Assam HS Class 11-12 Business Mathematics & Statistics common mistakes
Assam HS Class 11-12 Business Mathematics & Statistics · Understand the traps and how to avoid them.
When computing simple interest, the candidate multiplies P × R × T using the rate as a percentage number (e.g. 8) instead of a decimal (0.08), producing answers 100 times too large, and then 'fixes' the unit by guessing where to shift the decimal.
School arithmetic drills treat '8%' as the number 8, and the formula SI = PRT/100 is often memorised as a string without the candidate tracking that R is a rate per year expressed as a fraction. Under time pressure the /100 step is dropped or applied twice.
What to do instead
Write the formula once per problem as SI = P × (R/100) × T and substitute R as a decimal explicitly on the answer script. Sanity-check with a benchmark: interest on ₹10,000 at 8% for 1 year must be about ₹800, so any answer in the tens of thousands signals a factor-of-100 error. Practise 10 mixed problems where T is given in months (convert to T/12) and in days (use 365 or 360 as instructed).
In compound interest and annuity problems, the candidate counts the compounding periods wrongly - using the number of years as n when interest compounds half-yearly or quarterly, and using the annuity-due formula when instalments are paid at period end (or vice versa).
The formula (1+i)^n is memorised with n as 'years' because textbook examples are annual. When the question says 'compounded quarterly for 3 years', the candidate keeps n = 3 instead of 12 and keeps the annual rate instead of the quarterly rate. The due/ordinary distinction is invisible because both formulas look similar.
What to do instead
Before substituting, write down three lines: rate per period i = annual rate ÷ number of conversions per year; number of periods n = years × conversions per year; timing of each payment (beginning = annuity due, multiply the ordinary annuity factor by (1+i)). Do a two-step check: compute the annual-rate answer and the converted answer, and confirm the converted one is slightly larger for the same nominal rate. For every annuity word problem, underline 'at the end of each year/quarter' or 'in advance' before choosing the factor.
In present value and future value problems the candidate confuses the two, discounting when the question asks for accumulation (or the reverse), and mixes up the present value of a single sum with the present value of an annuity.
Both use the same symbols and the only cue is the wording 'today', 'now', 'after 5 years', 'accumulated'. Candidates scan for numbers and pick whichever formula they practised last; the annuity-versus-lump-sum distinction is lost because both are 'present value'.
What to do instead
Adopt a fixed translation routine: rewrite every question into one of four templates - accumulate a lump sum (FV = P(1+i)^n), discount a lump sum (PV = FV/(1+i)^n), accumulate a series of payments (future value of annuity), value a series of payments today (present value of annuity). Draw a timeline with the unknown marked. Then state in one sentence what the money is doing before touching the calculator.
When finding a maximum or minimum of a business function (profit, revenue, cost), the candidate differentiates, sets the derivative to zero, solves for quantity, and stops - reporting the quantity as the maximum profit, or never checking the second derivative.
Drill questions often ask only for the stationary point, so the candidate internalises 'set f′(x)=0 and solve' as the whole method. The distinction between the input (quantity) and the output (profit) collapses, and the second-derivative test is skipped as an extra step.
What to do instead
Use a three-line answer template every time: (1) stationary point x from f′(x)=0; (2) classify with f″(x) - negative means maximum, positive means minimum; (3) substitute x back into the original function to get the maximum or minimum value, and label units. If the domain is restricted (quantity cannot be negative), also evaluate the endpoints. Practise with cost, revenue and profit functions so the substitution step becomes automatic.
In matrices, the candidate multiplies two matrices entry-by-entry (element-wise) instead of using row-by-column dot products, or multiplies A × B when the question needs B × A, ignoring that matrix multiplication is not commutative.
Element-wise operations are what candidates know from addition and scalar multiplication, and the row-column rule is a procedure rather than an intuition, so it decays under exam pressure. Order matters in business models (price vector × quantity vector) but the candidate treats AB and BA as interchangeable.
What to do instead
Before multiplying, write the orders side by side: (m×n)(n×p) = m×p, and if the inner numbers do not match, stop - the product does not exist. Then compute each entry by physically covering the row of A and the column of B with a finger and summing the products aloud. For any applied problem, write the meaning of each matrix dimension first (rows = products, columns = years, etc.) so the correct order is forced by the labels.
When solving a system of linear equations by Cramer's rule or by determinants, the candidate computes the determinant of the coefficient matrix, gets zero, and still divides by it to produce a numerical answer instead of concluding the system is singular (no unique solution).
The formula x = D₁/D is applied mechanically; the candidate treats D as just another number in a fraction and does not connect D = 0 with the geometric and business meaning (parallel constraints, no unique equilibrium). Division by zero is often written as an ordinary result on the script.
What to do instead
Make 'compute D first and record its value before anything else' a fixed first step. If D = 0, write 'system is singular: either no solution or infinitely many' and test consistency by checking whether the other determinants also vanish. Also verify any non-zero answer by substituting back into the original equations - a two-line check that catches most arithmetic slips in determinants.
In permutations and combinations, the candidate cannot decide whether order matters and defaults to nCr for every selection problem, or forgets to divide by the factorial of repeated identical items and by r! for circular arrangements.
The words 'select', 'choose', 'arrange' and 'form a committee' are memorised as triggers, but real questions mix them ('select a president and secretary from 10 people' is a permutation). Circular and repeated-item cases are taught late and rarely revised, so they drop out.
What to do instead
Apply a test question before choosing: 'If I swap two chosen items, do I get a different outcome?' Yes means permutation (nPr), no means combination (nCr). For committees with named posts, multiply combinations or use nPr directly. Keep a short checklist for special cases: circular arrangements of n distinct items = (n−1)!; arrangements of a word with repeated letters = n!/(p!q!…); arrangements with restrictions handled by the gap or complement method. Work 15 mixed problems and label each P or C before computing.
In correlation and regression, the candidate reports a strong correlation as proof that one variable causes the other, and uses the regression of y on x to predict x from y (or reverses the slope and intercept in the equation).
The computational formula for r is drilled without interpretation, so the coefficient becomes a bare number. The two regression lines look symmetric in the formula sheet, and candidates do not notice that y on x minimises vertical deviations and x on y minimises horizontal ones.
What to do instead
Write the interpretation sentence explicitly after every r: direction (positive/negative), strength (using 0.7 and 0.3 as rough bands), and the caveat that correlation does not establish causation - mention a possible lurking variable. For regression, decide from the wording which variable is being predicted, write the line in the form ŷ = a + bx with the predicted variable on the left, and check that the sign of b matches the sign of r. Remember that bxy × byx = r² and use it as a self-check.
When computing the mean, variance or standard deviation from a grouped frequency distribution, the candidate uses class limits instead of midpoints, or applies the wrong denominator (n instead of n−1, or N instead of the sum of frequencies) and mixes the assumed-mean and direct methods.
The table format invites copying the class boundaries straight into the calculation, and the step-deviation method's d = (x − A)/h is remembered without the h² factor in the variance. The n versus n−1 distinction is glossed over because the syllabus context (descriptive versus sample) is not made explicit.
What to do instead
Build a working table with columns: class, midpoint x, frequency f, fx (or d and fd) before any formula. For step deviation, always restore the scale with variance = h² × [Σfd²/N − (Σfd/N)²]. Follow the instruction on the paper about population versus sample; when the data are a sample, use n−1 for variance and standard deviation. Check the mean lies inside the range of midpoints and that standard deviation is non-negative and smaller than the range.
In probability, the candidate adds probabilities for 'and' events and multiplies for 'or' events, and applies the multiplication rule to dependent events without conditioning (drawing without replacement from a lot).
The addition and multiplication rules are learned as paired formulas and get swapped under pressure. The phrase 'without replacement' is easy to skim, and the conditional probability P(B|A) is collapsed to P(B), which understates the answer.
What to do instead
Translate each question into set language first: 'and' means intersection, 'or' means union, and write P(A∪B) = P(A) + P(B) − P(A∩B) rather than a bare sum. For sequential draws, state whether the item is replaced before writing the second probability; without replacement, the denominator and possibly the numerator decrease. Verify that the final probability lies between 0 and 1, and that mutually exclusive events give zero for the intersection.