Types of vectors

A vector is not just "an arrow with a length" — different arrows earn different names depending on their length, their direction, and how they relate to other arrows in the same problem. CBSE Class 12 Vector Algebra leans on five such names constantly: the zero vector, the unit vector, equal vectors, parallel (collinear) vectors, and the negative of a vector. Once magnitude, direction, and position vectors are settled, questions stop saying "vector" in general and start saying "find a unit vector," "show these are collinear," or "prove a⃗=b⃗\vec a = \vec b" — and each phrase is really shorthand for a precise algebraic test. This article catalogues those tests so that the rest of the chapter — addition laws, the section formula, dot and cross products — can be read fluently without re-deriving vocabulary every time.

The core insight

Vectors are classified by magnitude and direction into standard types — zero, unit, equal, parallel/collinear, and negative — each with its own precise algebraic test, and recognising which type a problem names is usually the first step to solving it.

The core idea

Zero (null) vector, 0⃗\vec 0: magnitude 00, no specific direction. It shows up whenever the initial and terminal points coincide, e.g. AA→=0⃗\overrightarrow{AA} = \vec 0. It is still a legitimate vector — a⃗+(−a⃗)=0⃗\vec a + (-\vec a) = \vec 0 is a genuine vector equation, not an exception.

Unit vector: magnitude exactly 11. For any nonzero vector a⃗\vec a, the unit vector along it is

a^=a⃗∣a⃗∣,\hat a = \frac{\vec a}{|\vec a|},

where: a⃗\vec a is the given vector and ∣a⃗∣|\vec a| is its magnitude. Dividing a vector by its own length always produces a vector of length 11 pointing the same way — this is called normalising the vector. The axis vectors i^,j^,k^\hat i, \hat j, \hat k are the most familiar unit vectors, but any direction has one.

Equal vectors: a⃗=b⃗\vec a = \vec b means same magnitude and same direction, regardless of where each one is drawn from. In components this is three separate equalities at once — the xx, yy, and zz parts must all match.

Parallel (collinear) vectors: nonzero a⃗\vec a and b⃗\vec b are parallel if one is a scalar multiple of the other,

b⃗=λa⃗for some scalar λ≠0,\vec b = \lambda \vec a \quad \text{for some scalar } \lambda \neq 0,

where: λ\lambda is the scale factor linking the two vectors. If λ>0\lambda > 0 the vectors are like vectors (same way); if λ<0\lambda < 0 they are unlike vectors (opposite ways) — both count as parallel. Three or more points are collinear when the vectors joining them are all parallel to a common line.

Negative of a vector: −a⃗-\vec a has the same magnitude as a⃗\vec a but points the opposite way, so BA→=−AB→\overrightarrow{BA} = -\overrightarrow{AB}.

Coinitial vectors: vectors sharing the same starting point — a labelling convenience used when several vectors are drawn from one origin.

Figure
Five labelled arrows on a grid showing a zero vector as a dot a unit length arrow two equal parallel arrows pointing the same way and one arrow reversed to show a negative vector
Types of vectors at a glance zero vector 0 unit vector, length 1 equal vectors, same length and way a vector and its negative, reversed parallel vectors, lambda negative, opposite ways

Why the classification matters. Naming a vector's type turns a wordy instruction into an algebraic recipe: "unit vector along a⃗\vec a" means compute a^=a⃗/∣a⃗∣\hat a = \vec a/|\vec a|; "collinear" means test proportionality of components; "a⃗=b⃗\vec a = \vec b" means three separate component equations must all hold. Spotting the type is usually the first move in solving the problem.

Derivation AFinding the unit vector along any given vector

The goal is to show, starting only from the definition of a unit vector as one of magnitude exactly 11, why dividing any nonzero vector by its own magnitude always produces such a unit vector pointing in the same direction — the single fact that turns "find a unit vector along a⃗\vec a" into a two-line calculation every time.

Step 1. Write a⃗\vec a in components, a⃗=a1i^+a2j^+a3k^\vec a = a_1\hat i + a_2 \hat j + a_3 \hat k, and recall that its magnitude is ∣a⃗∣=a12+a22+a32|\vec a| = \sqrt{a_1^2+a_2^2+a_3^2}, a fixed positive number as long as a⃗≠0⃗\vec a \neq \vec 0.

Step 2. Define a^=a⃗∣a⃗∣=a1∣a⃗∣i^+a2∣a⃗∣j^+a3∣a⃗∣k^\hat a = \dfrac{\vec a}{|\vec a|} = \dfrac{a_1}{|\vec a|}\hat i + \dfrac{a_2}{|\vec a|}\hat j + \dfrac{a_3}{|\vec a|}\hat k. Dividing every component by the same positive scalar ∣a⃗∣|\vec a| does not change the direction the vector points — it only rescales its length.

Step 3. Check the magnitude of a^\hat a: ∣a^∣=(a1∣a⃗∣)2+(a2∣a⃗∣)2+(a3∣a⃗∣)2=1∣a⃗∣a12+a22+a32=∣a⃗∣∣a⃗∣=1|\hat a| = \sqrt{\left(\dfrac{a_1}{|\vec a|}\right)^2 + \left(\dfrac{a_2}{|\vec a|}\right)^2 + \left(\dfrac{a_3}{|\vec a|}\right)^2} = \dfrac{1}{|\vec a|}\sqrt{a_1^2+a_2^2+a_3^2} = \dfrac{|\vec a|}{|\vec a|} = 1. So a^\hat a has magnitude exactly 11 by construction, for any nonzero a⃗\vec a.

Step 4. Because dividing by a positive scalar preserves direction (Step 2) and always yields magnitude 11 (Step 3), a^\hat a is a genuine unit vector along a⃗\vec a — this is why a^=a⃗/∣a⃗∣\hat a = \vec a/|\vec a| works universally and needs no case-by-case justification.

Worked example

Find the unit vector in the direction of a⃗=3i^−4j^+12k^\vec a = 3\hat i - 4\hat j + 12\hat k, and verify it against a second vector to check the collinearity test as well.

Step 1 — magnitude. ∣a⃗∣=32+(−4)2+122=9+16+144=169=13.|\vec a| = \sqrt{3^2+(-4)^2+12^2} = \sqrt{9+16+144} = \sqrt{169} = 13.

Step 2 — divide by magnitude. a^=113(3i^−4j^+12k^)=313i^−413j^+1213k^.\hat a = \dfrac{1}{13}\left(3\hat i - 4\hat j + 12\hat k\right) = \dfrac{3}{13}\hat i - \dfrac{4}{13}\hat j + \dfrac{12}{13}\hat k.

Step 3 — check. ∣a^∣=(313)2+(−413)2+(1213)2=169169=1|\hat a| = \sqrt{\left(\tfrac{3}{13}\right)^2+\left(\tfrac{-4}{13}\right)^2+\left(\tfrac{12}{13}\right)^2} = \sqrt{\dfrac{169}{169}} = 1. Confirmed a unit vector. Answer: a^=313i^−413j^+1213k^\hat a = \dfrac{3}{13}\hat i - \dfrac{4}{13}\hat j + \dfrac{12}{13}\hat k.

Competitive stretch

Are p⃗=2i^−j^+4k^\vec p = 2\hat i - \hat j + 4\hat k and q⃗=−6i^+3j^−12k^\vec q = -6\hat i + 3\hat j - 12\hat k collinear, and if so, same or opposite direction?

Step 1. Compare component ratios: −62=−3\dfrac{-6}{2} = -3, 3−1=−3\dfrac{3}{-1} = -3, −124=−3\dfrac{-12}{4} = -3. All equal, so q⃗=−3p⃗\vec q = -3\vec p.

Step 2. Since λ=−3≠0\lambda = -3 \neq 0, the vectors are collinear; since λ<0\lambda < 0, they point in opposite directions — q⃗\vec q is three times as long as p⃗\vec p but reversed.

Common traps

  • Calling equal-length vectors "equal vectors." Equality needs the same direction too — two vectors of the same length pointing differently are not equal.
  • Treating 0⃗\vec 0 as "not a real vector" because it has no direction. It is a legitimate vector that satisfies real equations, such as a⃗+(−a⃗)=0⃗\vec a + (-\vec a) = \vec 0.
  • Ignoring the sign of λ\lambda. Parallel vectors can point the same way (λ>0\lambda>0) or opposite ways (λ<0\lambda<0) — both are "parallel/collinear," but direction differs.
  • Assuming only i^,j^,k^\hat i, \hat j, \hat k are unit vectors. Any vector of magnitude 11, in any direction, qualifies; the axis vectors are just the most common examples.

Ask the tutor: unsure how to use collinear vectors to prove three points lie on a straight line, or want more practice finding unit vectors quickly? Ask the AI tutor for a guided example.

Why does dividing a vector by its magnitude always give length exactly 1?

Because magnitude scales linearly with every component. If you divide each component of a⃗\vec a by the positive number ∣a⃗∣|\vec a|, the new magnitude is ∣a⃗∣/∣a⃗∣=1|\vec a|/|\vec a| = 1 by the definition of magnitude itself — the square root of the sum of squared components shrinks by exactly the same factor you divided by. This works for any nonzero vector, which is why a^=a⃗/∣a⃗∣\hat a = \vec a/|\vec a| is a universal recipe, not something you re-derive each time.

Is the zero vector parallel to every other vector?

This is a genuine subtlety, and CBSE convention treats 0⃗\vec 0 as a special case rather than forcing it through the b⃗=λa⃗\vec b = \lambda\vec a test, since 0⃗=0⋅a⃗\vec 0 = 0\cdot\vec a would require λ=0\lambda = 0, which the parallel-vector definition explicitly excludes (λ≠0\lambda \neq 0). In practice, questions about parallel or collinear vectors always assume both vectors are nonzero, so this edge case rarely needs to be argued in an exam answer — just note that the standard test applies to nonzero vectors.

How is showing three points are collinear different from showing two vectors are parallel?

They use the same core test but at different starting points. For two vectors, you check b⃗=λa⃗\vec b = \lambda \vec a directly. For three points AA, BB, CC, you first form two vectors between them, say AB→\overrightarrow{AB} and AC→\overrightarrow{AC}, and then check whether AC→=λ AB→\overrightarrow{AC} = \lambda\, \overrightarrow{AB} for some scalar λ\lambda. If it holds, the vectors are parallel and they share the point AA, which forces all three points onto one straight line — parallelism alone is not enough for points, you also need a shared point.

Before this

Where this leads