Common Mistakes in Algebra Fundamentals
392 common pitfalls across 98 concepts, including 98 recurring stuck points
These are the pitfalls students hit across all 98 algebra fundamentals concepts, from Variables, Expressions and Equations onward. Each entry names the concept it comes from, so you can tell a slip from a genuine gap and go straight to the explanation.
The procedure for variables is the easy part; the trap is treating as always meaning the same number across different problems. Asking "Is a letter being used to hold a number we don't know yet or one that can vary?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating as always meaning the same number across different problems - a variable's value is set by its own equation, not fixed forever.
Reading the letter as the answer instead of a placeholder - the variable names the unknown; solving reveals the number.
Thinking different letters must mean different numbers - and can happen to equal the same value.
The procedure for expressions is the easy part; the trap is trying to solve an expression. Asking "Is there a combination of terms with NO equals sign, so the only moves are simplify or evaluate?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Trying to solve an expression - with no equals sign there is nothing to make true, so you can only simplify or evaluate.
Combining terms that are not alike, like adding and - only like terms (same variable part) can be combined.
Dropping a sign when simplifying, e.g. becoming - the subtraction and the term stay attached.
The procedure for equations is the easy part; the trap is operating on one side only. Asking "Are two expressions joined by with an unknown I'm asked to make the statement true?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Operating on one side only - whatever you add, subtract, multiply, or divide must hit both sides to keep balance.
Forgetting to check the answer - substitute it back; a true statement confirms the solution.
Treating as 'compute the answer next' (calculator habit) - in algebra means the two sides are the same quantity.
The procedure for solving linear equations is the easy part; the trap is doing an operation to only one side. Asking "Is there an equals sign and a variable value to find?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Doing an operation to only one side — preserve equality by doing the same operation to both sides.
Combining unlike terms — only combine terms with the same variable part.
Stopping before checking — substitute the solution back into the original equation.
The procedure for inequalities is the easy part; the trap is forgetting to flip the symbol when multiplying or dividing by a negative. Asking "Is the relation 'less/greater than (or equal)' so the answer is a range, not a single value?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting to flip the symbol when multiplying or dividing by a negative - reverse to (and vice versa) in that step.
Writing one number as the answer - an inequality's solution is a range, shown on a number line or in interval form.
Confusing open and closed dots - use an open circle (not included); use a filled circle (included).
The procedure for coordinate plane is the easy part; the trap is reversing the coordinates. Asking "Am I locating or drawing a position using a horizontal value paired with a vertical value?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reversing the coordinates - is always (horizontal, vertical); .
Mixing up quadrant signs - quadrant II is , quadrant III is ; check both signs.
Starting from a point other than the origin - every address is measured from .
The procedure for slope is the easy part; the trap is computing instead of . Asking "Does the output change by the same amount for each equal step in the input?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Computing instead of — rise goes over run, output over input.
Subtracting the coordinates in a different order top and bottom — keep over with the same first point.
Calling a curved or unequal-step relationship a slope — slope only describes a straight line.
The procedure for linear functions is the easy part; the trap is calling every equation with linear. Asking "Does the output change by the same amount each time the input step is the same?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling every equation with linear — check for constant rate and no powers like .
Confusing slope with y-intercept — slope is the repeated change, while the y-intercept is the starting value when the input is 0.
Using two points without checking the whole pattern — a line through two points is linear, but a table or story still needs every equal input step to fit the same rate.
The procedure for systems of equations is the easy part; the trap is solving only one equation and ignoring the other. Asking "Does the answer need to make every equation true?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Solving only one equation and ignoring the other — a system solution must satisfy all equations.
Treating the intersection as just an value — for two-variable systems, the solution is an ordered pair.
Using substitution or elimination without aligning variables — keep equations equivalent at each step.
The procedure for quadratic functions is the easy part; the trap is forgetting may be negative. Asking "Is the highest power of the variable exactly 2, so the graph curves into a parabola?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting may be negative - then the parabola opens downward and has a maximum, not a minimum.
Reading the vertex from the standard form directly - use or convert to .
Expecting one x-value per y - a horizontal line can cross a parabola twice, so two inputs can share an output.
The procedure for quadratic formula is the easy part; the trap is using the formula before setting the equation to zero. Asking "Do I have a quadratic equation set to zero whose exact solutions I need?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using the formula before setting the equation to zero - rearrange to first.
Mishandling signs of in - if , then .
Dropping the and reporting one root - the produces two solutions.
The procedure for polynomials is the easy part; the trap is counting an expression with a negative or fractional exponent as a polynomial. Asking "Is every exponent on the variable a whole number with no variable in a denominator?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Counting an expression with a negative or fractional exponent as a polynomial - powers must be whole numbers .
Misreading the degree - it's the highest power present, not the number of terms.
Forgetting to combine like terms before naming degree or leading coefficient - simplify to standard form first.
The procedure for factoring is the easy part; the trap is forgetting to pull out the greatest common factor first. Asking "Am I rewriting an expression as a product of simpler factors that multiply back to it?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting to pull out the greatest common factor first - factor out the GCF before any other pattern.
Sign errors in the binomials - for both factors are negative: .
Assuming every quadratic factors over integers - if no integer pair works, switch to the quadratic formula.
The procedure for variable as placeholder is the easy part; the trap is thinking the placeholder can be any number. Asking "Does the condition pin the variable to one specific value I'm meant to find?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Thinking the placeholder can be any number - the equation constrains it to one specific value.
Treating the letter as the final answer - it names the unknown; solving reveals the number.
Forgetting to verify by substitution - plug the value back to confirm it fits the condition.
The procedure for variable as generalization is the easy part; the trap is solving a generalization for a value. Asking "Is the letter meant to stand for ANY value, making the statement true universally?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Solving a generalization for a value - there's no single solution; it holds for all values.
Plugging in one number and concluding it's proven - one case doesn't establish a 'for all' claim.
Confusing it with a placeholder - a generalization means ANY value, not one specific unknown.
The procedure for evaluation is the easy part; the trap is ignoring order of operations after substituting. Asking "Are the variable's values given so I just substitute and compute one number?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Ignoring order of operations after substituting - compute exponents and multiplication before addition.
Forgetting parentheses around the substituted value - write , not .
Confusing evaluating with solving - evaluation needs given values; solving finds unknown ones.
The procedure for substitution is the easy part; the trap is dropping parentheses around the substituted expression. Asking "Am I replacing a variable with an EQUAL expression everywhere it appears?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Dropping parentheses around the substituted expression - wrap it so grouping survives.
Replacing only some occurrences of the variable - substitute every occurrence consistently.
Substituting a value when the variable is still unknown - solve for it first, then substitute.
The procedure for identity vs equation is the easy part; the trap is calling every equation an identity. Asking "Does the equality hold for EVERY value of the variable, or only for special ones?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling every equation an identity - test more than one value; an identity must hold for all.
Interpreting as no solution - it means every value works (an identity).
Trying to find a single answer to an identity - there isn't one; it's universally true.
The procedure for solution concept is the easy part; the trap is skipping the check. Asking "Does plugging this value back into the original statement make it true?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Skipping the check - substitute the candidate back to confirm it makes the statement true.
Accepting an extraneous solution - values introduced by squaring or clearing denominators must be verified.
Confusing a solution with the expression's value - a solution makes a statement TRUE, not just produces a number.
The procedure for solution set is the easy part; the trap is giving one value when there are many. Asking "Am I describing EVERY value that satisfies the statement, not just one?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Giving one value when there are many - inequalities and quadratics often have whole ranges or multiple answers.
Forgetting the empty set is a valid answer - if nothing satisfies it, write .
Mixing up open and closed brackets in interval notation - excludes the endpoint, includes it.
The procedure for constraint system is the easy part; the trap is satisfying some constraints but not all. Asking "Must the same values satisfy multiple conditions at the same time?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Satisfying some constraints but not all - the answer must meet every condition simultaneously.
Ignoring inequality constraints once an equation is solved - check the solution against the inequalities too.
Treating the conditions as alternatives - they're joined by AND, so all must hold, not just one.
The procedure for proportional line is the easy part; the trap is calling a line with a nonzero -intercept proportional. Asking "Does the line pass through with the same for every point?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Calling a line with a nonzero -intercept proportional - proportional lines must pass through the origin.
Confusing the constant ratio with the intercept - is the slope here, and the intercept is 0.
Checking only one point for a constant ratio - must hold for every point on the line.
The procedure for rate of change (algebraic) is the easy part; the trap is subtracting the points in inconsistent order. Asking "Am I dividing a change in output by a change in input over an interval?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Subtracting the points in inconsistent order - keep with the same first point on top and bottom.
Treating the average over a curve as a single constant rate - it's only the average between the two chosen points.
Dividing inputs by outputs - it's change in output over change in input, not the reverse.
The procedure for algebraic representation is the easy part; the trap is using the same letter for two different quantities. Asking "Can I name the varying quantity with a letter and write the relationship as an equation?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using the same letter for two different quantities - give each varying quantity its own letter and write what it stands for.
Putting the fixed number where the rate belongs - the 'per each' amount multiplies the variable; the one-time amount stands alone.
Representing a constant as a variable - only the quantity that actually changes gets a letter.
The procedure for symbolic abstraction is the easy part; the trap is trying to solve a universal identity for a value. Asking "Am I making a claim meant to hold for every value, not just the numbers in front of me?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Trying to solve a universal identity for a value - an identity like is true for all values, so there is nothing to solve.
Generalizing from one example - a single case does not prove a 'for all' claim; the letters must cover every case.
Reusing one letter to mean two different things - each abstracted symbol must stand for the same quantity throughout.
The procedure for rewriting expressions is the easy part; the trap is distributing to only the first term. Asking "Are these two expressions equal at every value of the variable, just written differently?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Distributing to only the first term - , not ; the factor hits every term inside.
Combining unlike terms - does not simplify to ; only matching terms combine.
Changing the value instead of just the form - check by plugging in one number; both forms must agree.
The procedure for factoring intuition is the easy part; the trap is finding numbers that add to b but ignore the product. Asking "Am I turning a sum of terms into a product of simpler factors?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Finding numbers that add to b but ignore the product - the pair must satisfy both and .
Mixing up signs - for both numbers are negative (); check the sign of and .
Forgetting a common factor first - pull out shared factors before hunting for the pair, e.g. .
The procedure for expansion intuition is the easy part; the trap is dropping the cross term when squaring. Asking "Am I turning a product of factors into a sum of terms by distributing?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Dropping the cross term when squaring - has a middle , not just .
Distributing to only one term - every term in the first factor multiplies every term in the second.
Forgetting to combine like terms - after FOIL, add the inner and outer terms ().
The procedure for equivalence transformation is the easy part; the trap is changing only one side. Asking "Does this step act on both sides equally so the solution set is unchanged?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Changing only one side - whatever you add, subtract, multiply, or divide must hit both sides.
Multiplying or dividing by zero - the multiplier must be nonzero or solutions are lost.
Doing different operations to each side - the SAME operation with the SAME number must apply to both.
The procedure for isolating variable is the easy part; the trap is undoing multiplication before addition. Asking "Am I peeling operations off the variable until it stands alone on one side?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Undoing multiplication before addition - reverse the evaluation order: strip the added constant first, then the coefficient.
Dividing only part of a side - when you divide by the coefficient, divide every term on both sides.
Applying the inverse to one side only - each peel is a both-sides operation.
The procedure for dependent vs independent variables is the easy part; the trap is swapping the axes. Asking "Which quantity do I choose freely, and which one is then determined by it?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Swapping the axes - independent goes on the horizontal -axis, dependent on the vertical -axis.
Assuming the dependent variable can be chosen - only the independent input is free; the output follows the rule.
Confusing correlation direction - the dependent quantity responds to the independent one, not the reverse.
The procedure for modeling with equations is the easy part; the trap is flipping subtraction order. Asking "Am I turning a described situation into an equation I can then solve for the unknown?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Flipping subtraction order - '7 less than x' is , not ; translate meaning, not word order.
Skipping the 'let x =' definition - state exactly what the variable represents before writing the equation.
Ignoring a stated constraint - every condition in the story must appear in the model.
The procedure for parameter is the easy part; the trap is treating the parameter as the input variable. Asking "Is this quantity held constant within one case but changed to produce a different member of a family?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating the parameter as the input variable - within one line, and are fixed while varies.
Treating a parameter as a universal constant - parameters change between cases; constants like never do.
Solving for the parameter as if it were the unknown - to find you typically use given points, not isolate them from one equation.
The procedure for constant vs variable is the easy part; the trap is trying to solve for a named constant. Asking "Could this symbol's value be different in another situation, or is it locked forever?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Trying to solve for a named constant - and have fixed values; you never isolate them.
Assuming every letter is a variable - letters like or often stand for fixed unknown constants.
Forgetting context - the same letter can be a constant in one problem and a variable in another.
The procedure for degrees of freedom is the easy part; the trap is counting dependent equations as constraints. Asking "After applying all independent constraints, how many values can I still choose freely?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Counting dependent equations as constraints - only independent equations reduce ; redundant ones don't.
Forgetting that more variables than equations means free choices - gives infinitely many solutions, provided the equations do not contradict each other.
Confusing zero degrees with a guaranteed unique solution - only allows uniqueness; a contradiction can still make .
The procedure for linear system behavior is the easy part; the trap is equating 'two equations' with 'one solution'. Asking "Am I classifying how many solutions a linear system has by how its lines relate?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Equating 'two equations' with 'one solution' - parallel or coincident lines break that assumption.
Confusing parallel with coincident - equal slope and different intercept is parallel (none); equal everything is the same line (infinite).
Solving before checking - compare coefficient ratios first to spot no-solution or infinite-solution cases.
The procedure for consistency is the easy part; the trap is equating consistency with a unique solution. Asking "Is there at least one set of values that makes every equation true at the same time?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Equating consistency with a unique solution - infinitely many solutions still counts as consistent.
Calling a system inconsistent because it has free variables - free variables mean many solutions, not zero.
Missing the signal - a false numeric statement during elimination means inconsistent, stop solving.
The procedure for redundancy is the easy part; the trap is counting a redundant equation as a real constraint. Asking "Is this equation just a combination of the others, telling me nothing new?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Counting a redundant equation as a real constraint - it leaves unchanged; the system is less constrained than it looks.
Confusing redundant with contradictory - all ratios equal is redundant (); equal slopes but unequal constants is contradictory ().
Expecting a unique solution from duplicate equations - redundancy typically leaves free variables and infinitely many solutions.
The procedure for contradiction is the easy part; the trap is reading as a contradiction. Asking "Have I reached a statement that no values could ever make true?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as a contradiction - is always true (redundant); only with is a contradiction.
Continuing to solve after a contradiction - once you hit , stop: there is no solution.
Forgetting a contradiction can come from a setup error - recheck arithmetic before declaring no solution.
The procedure for algebraic symmetry is the easy part; the trap is assuming any expression with both variables is symmetric. Asking "If I swap the two variables, do I get back the exact same expression?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming any expression with both variables is symmetric - actually swap and compare; fails.
Confusing symmetry with the commutative property - symmetry is about an expression's invariance, not an operation's reordering.
Checking only one term - every term must survive the swap for the whole expression to be symmetric.
The procedure for dimensional consistency is the easy part; the trap is adding terms with different units. Asking "Does every term being added or equated carry the same units?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding terms with different units - only quantities with the same dimensions can be added or equated.
Ignoring units because the arithmetic works - matching numbers don't guarantee matching dimensions.
Forgetting to check both sides - must equal , not just the terms within one side.
The procedure for symbolic overload is the easy part; the trap is reading '' as automatically negative. Asking "Does this symbol have more than one possible meaning that only context resolves?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading '' as automatically negative - it means 'opposite of ,' which is positive when is negative.
Assuming a symbol's meaning is fixed everywhere - check the surrounding context each time.
Confusing the negative sign with subtraction - labels a number; is an operation between two.
The procedure for structure vs computation is the easy part; the trap is computing first and missing the shortcut. Asking "Can I solve this by recognizing its form instead of grinding through the arithmetic?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Computing first and missing the shortcut - scan for a recognizable form before doing arithmetic.
Forcing a pattern that isn't there - confirm the form actually matches before applying it.
Treating structural insight as 'not real math' - recognizing form is the higher-leverage skill, not a shortcut to skip.
The procedure for expression simplification is the easy part; the trap is combining unlike terms like into one term. Asking "Am I making one expression cleaner without changing its value (no equation to solve)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Combining unlike terms like into one term - only terms with the same variable AND same exponent combine.
Adding the exponents when combining like terms () - the exponent stays the same; only coefficients add.
Forgetting to distribute a leading minus or coefficient before combining - clear parentheses first, then combine.
The procedure for algebraic pattern is the easy part; the trap is forcing to factor as . Asking "Does this expression match the template of a known identity slot-for-slot?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forcing to factor as - only the DIFFERENCE of squares factors; the sum does not over the reals.
Mismatching the cube signs ( vs ) - SOAP: the middle term sign is Opposite the binomial sign, the last is Always Positive.
Missing that a coefficient is a perfect square () - check whether each piece is itself a square or cube before deciding the pattern fits.
The procedure for functional dependency is the easy part; the trap is allowing one input to give two outputs and still calling it a function. Asking "Does every allowed input produce exactly one output (never two)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Allowing one input to give two outputs and still calling it a function - apply the vertical line / one-output test.
Confusing dependency with causation - says determines mathematically, not that physically causes .
Swapping which variable is the input - the dependent (output) variable is the one being computed FROM the independent (input) one.
The procedure for abstraction level is the easy part; the trap is proving a universal claim by checking a few examples. Asking "Is this claim about one particular case, or about every object of its kind?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Proving a universal claim by checking a few examples - examples can suggest but never prove a 'for all' statement.
Disproving a general rule and thinking you must check more cases - a single counterexample is enough to kill a universal claim.
Confusing a concrete instance with the general law it illustrates - is one case of commutativity, not the law itself.
The procedure for algebra as language is the easy part; the trap is reading as a concatenated number or 'five and x'. Asking "Am I treating these symbols as a language with grammar that must be read or written correctly?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading as a concatenated number or 'five and x' - adjacency in algebra means multiplication.
Translating 'less than' in original word order - '3 less than x' is , not ; the language reverses it.
Writing grammatically broken notation like - every expression must be a well-formed sentence in the language.
The procedure for algebra as structure is the easy part; the trap is assuming every operation is commutative or associative. Asking "Is the question about the properties of an operation, rather than computing a number?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming every operation is commutative or associative - structure means you must CHECK each property, not assume it (matrix multiplication is not commutative).
Confusing the identity element with zero specifically - the identity is whatever leaves elements unchanged under THAT operation (1 for multiplication).
Treating structure as just harder number-crunching - the object of study is the property/behavior, not a numerical answer.
The procedure for algebraic invariance is the easy part; the trap is assuming all features are invariant under a transformation. Asking "Does this quantity stay exactly the same after the allowed transformation is applied?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Assuming all features are invariant under a transformation - state WHICH transformation and check each quantity separately.
Confusing 'the expression changed' with 'the invariant changed' - the form can change while the invariant (e.g. degree) stays fixed.
Calling a number invariant without naming the transformation - invariance is always relative to a specified allowed operation.
The procedure for algebraic constraint is the easy part; the trap is ignoring implicit constraints like denominator or radicand . Asking "Does this condition restrict the set of permissible values rather than ask for a single computation?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Ignoring implicit constraints like denominator or radicand - always exclude values that break the expression.
Treating a constraint as a single answer - a constraint defines a SET of allowed values, often infinitely many.
Mixing up and at the boundary - a closed inequality includes the boundary value, a strict one excludes it.
The procedure for binomial theorem is the easy part; the trap is distributing the exponent as . Asking "Am I raising a two-term expression to a whole-number power and want its expansion or a single term?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Distributing the exponent as - every cross term with is required.
Letting the and exponents not sum to - in each term the powers must total .
Forgetting the coefficient on a chosen term - the term is , not just .
The procedure for quadratic standard form is the easy part; the trap is forgetting . Asking "Is this a single-variable degree-2 equation written as everything-minus-everything with ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting - if the coefficient is 0 it is linear, not quadratic.
Reading off before collecting like terms or moving everything to one side - get it to first.
Misreading when the equation is not set to zero - subtract the right side over so the constant on the left is the true .
The procedure for quadratic vertex form is the easy part; the trap is taking with the wrong sign. Asking "Is the quadratic written as a squared binomial plus a constant, and do I want its turning point?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Taking with the wrong sign - means is the value that makes the inside zero, so has .
Forgetting also flips/stretches - a negative opens the parabola downward (vertex is a maximum).
Reading from standard form - convert to first; the vertex is not the and .
The procedure for quadratic factored form is the easy part; the trap is reading roots with the wrong sign. Asking "Is the quadratic written as a product of linear factors, and do I want where it equals zero?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reading roots with the wrong sign - the root of is , so gives root .
Ignoring the leading factor - does not change the roots but is needed to match the full function.
Setting the whole product equal to a nonzero number and 'solving' each factor - the zero-product trick only works when the product equals 0.
The procedure for completing the square is the easy part; the trap is adding without also subtracting it. Asking "Am I manufacturing a perfect-square trinomial by adding and subtracting the same value to keep it equal?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding without also subtracting it - you must keep the expression equal; add and subtract the same amount.
Adding without first factoring out a leading coefficient — factor out , complete the square inside, then distribute.
Halving but not squaring it - the value added is the SQUARE of half the middle coefficient.
The procedure for discriminant is the easy part; the trap is computing with the wrong sign on . Asking "Do I only need to know how many/what kind of roots a quadratic has, not their values?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Computing with the wrong sign on - is always nonnegative, so square the value of carefully (e.g. ).
Dropping the - the discriminant is the whole , not just .
Saying 'no solutions' when - there are no REAL solutions, but two complex ones.
The procedure for graphing parabolas is the easy part; the trap is getting the opening direction wrong. Asking "Am I producing or reading a picture of a quadratic, using vertex, axis, and intercepts?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Getting the opening direction wrong - opens up, opens down; check the sign first.
Using for the axis - the axis is (note the minus).
Connecting points with straight segments - a parabola is a smooth curve, not a polygon.
The procedure for vertex and axis of symmetry is the easy part; the trap is reporting only the x-coordinate as 'the vertex'. Asking "Am I after the turning point or the mirror line of a parabola?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reporting only the x-coordinate as 'the vertex' - the vertex is the point ; plug back to get .
Writing the axis of symmetry as a number instead of an equation - it is the line .
Using - the axis/vertex x-coordinate is .
The procedure for zeros of a quadratic is the easy part; the trap is taking only the branch of . Asking "Am I looking for the x-values that make the quadratic equal zero?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Taking only the branch of - a quadratic generally has two zeros; keep both signs.
Reporting the zeros with flipped signs from factors - gives zero , gives zero .
Calling 'no real zeros' an error - when the parabola simply does not cross the x-axis.
The procedure for polynomial addition and subtraction is the easy part; the trap is adding exponents instead of coefficients. Asking "Am I joining two polynomials with or and merging only same-power terms?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding exponents instead of coefficients - ; the exponent stays.
Not distributing the subtraction sign to every term of the second polynomial - .
Combining unlike terms like and - only identical variable-and-power terms merge.
The procedure for polynomial multiplication is the easy part; the trap is squaring a binomial as . Asking "Am I multiplying expressions so that every term in one meets every term in the other?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Squaring a binomial as - includes the cross term.
Missing a cross-product term (forgetting Outer or Inner in FOIL) - every term must meet every term.
Adding exponents wrong when multiplying terms - (add exponents only when MULTIPLYING).
The procedure for factoring out the gcf is the easy part; the trap is taking the highest variable power instead of the lowest. Asking "Do all terms share a numerical and/or variable factor I can lift to the front?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Taking the highest variable power instead of the lowest - the common variable factor is the SMALLEST exponent present.
Forgetting a term's leftover, e.g. factoring as drops the ; it should be - every original term must reappear inside the parentheses.
Not pulling the largest numerical factor - has GCF 3, not 1; take the greatest common number too.
The procedure for factoring difference of squares is the easy part; the trap is trying to factor the same way. Asking "Are both terms perfect squares with a minus sign between them and nothing in the middle?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Trying to factor the same way — a sum of squares does not factor over the reals; only the difference does.
Stopping after one step on — refactor the new difference of squares: .
Forgetting a coefficient is a square too — has , giving , not .
The procedure for factoring trinomials is the easy part; the trap is using instead of when . Asking "Can I find two numbers that multiply to (or ) and add to ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using instead of when — the AC method needs the product , then split the middle and group.
Getting the signs of and wrong — match signs to (same sign if , opposite if ) and to .
Forgetting to pull a GCF first — factor out the common factor before searching for the pair, e.g. .
The procedure for factoring by grouping is the easy part; the trap is sign error pulling the GCF from the second pair. Asking "After factoring the GCF from each pair, do both pairs leave behind the identical binomial?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Sign error pulling the GCF from the second pair — when the pair is , factor out to get so the binomial matches the first pair.
Giving up when pairs do not match — reorder the four terms (group differently) before deciding it is unfactorable.
Forgetting the final factor-out step — after you must write , not leave it half-done.
The procedure for simplifying radicals is the easy part; the trap is using a non-perfect-square factor. Asking "Does the number under the root have any perfect-square factor bigger than 1?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using a non-perfect-square factor — , not ; pick the factor pair where one factor is a perfect square.
Distributing the root over a sum — ; the product rule applies only to multiplication.
Leaving a perfect square inside — is not finished; keep factoring until .
The procedure for radical operations is the easy part; the trap is adding the radicands when multiplying. Asking "For or , do the radicands match — and have I simplified first to check?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding the radicands when multiplying — , not ; multiply under one root, do not add.
Combining unlike radicals — stays as is; only same-radicand terms combine.
Skipping simplification — looks unlike but becomes after simplifying.
The procedure for rationalizing denominators is the easy part; the trap is multiplying only the denominator. Asking "Is there a square root in the denominator, and is it a single term or a binomial?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Multiplying only the denominator — you must multiply numerator AND denominator by the same factor to keep the value unchanged (you are multiplying by 1).
Using the same binomial instead of the conjugate — for use so the cross term cancels via difference of squares.
Forgetting to simplify after — should reduce to .
The procedure for radical equations is the easy part; the trap is skipping the check. Asking "Is the unknown trapped under a radical that I undo by squaring, and did I check for extraneous roots?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Skipping the check — squaring can introduce extraneous roots, so substitute every solution back into the ORIGINAL equation.
Squaring a sum termwise — , not ; isolate the radical first to avoid this.
Accepting a negative result of a principal root — if isolating gives , there is no real solution since a principal root is nonnegative.
The procedure for simplifying rational expressions is the easy part; the trap is canceling terms instead of factors. Asking "Are the things I want to cancel FACTORS (multiplied) and not TERMS (added)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Canceling terms instead of factors — does not simplify to ; only common factors cancel, and is a sum.
Forgetting domain restrictions — the canceled in still requires in the answer.
Not fully factoring first — must become before canceling.
The procedure for multiplying and dividing rational expressions is the easy part; the trap is flipping the wrong fraction. Asking "Are the fractions joined by or (so I cancel and multiply across) rather than or ?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Flipping the wrong fraction — when dividing, take the reciprocal of the DIVISOR (the second fraction), not the first.
Multiplying before factoring — factor every numerator and denominator first so cross-cancellation simplifies the work.
Forgetting domain restrictions from the divisor — in division, values that make the original divisor zero are also excluded.
The procedure for adding and subtracting rational expressions is the easy part; the trap is adding denominators too. Asking "Do the denominators match yet — and if not, what is the LCD I must rewrite both over?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding denominators too — , the denominator stays , it does not become .
Distributing the subtraction sign incompletely — requires subtracting EVERY term of .
Using a denominator that is not the least common — multiplying all denominators works but creates extra simplifying; build the LCD.
The procedure for solving rational equations is the easy part; the trap is multiplying only some terms by the LCD. Asking "Is this an EQUATION (has ) with the variable in a denominator that I clear by the LCD and then check?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Multiplying only some terms by the LCD — every term, including non-fraction ones, must be multiplied to keep the equation balanced.
Forgetting to check excluded values — a solved that zeros any original denominator is extraneous and rejected.
Treating it like simplification — with an equals sign you clear all denominators, you do not just rewrite over the LCD.
The procedure for multi-step equations is the easy part; the trap is distributing only to the first term. Asking "Does isolating take more than one step — distributing, combining, or moving variables first?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Distributing only to the first term — , not ; multiply the outside number by EVERY term inside.
Performing inverse operations out of order — undo addition/subtraction before multiplication/division (reverse of order of operations).
Sign error moving a variable across — subtracting from both sides of gives , not .
The procedure for writing equations from context is the easy part; the trap is skipping the variable definition. Asking "Can I name the unknown and write two expressions that must be equal in this situation?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Skipping the variable definition — write what the variable counts and include the units before building the equation.
Reversing subtraction or comparison language — translate from the quantity being described, not from the order of the words alone.
Writing an expression when the context needs an equation — look for the total, equality, or condition that completes the mathematical sentence.
The procedure for matrix definition is the easy part; the trap is stating dimensions columns-first. Asking "Are the numbers arranged in a fixed rectangle of rows and columns with a stated size?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Stating dimensions columns-first — always write as rows columns.
Mixing up the entry subscript order — is row then column , not the reverse.
Calling a non-rectangular set of numbers a matrix — every row must have the same number of entries.
The procedure for matrix addition, subtraction, and scalar multiplication is the easy part; the trap is adding mismatched-size matrices. Asking "For , do the two matrices have exactly the same dimensions?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding mismatched-size matrices — addition/subtraction only works when dimensions are identical.
Scaling only the first row or entry — a scalar multiplies EVERY entry of the matrix.
Confusing with — a scalar is a single number, not another matrix; do not row-by-column it.
The procedure for matrix multiplication is the easy part; the trap is multiplying when inner dimensions disagree. Asking "Does the column count of equal the row count of , and am I dotting rows with columns?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Multiplying when inner dimensions disagree — is undefined; the inner numbers must match.
Assuming — matrix multiplication is generally non-commutative, so order matters.
Multiplying entrywise — each result entry is a row-by-column SUM of products, not a single product.
The procedure for determinant is the easy part; the trap is computing instead of . Asking "Is the matrix square, and am I asking whether it is invertible or how it scales area?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Computing instead of — the off-diagonal product is SUBTRACTED.
Trying to take a determinant of a non-square matrix — determinants exist only for square matrices.
Assuming a negative determinant means an error — a determinant can be negative; only zero means non-invertible.
The procedure for inverse matrix is the easy part; the trap is forgetting the factor. Asking "Is the matrix square with nonzero determinant, so an undo-matrix exists?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Forgetting the factor — the inverse scales the adjugate by .
Mis-swapping entries — for , swap and , negate and : .
Attempting to invert when — a singular matrix has no inverse; check the determinant first.
The procedure for solving systems of equations with matrices is the easy part; the trap is using when . Asking "Can I write the system as and solve mechanically by row ops, inverse, or determinants?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using when — there is no unique solution; switch to row reduction to classify it.
Setting up the augmented matrix in the wrong order — align coefficients column by variable and put constants after the bar.
Mishandling a row operation — only swap rows, scale a row, or add a multiple of one row to another.
The procedure for vector addition, subtraction, and scalar multiplication is the easy part; the trap is adding magnitudes instead of components. Asking "Am I adding/subtracting matching components, or multiplying one vector by a single number?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding magnitudes instead of components — , not a length of 7.
Scaling only one component — multiplies EVERY component by .
Treating subtraction as commutative — ; they point opposite ways.
The procedure for vector magnitude and direction is the easy part; the trap is adding components for length. Asking "Am I asked how long the arrow is or which way it points, rather than how to combine arrows?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding components for length — magnitude squares, sums, then square-roots; it is not .
Forgetting to normalize for a unit vector — divide the whole vector by its magnitude so the length becomes exactly 1.
Ignoring the quadrant for the direction angle — may need an adjustment depending on the signs of the components.
The procedure for dot product is the easy part; the trap is reporting a vector as the answer. Asking "Do I have two vectors and need one number measuring their directional agreement (not a new vector)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Reporting a vector as the answer - the dot product is always a single scalar; multiply matching components and add them
Pairing components crosswise like - pair like-with-like:
Forgetting that form needs magnitudes - , so solve for the angle
The procedure for cross product is the easy part; the trap is returning a scalar. Asking "Do I have two 3D vectors and need a new vector perpendicular to both (or the area they span)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Returning a scalar - the cross product is a vector; build all three components
Reversing the order without flipping sign - , so order matters
Forgetting the middle component's sign pattern - the determinant expansion makes the term negative; the formula already bakes this in as
The procedure for algebraic manipulation is the easy part; the trap is changing only one side of an equation. Asking "Does every step keep exactly the same set of true values, just written differently?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Changing only one side of an equation - whatever you do to one side you must do to the other to keep it equivalent
Cancelling a factor that could be zero - dividing both sides by can delete the solution ; factor instead
Distributing a sign or exponent incorrectly - ; expand carefully so the rewrite stays equivalent
The procedure for linear programming is the easy part; the trap is searching the interior for the best value. Asking "Am I optimizing a linear objective over a region bounded by linear inequalities, with the answer expected at a corner?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Searching the interior for the best value - the optimum of a linear objective lies at a vertex of the feasible region; evaluate the objective at each corner
Forgetting hidden constraints like - real quantities are often nonnegative, which changes the corners
Mixing up maximize and minimize - read whether you want the largest or smallest objective value before comparing corners
The procedure for algebraic identities is the easy part; the trap is dropping the middle term: writing . Asking "Is this equality true for every value of the variable (an always-true pattern), rather than only for special values?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Dropping the middle term: writing - the perfect-square identity has :
Trying to 'solve' an identity for a unique value - every value satisfies it, so there is nothing to solve
Mismatching the sign pattern - , but does not factor over the reals
The procedure for checking solutions is the easy part; the trap is checking against a transformed equation instead of the original. Asking "Do I have a candidate value that I plug into the original condition to confirm both sides are equal?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Checking against a transformed equation instead of the original - extraneous roots pass the squared step but fail the original; always substitute into the starting equation
Skipping the check after squaring or clearing denominators - these are exactly the steps that introduce false solutions
Evaluating only one side - compute the left and right sides separately and confirm they are equal, not just that one side computes
The procedure for interval notation is the easy part; the trap is using a bracket on infinity. Asking "Am I describing a continuous range of reals where I must mark each endpoint as included (bracket) or excluded (parenthesis)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using a bracket on infinity - and are never included, so always use a parenthesis:
Mixing up which symbol includes - square bracket includes the endpoint (); parenthesis excludes it ()
Reading as a point - in interval context it means all reals strictly between 2 and 5, not a coordinate
The procedure for vector addition is the easy part; the trap is adding the lengths instead of the components. Asking "Do I have two vectors acting together and want the single combined (resultant) vector?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Adding the lengths instead of the components - is not ; add matching components first, then find the length
Pairing components crosswise - add to and to :
Joining the tips when the arrows are drawn tail-to-tail - the arrow from the tip of to the tip of is the difference ; for the sum, place them tip-to-tail, or tail-to-tail and take the parallelogram diagonal from the shared tail
The procedure for absolute value equations is the easy part; the trap is keeping only the positive case. Asking "Is an expression inside absolute-value bars set equal to a constant, asking which values are that distance away?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Keeping only the positive case - means or ; solve both
Solving when the right side is negative - has no solution because distance can't be negative
Splitting before isolating the bars - first get alone (e.g. ), then split into the two cases
The procedure for absolute value inequalities is the easy part; the trap is treating like . Asking "Are absolute-value bars set or a value, asking for a range within or outside a distance (not an exact distance)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Treating like - gives one interval , but gives two rays or
Flipping into a sandwich for a 'greater than' - is the solution of , the opposite set, and has no solutions; for write two rays, or
Splitting before isolating the bars - first isolate , then convert; e.g. becomes first
The procedure for graphing inequalities is the easy part; the trap is using a solid line for a strict inequality. Asking "Is the solution of a two-variable inequality a shaded region of the plane bounded by a line?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Using a solid line for a strict inequality - and get a dashed boundary; only and get a solid one
Shading the wrong half-plane - pick a test point not on the line (often ); shade its side only if it makes the inequality true
Forgetting to shade at all - the solution is the whole half-plane region, not just the boundary line
The procedure for ordered pairs is the easy part; the trap is plotting up first then across. Asking "Am I naming a single point with a horizontal value first and a vertical value second?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Plotting up first then across - the first number is horizontal (), the second is vertical ()
Treating and as the same point - order matters, so they are different locations
Confusing the point with the interval - in coordinate context it's one point, not a range of numbers
The procedure for simple interest is the easy part; the trap is compounding by accident. Asking "Is each period's interest computed on the original principal alone (so the yearly interest never changes)?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Compounding by accident - simple interest uses the original every year: , not interest on a growing balance
Leaving the rate as a whole percent - convert to a decimal first (5% becomes )
Mismatching the time units - is in years to match an annual rate; 6 months is
The procedure for function tables and graphs is the easy part; the trap is plotting instead of . Asking "Can I match each table row to a graph point?" first is what keeps a correct-looking calculation from being attached to the wrong concept.
Plotting instead of — input is the horizontal coordinate.
Reading graph points between listed table values without checking the context — some patterns are discrete.
Ignoring scale on graph axes — a visually steep line may just use a different scale.