The Central Idea
Suppose a solving process produces
\[ x=3. \]That value is a candidate solution until we confirm that it actually satisfies the original equation.
For example, if the original equation is
\[ 2x+1=7, \]substitute \(x=3\):
\[ 2(3)+1=7. \]Since
\[ 7=7, \]the candidate is verified.
A solution is not merely a value reached by symbolic manipulation. It is a value that makes the original mathematical statement true.
Candidate Versus Verified Solution
A candidate solution is a value produced during the solving process that might satisfy the original problem.
A verified solution is a candidate that has been checked and shown to satisfy:
- the original equation or inequality;
- the original domain;
- and any contextual restrictions.
Only verified candidates belong in the final solution set.
The Original Equation Is the Final Authority
During solving, the equation may be rewritten many times.
But the final check should usually return to the original equation, not merely the last transformed version.
This matters because some transformations can create or hide invalid cases.
Direct Substitution
The most basic verification method is direct substitution.
Suppose
\[ x=4 \]is a candidate for
\[ 3x-5=7. \]Substitute:
\[ 3(4)-5=7. \]Simplify:
\[ 12-5=7, \] \[ 7=7. \]So \(x=4\) is valid.
Rejecting an Invalid Candidate
Suppose
\[ x=2 \]is proposed as a solution to
\[ x^2=9. \]Substitute:
\[ 2^2=9, \] \[ 4=9. \]This is false.
Therefore, \(x=2\) is not a solution.
Verification Is More Than Arithmetic
A candidate may fail even when the arithmetic looks reasonable.
Verification may require checking:
- whether a denominator becomes zero;
- whether a logarithm receives a positive argument;
- whether an even root has a valid radicand;
- whether a principal root has the required sign;
- whether a physical quantity is allowed to be negative;
- or whether a count must be an integer.
Domain Check Before Substitution
Suppose the original equation contains
\[ \frac{1}{x-3}. \]Then
\[ x\ne3. \]If algebra later produces the candidate
\[ x=3, \]we can reject it immediately because it is outside the original domain.
No further arithmetic is needed.
Rational Equation Example
Solve
\[ \frac{x+1}{x-1}=\frac{2}{x-1}. \]The original restriction is
\[ x\ne1. \]Clearing the denominator gives
\[ x+1=2, \]so
\[ x=1. \]But this candidate makes the original denominator zero.
Therefore,
\[ \boxed{\varnothing}. \]Why Checking the Transformed Equation Is Not Enough
In the previous example, \(x=1\) satisfies the transformed equation
\[ x+1=2. \]But it does not belong to the domain of the original rational equation.
So checking only the transformed equation would incorrectly accept it.
Radical Equations and Extraneous Solutions
Radical equations often require verification because squaring can introduce extra candidates.
Consider
\[ \sqrt{x+1}=x-1. \]Squaring gives
\[ x+1=(x-1)^2. \]This simplifies to
\[ x(x-3)=0. \]So the candidates are
\[ x=0 \]and
\[ x=3. \]Verify the Radical Candidates
Check \(x=0\):
\[ \sqrt{0+1}=0-1, \] \[ 1=-1. \]False.
So \(x=0\) is extraneous.
Check \(x=3\):
\[ \sqrt{3+1}=3-1, \] \[ 2=2. \]True.
Therefore,
\[ \boxed{x=3}. \]Extraneous Solutions
An extraneous solution is a candidate produced by the solving process that does not satisfy the original equation.
Common causes include:
- squaring both sides;
- raising both sides to another even power;
- clearing variable denominators;
- combining logarithms and then solving outside the original domain;
- or other one-way transformations.
Squaring Is a Common Source of Extraneous Candidates
If
\[ A=B, \]then
\[ A^2=B^2. \]But the reverse is not always true because
\[ A^2=B^2 \]allows both
\[ A=B \]and
\[ A=-B. \]Squaring can therefore enlarge the candidate set.
Logarithmic Equations Need Domain Verification
Consider
\[ \log(x-1)+\log(x-4)=1. \]The original domain requires
\[ x>4. \]After combining and solving, algebra may produce several candidates.
Any candidate with
\[ x\le4 \]must be rejected even if it satisfies the transformed polynomial equation.
Example: Logarithmic Candidate Rejection
Suppose solving gives
\[ x=2 \]and
\[ x=5. \]The domain condition
\[ x>4 \]immediately rejects \(x=2\).
Then \(x=5\) should be checked in the original logarithmic equation.
Absolute-Value Equations
Absolute-value equations solved by the correct case split are usually logically equivalent to the original equation.
For example,
\[ |x-2|=5 \]is equivalent to
\[ x-2=5 \]or
\[ x-2=-5. \]This produces
\[ x=7 \]or
\[ x=-3. \]Verification is still useful for catching arithmetic or sign mistakes.
Verification Can Be Logical or Practical
Sometimes verification is logically essential.
This happens when the solving process uses transformations that may introduce candidates.
Other times verification is prudent but not logically necessary.
If every step is reversible and solution-preserving, substitution serves mainly as an error check.
When Verification Is Logically Essential
Verification is especially important after:
- squaring;
- raising to even powers;
- clearing variable denominators;
- canceling factors with domain restrictions;
- combining logarithmic expressions;
- using inverse operations with restricted ranges;
- or making approximations.
When Verification Is Mainly an Error Check
Consider
\[ 2x+3=9. \]The chain
\[ 2x+3=9 \] \[ \Longleftrightarrow \] \[ 2x=6 \] \[ \Longleftrightarrow \] \[ x=3 \]uses reversible transformations.
So \(x=3\) is guaranteed by the logic of the chain.
Substitution is still useful because it can catch arithmetic mistakes.
Verification Cannot Recover Lost Solutions
Suppose
\[ x(x-4)=0. \]The valid solutions are
\[ x=0,\qquad x=4. \]If we divide by \(x\), we obtain
\[ x-4=0, \]and keep only
\[ x=4. \]Checking \(x=4\) verifies that it is a solution, but verification does not reveal the lost solution \(x=0\).
Added Candidates Versus Lost Solutions
These are different failure modes.
Added candidates can often be removed by checking the original equation.
Lost solutions cannot be recovered unless we notice that an earlier transformation discarded a valid case.
This is why preserving equivalence during solving is more important than relying on verification alone.
Check Every Candidate, Not Just One
Suppose factoring gives
\[ x=-2,\qquad x=1,\qquad x=5. \]Each value is a separate candidate.
If the solving process could have introduced invalid values, every candidate must be tested.
Verifying one does not automatically validate the others.
Verification of Ordered Pairs
For an equation in two variables, a candidate solution is an ordered pair.
Suppose
\[ 2x+y=7 \]and the candidate is
\[ (2,3). \]Substitute both coordinates:
\[ 2(2)+3=7. \]Since
\[ 7=7, \]the ordered pair is a solution.
Verifying a Point Against Several Constraints
Suppose the system of inequalities is
\[ x\ge0, \] \[ y\ge0, \] \[ x+y\le6. \]To verify
\[ (2,3), \]check all three:
\[ 2\ge0, \] \[ 3\ge0, \] \[ 2+3\le6. \]All are true, so the point is feasible.
One Failed Constraint Rejects the Candidate
For the same feasible region, test
\[ (4,4). \]The nonnegativity conditions are satisfied, but
\[ 4+4\le6 \]is false.
Therefore, the point is infeasible.
A candidate must satisfy every required condition.
Inequality Verification
For an interval solution, verification usually means testing representative points and boundaries rather than every value.
Suppose the proposed solution is
\[ x<3. \]Useful checks include:
- an interior point such as \(x=0\);
- a point outside such as \(x=5\);
- the boundary point \(x=3\).
Boundary Verification
If the proposed solution is
\[ x\le3, \]the boundary value \(x=3\) should satisfy the original inequality.
If the proposed solution is
\[ x<3, \]the boundary should fail the strict comparison or produce equality rather than strict inequality.
Interval Verification
Suppose sign analysis gives
\[ (-\infty,-2]\cup[4,\infty). \]A strong check uses:
- one value from the left interval;
- one value from the excluded middle interval;
- one value from the right interval;
- and both boundary points.
This tests the overall sign pattern and endpoint inclusion.
Contextual Verification
Suppose algebra gives
\[ t=-3 \]for a problem where \(t\) represents elapsed time.
The equation may be algebraically satisfied, but the model may require
\[ t\ge0. \]Then \(t=-3\) must be rejected as contextually invalid.
Counts and Integer Restrictions
Suppose algebra gives
\[ n=4.5 \]where \(n\) represents a number of people.
If the context requires whole people, then
\[ n\in\mathbb{Z}_{\ge0}. \]The value \(4.5\) is not an allowable solution to the modeled problem.
Units as a Verification Tool
Suppose a rate is calculated as
\[ r=\frac{d}{t}. \]If \(d\) is measured in miles and \(t\) in hours, then \(r\) should have units
\[ \frac{\text{miles}}{\text{hour}}. \]A unit mismatch can reveal that a formula was rearranged or applied incorrectly.
Dimensional verification complements algebraic substitution.
Sign and Range Checks
Some expressions have built-in output restrictions.
For example:
\[ \sqrt{x}\ge0, \] \[ |x|\ge0, \]and
\[ e^x>0. \]If a candidate would require one of these expressions to equal an impossible value, it can be rejected immediately.
Principal Root Checks
Consider
\[ \sqrt{x}= -3. \]There is no real solution because the principal square root is never negative.
This conclusion comes from range verification before any algebraic manipulation.
Inverse-Function Checks
If solving involves an inverse function, verify that the original value lies in the appropriate domain and that the inverse relation is valid.
For example,
\[ \ln x=2 \]requires
\[ x>0. \]The solution
\[ x=e^2 \]passes that domain check automatically.
Approximate Solutions Need Numerical Verification
Suppose a calculator or numerical method gives
\[ x\approx1.73. \]Substitute the approximation into the original equation and compare the two sides.
Because the value is approximate, we usually expect the two sides to agree within some tolerance rather than exactly.
Residuals
For an equation
\[ f(x)=g(x), \]define the residual
\[ R(x)=f(x)-g(x). \]An exact solution satisfies
\[ R(x)=0. \]An approximate numerical solution should make
\[ |R(x)| \]small.
Residual Example
Suppose we estimate a solution to
\[ x^2=3 \]as
\[ x\approx1.732. \]The residual is
\[ R(x)=x^2-3. \]Substituting:
\[ R(1.732)=1.732^2-3. \]The result is close to zero, confirming that the approximation is reasonable.
Graphical Verification
For
\[ f(x)=g(x), \]a candidate solution should correspond to an intersection of the graphs
\[ y=f(x) \]and
\[ y=g(x). \]A graph can reveal missing or extra candidates, although exact symbolic or numerical checking should still be used when precision matters.
Independent Verification Methods
Confidence increases when a result can be checked in more than one way.
Possible independent checks include:
- direct substitution;
- graphical intersection;
- dimensional analysis;
- reverse algebra;
- an alternative solving method;
- numerical residuals;
- or contextual reasonableness.
Reverse Algebra as Verification
Suppose a formula was rearranged from
\[ A=bh \]to
\[ h=\frac{A}{b}. \]Multiply the rearranged form by \(b\):
\[ bh=A. \]By symmetry of equality,
\[ A=bh. \]The original formula is recovered, provided
\[ b\ne0. \]Check the Conditions of the Reverse Step
Recovering the original expression is not enough if the reverse step requires conditions.
In
\[ h=\frac{A}{b}, \]the condition
\[ b\ne0 \]must remain explicit.
Verification includes checking the legality of the transformations, not only the symbolic appearance of the result.
Reasonableness Checks
Before accepting a result, ask whether it makes sense.
Examples:
- Should a length be negative?
- Should a probability exceed \(1\)?
- Should a count be fractional?
- Should a rate have the expected sign?
- Is the magnitude plausible?
Reasonableness does not replace proof, but it can expose obvious errors quickly.
Verification in Word Problems
For an applied problem, substitute the candidate back into the original relationship described by the problem, not just a derived equation.
Then check whether the result satisfies all contextual conditions.
This helps ensure that the mathematical solution actually answers the question that was asked.
Worked Example: Linear Equation
Solve and verify
\[ 4x-7=13. \]Add \(7\):
\[ 4x=20. \]So
\[ x=5. \]Verify:
\[ 4(5)-7=13, \] \[ 20-7=13, \] \[ 13=13. \]Therefore,
\[ \boxed{x=5}. \]Worked Example: Rational Equation
Solve and verify
\[ \frac{x+2}{x-3}=4. \]Domain:
\[ x\ne3. \]Multiply by \(x-3\):
\[ x+2=4x-12. \]Then
\[ 14=3x, \]so
\[ x=\frac{14}{3}. \]This does not violate the domain.
Substitute into the original equation to confirm the equality.
Worked Example: Radical Equation
Solve and verify
\[ \sqrt{x+5}=x-1. \]The right side must satisfy
\[ x\ge1. \]Square:
\[ x+5=(x-1)^2. \]Expand:
\[ x+5=x^2-2x+1. \]Rearrange:
\[ x^2-3x-4=0. \]Factor:
\[ (x-4)(x+1)=0. \]Candidates:
\[ x=4,\qquad x=-1. \]The value \(x=-1\) violates \(x\ge1\).
Check \(x=4\):
\[ \sqrt{9}=3, \] \[ 4-1=3. \]Therefore,
\[ \boxed{x=4}. \]Worked Example: Logarithmic Equation
Solve and verify
\[ \ln(x-2)=0. \]Domain:
\[ x>2. \]Rewrite exponentially:
\[ x-2=e^0=1. \]So
\[ x=3. \]Verify:
\[ \ln(3-2)=\ln1=0. \]The domain is satisfied, so
\[ \boxed{x=3}. \]Worked Example: Feasible Point
Verify whether
\[ (3,1) \]belongs to the region
\[ x\ge0, \] \[ y\ge0, \] \[ 2x+y\le7. \]Check:
\[ 3\ge0, \] \[ 1\ge0, \]and
\[ 2(3)+1=7\le7. \]All constraints hold, so the point is feasible.
A Reliable Verification Workflow
- Return to the original equation, inequality, or model.
- Check the original domain.
- Substitute the candidate exactly when possible.
- Simplify both sides independently.
- Confirm that the required equality or inequality is true.
- Check contextual restrictions such as sign, units, or integrality.
- Reject any candidate that fails any required condition.
- List only verified candidates in the final solution set.
Verification Checklist by Problem Type
| Problem Type | Important Check |
|---|---|
| Linear equation | Substitute and compare both sides |
| Rational equation | Check denominator restrictions |
| Radical equation | Check original equation after power transformations |
| Logarithmic equation | Check every log argument is positive |
| Inequality | Check boundaries and representative intervals |
| Applied model | Check context, units, and allowed values |
| Numerical approximation | Check residual size |
Common Errors
Checking Only the Final Transformed Equation
A candidate may satisfy a transformed equation but fail the original domain or equation.
Forgetting Domain Restrictions
Denominators, logarithms, and even roots may exclude candidates before substitution.
Checking Only One Candidate
Every candidate must be verified independently when the solving process can introduce invalid values.
Assuming Verification Can Recover Lost Solutions
Checking can reject added candidates, but it cannot restore solutions discarded by invalid division or cancellation.
Ignoring Context
An algebraic solution may fail physical, integer, sign, or unit restrictions.
Rounding Too Early
Premature approximation can create verification error. Keep exact forms as long as practical.
Using Reasonableness Instead of Proof
A value that seems plausible still needs mathematical verification.
Practice
1. Direct Substitution
Verify whether
\[ x=4 \]solves
\[ 2x+3=11. \]Check Answer
\[ 2(4)+3=11, \] \[ 8+3=11. \]True. So \(x=4\) is a verified solution.
2. Reject a Candidate
Does
\[ x=2 \]solve
\[ x^2=5? \]Check Answer
\[ 2^2=4\ne5. \]No.
3. Domain Restriction
Can
\[ x=3 \]solve an equation containing
\[ \frac{1}{x-3}? \]Check Answer
No. The denominator becomes zero, so \(x=3\) is outside the domain.
4. Radical Candidate
Check whether
\[ x=0 \]solves
\[ \sqrt{x+1}=x-1. \]Check Answer
\[ \sqrt{1}= -1, \] \[ 1=-1. \]False. The candidate is extraneous.
5. Logarithmic Domain
Can
\[ x=1 \]solve
\[ \ln(x-2)=0? \]Check Answer
No. The logarithm requires
\[ x-2>0, \]so \(x=1\) is outside the domain.
6. Feasible Point
Does
\[ (2,2) \]satisfy
\[ x\ge0,\qquad y\ge0,\qquad x+y\le5? \]Check Answer
Yes:
\[ 2\ge0, \] \[ 2\ge0, \]and
\[ 2+2=4\le5. \]7. Contextual Check
A calculation gives
\[ n=3.7 \]for the number of buses required. Is this a valid final answer?
Check Answer
Not directly. The context requires a whole number of buses, and if at least \(3.7\) buses are required, the practical answer is usually \(4\) buses.
Key Ideas
- A candidate solution is not automatically a verified solution.
- The original equation or model is the final authority.
- Direct substitution is the most basic verification method.
- Domain restrictions should be checked before or during substitution.
- Extraneous candidates often arise from nonreversible transformations such as squaring.
- Checking a transformed equation is not enough when the original problem had restrictions.
- Verification can remove added candidates but cannot recover lost solutions.
- Every candidate should be checked when the solving method may introduce invalid values.
- Inequality verification uses boundary and interval test points.
- Applied problems require contextual checks involving units, signs, magnitudes, and discrete values.
- Approximate numerical solutions can be checked with residuals.
- Independent verification methods increase confidence and reveal different kinds of errors.
Where This Leads
Verification turns equation solving from a sequence of manipulations into a complete reasoning process: generate candidates, preserve valid cases, reject invalid ones, and confirm that the final solution set actually answers the original problem.
This habit carries directly into systems, numerical methods, proofs, modeling, optimization, and every later setting where mathematical results must be checked against their assumptions and definitions.