← Back to all lessons
Class 12Algebra14:14Published 22 Dec 2024

Class 12 Board MCQ Solutions: 20 Questions

Worked solutions to twenty one-mark MCQ questions from Class 12 board papers, spanning inverse trigonometry, matrices, determinants, vectors, and differential equations.

This revision session walks through twenty multiple-choice questions drawn from Class 12 board question papers, writing the answer first and then the steps. The problems cover inverse trigonometric values, identity and adjoint matrix results, determinants of order three, the inverse of a rotation matrix, dot and cross products of vectors, direction cosines, and the order and degree of differential equations. Each is solved quickly with the key rule stated, so it doubles as a fast formula recap before the exam.

What you'll learn

  • How to evaluate combinations of inverse trigonometric values using their principal ranges
  • How a determinant of the adjoint relates to the determinant of the original matrix of order three
  • How to find the inverse of a two by two rotation matrix by swapping entries and changing signs
  • How to read off the order and degree of a differential equation

Lesson chapters

0:00Inverse trigonometry MCQs
1:22Matrices and determinants
2:46Inverse of a rotation matrix
4:41Degree of a differential equation
5:07Vectors: angle, projection, area
10:45Direction cosines and order and degree

Lesson notes

This lesson works through twenty one-mark MCQ questions from Class 12 board papers, writing each answer first and then the supporting steps. The questions cover inverse trigonometry, matrices and determinants, vectors, and differential equations.

Combining inverse trigonometric values

Evaluate 3sin⁡−112+2cos⁡−1123\sin^{-1}\tfrac{1}{2} + 2\cos^{-1}\tfrac{1}{2}.

Since sin⁡−112=π6\sin^{-1}\tfrac{1}{2} = \tfrac{\pi}{6} and cos⁡−112=π3\cos^{-1}\tfrac{1}{2} = \tfrac{\pi}{3},

3⋅π6+2⋅π3=π2+2π3=3π+4π6=7π6.3\cdot\tfrac{\pi}{6} + 2\cdot\tfrac{\pi}{3} = \tfrac{\pi}{2} + \tfrac{2\pi}{3} = \tfrac{3\pi + 4\pi}{6} = \tfrac{7\pi}{6}.

A nested inverse trig expression

Evaluate tan⁡−1 ⁣(2cos⁡(2sin⁡−112))\tan^{-1}\!\big(2\cos(2\sin^{-1}\tfrac{1}{2})\big).

Here sin⁡−112=π6\sin^{-1}\tfrac{1}{2} = \tfrac{\pi}{6}, so 2sin⁡−112=π32\sin^{-1}\tfrac{1}{2} = \tfrac{\pi}{3} and cos⁡π3=12\cos\tfrac{\pi}{3} = \tfrac{1}{2}.

tan⁡−1 ⁣(2⋅12)=tan⁡−1(1)=π4.\tan^{-1}\!\big(2\cdot\tfrac{1}{2}\big) = \tan^{-1}(1) = \tfrac{\pi}{4}.

Identity matrix times a column

If [100010001][xyz]=[1−10]\begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix}\begin{bmatrix} x \\ y \\ z \end{bmatrix} = \begin{bmatrix} 1 \\ -1 \\ 0 \end{bmatrix}, find x+y+zx + y + z.

The matrix is the identity II, so I[xyz]=[xyz]I\begin{bmatrix} x \\ y \\ z \end{bmatrix} = \begin{bmatrix} x \\ y \\ z \end{bmatrix}. Hence x=1x = 1, y=−1y = -1, z=0z = 0, and

x+y+z=1+(−1)+0=0.x + y + z = 1 + (-1) + 0 = 0.

Determinant of the adjoint

If AA is a square matrix of order 33 with det⁡(adj⁡A)=225\det(\operatorname{adj} A) = 225, find det⁡A\det A.

Using det⁡(adj⁡A)=(det⁡A)n−1\det(\operatorname{adj} A) = (\det A)^{n-1} with n=3n = 3,

225=(det⁡A)2  ⟹  det⁡A=225=15.225 = (\det A)^{2} \implies \det A = \sqrt{225} = 15.

Inverse of a rotation matrix

Find the inverse of A=[cos⁡θsin⁡θ−sin⁡θcos⁡θ]A = \begin{bmatrix} \cos\theta & \sin\theta \\ -\sin\theta & \cos\theta \end{bmatrix}.

Determinant.

det⁡A=cos⁡2θ−(−sin⁡2θ)=cos⁡2θ+sin⁡2θ=1.\det A = \cos^2\theta - (-\sin^2\theta) = \cos^2\theta + \sin^2\theta = 1.

Adjoint. For a 2×22\times 2 matrix, swap the diagonal entries and change the sign of the off-diagonal entries:

adj⁡A=[cos⁡θ−sin⁡θsin⁡θcos⁡θ].\operatorname{adj} A = \begin{bmatrix} \cos\theta & -\sin\theta \\ \sin\theta & \cos\theta \end{bmatrix}.

Inverse. Since A−1=adj⁡Adet⁡AA^{-1} = \dfrac{\operatorname{adj} A}{\det A} and det⁡A=1\det A = 1,

A−1=[cos⁡θ−sin⁡θsin⁡θcos⁡θ].A^{-1} = \begin{bmatrix} \cos\theta & -\sin\theta \\ \sin\theta & \cos\theta \end{bmatrix}.

Degree of a differential equation

For (d2ydx2)2−dydx+1=0\left(\dfrac{d^2y}{dx^2}\right)^2 - \dfrac{dy}{dx} + 1 = 0, the degree is the power of the highest order derivative. The highest order term d2ydx2\dfrac{d^2y}{dx^2} appears to the power 22, so the degree is 22.

Angle from a cross product

Given ∣a⃗∣=3|\vec a| = 3, ∣b⃗∣=23|\vec b| = \tfrac{2}{3}, and a⃗×b⃗\vec a \times \vec b is a unit vector, find the angle between them.

sin⁡θ=∣a⃗×b⃗∣∣a⃗∣ ∣b⃗∣=13⋅23=12  ⟹  θ=π6.\sin\theta = \frac{|\vec a \times \vec b|}{|\vec a|\,|\vec b|} = \frac{1}{3 \cdot \tfrac{2}{3}} = \frac{1}{2} \implies \theta = \frac{\pi}{6}.

Projection of one vector on another

For a⃗=7ı^+ȷ^−4k^\vec a = 7\hat\imath + \hat\jmath - 4\hat k and b⃗=2ı^+6ȷ^+3k^\vec b = 2\hat\imath + 6\hat\jmath + 3\hat k, find the projection of a⃗\vec a on b⃗\vec b.

a⃗⋅b⃗=7(2)+1(6)+(−4)(3)=14+6−12=8.\vec a \cdot \vec b = 7(2) + 1(6) + (-4)(3) = 14 + 6 - 12 = 8.

∣b⃗∣=22+62+32=49=7.|\vec b| = \sqrt{2^2 + 6^2 + 3^2} = \sqrt{49} = 7.

projection=a⃗⋅b⃗∣b⃗∣=87.\text{projection} = \frac{\vec a \cdot \vec b}{|\vec b|} = \frac{8}{7}.

A principal value of inverse cosine

Find cos⁡−1 ⁣(−32)\cos^{-1}\!\left(-\tfrac{\sqrt 3}{2}\right).

The value is negative, so the angle lies in the second quadrant. Writing −32=cos⁡ ⁣(π−π6)-\tfrac{\sqrt 3}{2} = \cos\!\left(\pi - \tfrac{\pi}{6}\right),

cos⁡−1 ⁣(−32)=π−π6=5π6.\cos^{-1}\!\left(-\tfrac{\sqrt 3}{2}\right) = \pi - \tfrac{\pi}{6} = \tfrac{5\pi}{6}.

Solving a determinant equation

Find xx if ∣34−52∣=∣2x4−53∣\begin{vmatrix} 3 & 4 \\ -5 & 2 \end{vmatrix} = \begin{vmatrix} 2x & 4 \\ -5 & 3 \end{vmatrix}.

3(2)−4(−5)=3(2x)−4(−5)3(2) - 4(-5) = 3(2x) - 4(-5)

6+20=6x+20  ⟹  6=6x  ⟹  x=1.6 + 20 = 6x + 20 \implies 6 = 6x \implies x = 1.

Another adjoint determinant

If AA is of order 33 with det⁡(adj⁡A)=64\det(\operatorname{adj} A) = 64, find det⁡A′\det A'.

64=(det⁡A)2  ⟹  det⁡A=±8.64 = (\det A)^{2} \implies \det A = \pm 8.

Since det⁡A′=det⁡A\det A' = \det A for the transpose, det⁡A′=±8\det A' = \pm 8.

Inverse when A2=IA^2 = I

If A2=IA^2 = I, find A−1A^{-1}. Pre-multiplying A2=IA^2 = I by A−1A^{-1} gives A−1A2=A−1IA^{-1}A^2 = A^{-1}I, so A=A−1A = A^{-1}.

Degree of a second differential equation

For y d2ydx2+(dydx)2=x(d2ydx2)2y\,\dfrac{d^2y}{dx^2} + \left(\dfrac{dy}{dx}\right)^2 = x\left(\dfrac{d^2y}{dx^2}\right)^2, the highest order derivative is d2ydx2\dfrac{d^2y}{dx^2} and its highest power is 22, so the degree is 22.

Area of a parallelogram

If a⃗\vec a and b⃗\vec b are two adjacent sides of a parallelogram, its area is ∣a⃗×b⃗∣|\vec a \times \vec b|.

Angle from a dot product

Given ∣a⃗∣=3|\vec a| = 3, ∣b⃗∣=4|\vec b| = 4, and a⃗⋅b⃗=6\vec a \cdot \vec b = 6,

cos⁡θ=a⃗⋅b⃗∣a⃗∣ ∣b⃗∣=63⋅4=12  ⟹  θ=π3.\cos\theta = \frac{\vec a \cdot \vec b}{|\vec a|\,|\vec b|} = \frac{6}{3 \cdot 4} = \frac{1}{2} \implies \theta = \frac{\pi}{3}.

Direction cosines making equal angles

A line through the origin in the first quadrant makes equal angles with the three axes. With l2+m2+n2=1l^2 + m^2 + n^2 = 1 and equal angles, 3cos⁡2α=13\cos^2\alpha = 1, so cos⁡α=13\cos\alpha = \tfrac{1}{\sqrt 3} (positive, since the line is in the first quadrant). The direction cosines are (13,13,13)\left(\tfrac{1}{\sqrt 3}, \tfrac{1}{\sqrt 3}, \tfrac{1}{\sqrt 3}\right).

Perpendicular vectors

Find λ\lambda so that a⃗=2ı^+λȷ^+k^\vec a = 2\hat\imath + \lambda\hat\jmath + \hat k and b⃗=ı^−2ȷ^+3k^\vec b = \hat\imath - 2\hat\jmath + 3\hat k are perpendicular. Then a⃗⋅b⃗=0\vec a \cdot \vec b = 0:

2(1)+λ(−2)+1(3)=0  ⟹  5−2λ=0  ⟹  λ=52.2(1) + \lambda(-2) + 1(3) = 0 \implies 5 - 2\lambda = 0 \implies \lambda = \frac{5}{2}.

Order and degree

For (dydx)2+3y−d2ydx2=0\left(\dfrac{dy}{dx}\right)^2 + 3y - \dfrac{d^2y}{dx^2} = 0, the highest order derivative is d2ydx2\dfrac{d^2y}{dx^2} with power 11, so the order is 22 and the degree is 11.

Equating matrices

If 2[120x]+[y012]=[5618]2\begin{bmatrix} 1 & 2 \\ 0 & x \end{bmatrix} + \begin{bmatrix} y & 0 \\ 1 & 2 \end{bmatrix} = \begin{bmatrix} 5 & 6 \\ 1 & 8 \end{bmatrix}, find xx and yy. Comparing entries, 2+y=52 + y = 5 gives y=3y = 3, and 2x+2=82x + 2 = 8 gives x=3x = 3.

Key takeaways

  • For inverse trig MCQs, convert to principal-range angles first, then simplify.
  • For order nn, det⁡(adj⁡A)=(det⁡A)n−1\det(\operatorname{adj} A) = (\det A)^{n-1}, and det⁡A′=det⁡A\det A' = \det A.
  • The order is the highest derivative present; the degree is the power of that highest order derivative once the equation is polynomial in the derivatives.