, but the notation √9 names one particular nonnegative number.
This distinction supplies a sign condition in radical equations. If √(x + 6) = x, the left side cannot be negative. Therefore x must be nonnegative. The radicand must also be nonnegative, so x + 6 ≥ 0. Together these conditions reduce to x ≥ 0.
The radicand condition and the side-sign condition do different jobs. The first makes the real square root defined. The second makes equality with the right side possible. Satisfying only x ≥ −6 is not enough to solve the original equation.
Do not add a ± sign to a principal square-root symbol as a repair for an unwanted candidate. That changes the original expression. Instead, preserve the meaning of the root and reject candidates that fail it. The notation itself carries information that squaring can erase.
Work over the real numbers. The square root requires x + 6 ≥ 0, and its nonnegative value requires x ≥ 0 if it equals x. State x ≥ 0 as the combined necessary condition before squaring.
Square both sides to obtain x + 6 = x². Rearrange: x² − x − 6 = 0. Factorise: (x − 3)(x + 2) = 0. The squared equation therefore gives candidates x = 3 and x = −2.
Check the original equation. For x = 3, the left side is √9 = 3 and the right side is 3, so the value works. For x = −2, the left side is √4 = 2 and the right side is −2, so the equality fails. Negative two also violates the previously stated x ≥ 0 condition.
The original solution is x = 3 only. The quadratic factorisation was correct. The rejected root is not evidence of bad factorisation; it is evidence that the squared equation permits the opposite-sign branch that the original principal-root equation does not permit.
Keep the two equations visible when checking candidates. A value can satisfy the squared equation perfectly and still fail the original. That is exactly what an extraneous candidate means here.
| Candidate | Original left side √(x + 6) | Original right side x | Original equality | Squared equality x + 6 = x² |
|---|---|---|---|---|
| 3 | 3 | 3 | True | 9 = 9 |
| −2 | 2 | −2 | False | 4 = 4 |
The second row is the important one. Squaring turns the original left and right values, two and negative two, into the same number four. It removes the distinction responsible for the failed original equality.
This table is not an alternative method for solving the quadratic. It is a verification record for the candidates already found. The original equation decides membership in the final solution set.
When writing a solution by hand, you do not need a large table every time. A short substitution line can be enough. But it must substitute into the original equation, not merely into the equation obtained after squaring. Checking the transformed equation again cannot detect the information that transformation has lost.
The domain of √(x + 6) requires x ≥ −6. At x = −2, that expression is defined. Therefore it would be inaccurate to reject negative two by saying “the square root does not exist”. It does exist and equals two; it simply does not equal the negative right side.
This is a useful diagnostic distinction. A domain failure means an original expression is undefined for the proposed input. A sign or equality failure means the expressions are defined but their values do not match. Both reject a candidate, but for different reasons.
For √(x − 1) = x − 3, the domain condition is x ≥ 1 and the right-side sign condition is x ≥ 3. Combining them gives x ≥ 3 as a necessary condition. Solving the squared equation without the stronger condition can produce an unwanted candidate inside the domain.
Write each condition in words if needed: “the radicand is nonnegative” and “the right side must match a nonnegative root”. This is more reliable than a vague instruction that x itself must always be positive whenever a square root appears. The required inequality depends on the actual expressions.
If A and B are both nonnegative real numbers, A² = B² implies A = B. One explanation is that A = −B would require both to be zero when both are nonnegative. Equivalently, the squaring function is one-to-one on the nonnegative real numbers.
Thus an equation such as √(x + 6) = x can be squared equivalently if the condition x ≥ 0 is retained throughout. The transformed statement is not merely x + 6 = x²; it is that equation together with x ≥ 0.
Both sides being nonpositive can also support reversal, because squaring is one-to-one when restricted to that range. The most common radical-equation use here is the nonnegative case because the principal square root already supplies one side’s sign.
You do not need to use the phrase one-to-one to apply the idea. Ask whether the permitted signs prevent two opposite values from being mistaken for one another after squaring. If they do, the missing information has been controlled. If not, keep candidates and verify them rather than assuming equivalence.
Solve √(2x + 3) = x over the real numbers. The radicand requires x ≥ −3/2, while equality with the principal root requires x ≥ 0. The combined necessary condition is x ≥ 0.
Squaring gives 2x + 3 = x², or x² − 2x − 3 = 0. Factorising gives (x − 3)(x + 1) = 0, with candidates three and negative one. Only three meets the sign condition.
Check explicitly: at x = 3, √9 = 3. At x = −1, √1 = 1, which is not −1. Therefore the original solution is x = 3. Negative one is defined in the original expressions but fails equality.
Compare this with the first worked example. The exact coefficients changed, but the structure did not: nonnegative root, right-side sign condition, squared quadratic, candidate filtering. Recognising that structure is more useful than memorising that a particular negative number was rejected. In another equation, a negative x may be perfectly valid if the complete right-side expression is nonnegative.
Solve √(x + 2) = x + 2. Here the right side is x + 2, not x. Both the domain and the necessary nonnegative-right-side condition require x ≥ −2. Negative values of x are not automatically excluded.
Squaring gives x + 2 = (x + 2)². Let y = x + 2, with y ≥ 0. Then y = y², so y(y − 1) = 0. Thus y = 0 or y = 1, giving x = −2 or x = −1.
Both candidates work. At x = −2, the equation reads √0 = 0. At x = −1, it reads √1 = 1. The final solution set contains two negative values, despite the presence of a principal square root.
This example corrects the shortcut “reject all negative x in square-root equations”. That shortcut confuses the sign of x with the sign of the expressions in the equation. The principal root is nonnegative, and the right side must match it. The input variable itself need not be nonnegative unless the actual expressions force that condition.
Consider √(x + 4) = −3. Whenever the left side is defined over the real numbers, it is nonnegative. The right side is negative. Therefore the equation has no real solution, without any need to square it.
If you square anyway, you obtain x + 4 = 9, so x = 5. Substitution into the original gives √9 = 3 on the left and −3 on the right. The only candidate from the squared equation is extraneous.
This illustrates an important strategic point: inspect signs before doing algebra. A sign contradiction can finish the problem immediately. Producing and rejecting a candidate is possible, but it is unnecessary work if the contradiction was already visible.
It also shows that squaring can change an equation with no solutions into one with a solution. The transformed equation is not a reliable substitute for the original unless the lost sign information is retained. A neat numerical candidate is not evidence that the original problem must have an answer; the original equality remains the final test.
Consider x² = 9 as an original equation. Solving it gives x = 3 or x = −3. Neither is extraneous because both satisfy the original statement. The presence of a negative root is not itself a warning sign; the issue is whether a transformation created a value that was not originally allowed.
Now compare |x| = 3. Squaring gives x² = 9, and both roots again satisfy the original absolute-value equation. Its left side was already nonnegative and its right side positive, so squaring does not introduce an unwanted sign branch in this case.
The number of surviving candidates depends on the original equation. Some squared problems produce no extra candidates, some produce one or more, and some produce only extraneous candidates. Do not assume that every squaring step must create an error or that every quadratic must lose one root.
Use the word can: squaring can introduce extra solutions. It does not inevitably do so. That phrasing accurately reflects the mathematics and prevents a learner from mechanically discarding a root just because an example earlier in the lesson rejected one.
Solve √(x − 1) = x − 3. The domain requires x ≥ 1. The right side must be nonnegative, giving x ≥ 3. Keep the stronger combined condition x ≥ 3.
Squaring gives x − 1 = (x − 3)² = x² − 6x + 9. Rearranging produces x² − 7x + 10 = 0, or (x − 5)(x − 2) = 0. The candidates are five and two.
At x = 5, the original equation reads √4 = 2, so it works. At x = 2, it reads √1 = −1, which fails. The rejected candidate is positive, showing that simply deleting negative roots is not a general checking method.
The relevant sign was x − 3, not x. Candidate two lies in the root’s domain but makes the right side negative. This example is particularly useful after the first two negative-candidate problems: it tests whether the learner has understood sign conditions or merely copied the habit of crossing out a minus sign in the final list.
When both sides contain sums, correct squaring still requires correct algebra. The identity (a + b)² = a² + 2ab + b² includes a cross term. It is not generally a² + b².
For example, (x − 3)² = x² − 6x + 9. Omitting −6x changes the squared equation itself, producing a different algebraic problem before candidate checking even begins. Substitution into the original can reject some errors, but it does not guarantee recovery of a genuine solution that faulty algebra never produced.
At x = 5, (x − 3)² equals four. The incorrect expression x² + 9 equals thirty-four. A small numerical test makes the missing cross term obvious, while distribution explains the correct identity: (x − 3)(x − 3) = x² − 3x − 3x + 9.
Keep two checks separate. First, did I square and rearrange correctly? Second, do the resulting candidates satisfy the original equation? The second check protects against nonreversibility, but it is not permission to be imprecise in the first. A reliable solution needs both correct transformations and correct filtering.
Decide whether each implication is valid over the real numbers. If A = B, then A² = B²: valid. If A² = B², then A = B: not valid without further conditions. If A and B are nonnegative and A² = B², then A = B: valid.
For the invalid implication, give the counterexample A = 2 and B = −2. Their squares are equal, but the original values are not. One counterexample is sufficient because the claim asserted validity for all real A and B.
Now decide whether x = −2 satisfies √(x + 6) = x. The left side equals two, so it does not. Does it satisfy x + 6 = x²? Yes, because both sides equal four. These answers are consistent because the second equation is weaker than the first without its sign condition.
This practice focuses on meaning rather than quadratic technique. If the logical directions remain unclear, spend time with numerical A and B examples before attempting more elaborate radical equations. A harder factorisation will not fix uncertainty about what equal squares actually imply.
Solve √(x + 12) = x. The root is nonnegative, so x ≥ 0 is necessary. Squaring gives x + 12 = x², then x² − x − 12 = 0. Factorisation gives (x − 4)(x + 3) = 0.
The candidates are four and negative three. At four, the original equation gives √16 = 4, so it works. At negative three, it gives √9 = −3, or 3 = −3, which is false. The only solution is x = 4.
For a second problem, solve √(x + 5) = −2. The left side cannot be negative, so there is no real solution. Squaring would give x = −1, but substitution produces 2 = −2. This is a candidate generated by sign loss, not an original solution.
Write a final sentence for each problem. A list of quadratic roots is not a finished answer unless every listed root has survived the original conditions. The filtering stage should be visible enough that a reader can tell whether a candidate was rejected for being undefined or for failing equality.
Solve √(x + 3) = x + 1. The domain requires x ≥ −3 and the nonnegative right side requires x ≥ −1. Squaring gives x + 3 = (x + 1)², so x² + x − 2 = 0.
Factorising gives (x + 2)(x − 1) = 0. The candidates are negative two and one. Negative two violates x ≥ −1 and gives √1 = −1 in the original, so it fails. At one, √4 = 2, so it works. The only solution is x = 1.
Compare with the earlier √(x + 2) = x + 2, whose solutions were negative two and negative one. Negative inputs are not universally disallowed. What matters is whether the radicand is defined and whether the complete sides match.
If you are using a rule such as “keep the positive root”, replace it with a more precise procedure: inspect the original expressions at each candidate. A positive candidate can be extraneous, and a negative candidate can be valid. The original equation, not the sign of the answer alone, decides.
Consider √(x + 1) + 1 = x. Isolate the radical first: √(x + 1) = x − 1. The right side must be nonnegative, so x ≥ 1. This also ensures the radicand is defined.
Squaring gives x + 1 = (x − 1)² = x² − 2x + 1. Rearranging gives x² − 3x = 0, so the candidates are zero and three. Zero violates x ≥ 1 and fails the original equation, since √1 + 1 = 2 rather than zero.
At three, the original left side is √4 + 1 = 3, matching the right side. The solution is x = 3. Isolating the radical made both the sign condition and the squaring simpler.
You could square the unisolated equation, but then the square of the left side contains a cross term involving the radical. That usually creates more algebra rather than removing the difficulty immediately. Strategy matters: choose a valid transformation that exposes the conditions and reduces complexity, not merely the first operation that is technically possible.
For real x, √(x²) = |x|. It is not always x. Squaring makes x² nonnegative, and taking the principal square root returns the nonnegative magnitude of the original x.
At x = −5, √(x²) = √25 = 5, while x itself is −5. If a solver cancels the square and root without considering sign, they lose the absolute value that preserves the correct principal-root meaning.
This connection helps explain why undoing a square needs conditions. On x ≥ 0, |x| = x, so the familiar cancellation is valid. On x ≤ 0, |x| = −x. Across all real x, the absolute-value expression covers both branches.
For example, solving √(x²) = 4 means solving |x| = 4, giving x = 4 or x = −4. Neither is extraneous. By contrast, solving √(x²) = −4 has no real solution because the left side is nonnegative. Squaring that last equation would produce x² = 16 and two candidates, both of which fail the original negative-right-side equation.
Some equations involving more than one radical require more than one squaring step. Each step can lose sign information unless suitable restrictions make it reversible. A candidate obtained after several transformations must satisfy all original domain conditions and the original equality.
This does not mean one must substitute into every intermediate line to finish a solution. Checking the original equation directly is decisive for candidate validity, provided the candidate list was obtained by correct forward transformations and all original solutions were retained in generating it.
However, keeping intermediate sign conditions can reduce the candidate list and explain why certain values are impossible. It also makes lengthy working easier to audit. Without those notes, a final polynomial may seem disconnected from the radical equation it was meant to solve.
Repeated-squaring techniques belong only where the learner’s course includes them. The central habit is already complete in the simple examples: squaring may widen possibilities, and original checks restore the exact target. More complicated algebra changes the workload, not the reason verification is necessary.
Squaring defined real sides preserves every original solution in the forward direction, though it may add candidates. Dividing by an expression that can be zero has a different risk: it may remove genuine solutions if the zero case is discarded without consideration.
For instance, x(x − 1) = 0 has solutions zero and one. Dividing by x and solving x − 1 = 0 finds only one, unless the case x = 0 is handled separately. This is not an extraneous-solution problem; it is a lost-solution problem.
The two risks suggest different repairs. After a nonreversible squaring step, check candidates against the original. Before dividing by a variable expression, consider whether it can be zero and separate that case if needed. A final substitution cannot recover a genuine solution that was never included in the candidate list.
The equivalent equations guide develops this wider distinction. An equation-solving step should be examined not only for whether it looks balanced, but for whether it preserves exactly the same solution set under its stated conditions.
Write the original left side and right side separately, substitute the candidate, and compare exact values where practical. For x = −2 in √(x + 6) = x, write left = √4 = 2 and right = −2. The mismatch is immediate.
Do not begin the check by squaring both resulting sides again. That repeats the very operation that concealed the sign mismatch. A check should test the original requirement, not recreate the weaker transformed requirement.
Exact checking is preferable when radicals simplify cleanly. If approximate values are needed, use sufficient precision and do not mistake rounded agreement for exact equality. For symbolic candidates involving surds, algebraic verification may be more informative than decimal approximations.
For each candidate, also check that the original expressions are defined. A denominator of zero or a negative radicand in a real-valued square root rejects the input before any equality comparison. State the reason accurately. “Undefined” and “unequal” are different conclusions, even though both exclude a candidate from the final real solution set.
Suppose a learner correctly squares √(x + 6) = x, correctly factorises the quadratic and reports x = 3 or x = −2. The algebra is not wholly wrong. The missing step is the transfer from candidates of the squared equation back to solutions of the original.
A precise correction is: “These are the roots of the squared equation. Check them in the original because squaring loses sign information. Negative two gives a positive root on the left and a negative value on the right, so it is not an original solution.”
This feedback preserves the valid factorisation and identifies the conceptual gap. Telling the learner simply to check more carefully does not explain why a check is essential here or why the rejected root appeared.
Next use the positive extraneous candidate from √(x − 1) = x − 3. If the learner correctly rejects two because x − 3 is negative, they are using the original-side condition rather than a memorised ban on negative roots. That is a stronger sign of understanding than repeating the first example with different coefficients.
The entry point for every pathway is the numerical fact that opposite real values have equal squares. Use that to discuss equality, signs and checking. Quadratic radical equations, absolute-value notation and repeated squaring should be introduced only when consistent with the student’s course and readiness.
For official course context, consult the G1 syllabus, G2 syllabus or G3 syllabus for the relevant 2027 Mathematics course. These references establish scope; this article does not claim that its optional extensions are common requirements across all three.
The broader eduKate guide on extraneous solutions covers the general problem of candidates that do not belong. This article focuses on one mechanism: sign information lost through squaring.
Return to the Mathematics explainer directory for related lessons on principal roots, factors, brackets and equality. Choose the connection that repairs the current mistake. If expansion is incorrect, practise distribution; if candidates are accepted too soon, practise original-equation verification.
Which original equations accept both x = 3 and x = −3: x² = 9, √(x²) = 3, or √9 = x? The first two accept both. The third accepts only positive three because √9 names the principal root three.
Now solve √(x + 20) = x. The necessary condition is x ≥ 0. Squaring gives x² − x − 20 = 0, with factors (x − 5)(x + 4). Candidates are five and negative four. Substitution keeps five and rejects negative four, giving the single solution x = 5.
Finally, decide whether √(x + 1) = −1 has any real solution. It has none because the left side is nonnegative wherever defined. Squaring produces candidate zero, but the original at zero reads 1 = −1. The candidate is extraneous.
These questions test different parts of the same reasoning: principal-root meaning, sign restrictions, correct factorisation and checking. If one part is insecure, practise that part directly rather than assuming more difficult equations will automatically strengthen it.
One useful self-test is to explain why checking matters without mentioning a particular example. Say: “The squared equation accepts equal or opposite original sides. The original equation accepts equal sides only.” That sentence identifies the information lost in the transformation. It also explains why a candidate may pass every quadratic calculation yet still need to be rejected. A check is not an optional confidence ritual here; it tests a condition the transformed equation no longer fully records.
Before squaring an equation, ask whether both original sides are defined and what signs they can have. Record the relevant restrictions. Then square correctly, solve the resulting equation and treat the resulting values as candidates unless equivalence has been established under the retained conditions.
Check each candidate in the original equation, especially when a sign condition was not carried through the algebra. Keep all candidates that satisfy it and reject all that do not. Do not discard a value merely because it is negative, and do not accept one merely because it solves the quadratic.
The logical distinction is the heart of the lesson: A = B gives A² = B², but A² = B² can also arise from A = −B. Squaring preserves equality forward while potentially forgetting the sign information needed to reverse it.
Once that distinction is clear, extra candidates stop looking like mysterious algebra failures. They are predictable consequences of a transformation that merged two possibilities. Original-equation checking restores the distinction and finishes the task that the transformed equation alone could not finish.
These answers summarise the checking habits developed in the examples. They distinguish the original equation from the intermediate squared equation and avoid using the sign of x as a substitute for testing the actual expressions.
Is squaring both sides an invalid operation?
No. It is valid as a forward implication when the real expressions are defined. The caution is that it may not be reversible without conditions, so the squared equation may have additional candidates that must be checked.
Are all negative roots extraneous?
No. A negative input can satisfy an equation whose complete sides are nonnegative. For example, both negative two and negative one solve √(x + 2) = x + 2. Always test the expressions in the original equation.
Can a positive candidate be extraneous?
Yes. In √(x − 1) = x − 3, candidate x = 2 from the squared equation makes the original sides one and negative one. It fails even though x itself is positive.
Why is checking the quadratic not enough?
The quadratic is the transformed condition that may already include extra possibilities. A value satisfying it has only passed that weaker test. Substitution into the original equation checks the actual equality the problem asked you to solve.
Continue the G1/G2/G3 Mathematics series
All G1/G2/G3 Mathematics explainers
What Is G1 Mathematics | Why Profit Percentage Uses Cost Price as Its Base
What Is G2 Mathematics | Why x² + x Cannot Be Collected as 2x²
What Is G3 Mathematics | Why a Probability of 0.6 Does Not Guarantee Six Successes in Ten Trials