Opening the book…
A unit vector is a vector whose magnitude is exactly \(1\) (dimensionless). Its only job is to indicate a direction. Unit vectors are typically written with a “hat” rather than an arrow: \(\hat{A}\) is read “\(A\)-hat.”
The unit vector in the direction of a given vector
Given any nonzero vector \(\vec{A}\), the unit vector pointing in the same direction is obtained by dividing \(\vec{A}\) by its own magnitude:
\[\hat{A} = \frac{\vec{A}}{|\vec{A}|} = \frac{\vec{A}}{A}. \tag{3.5}\]
The division of a vector by a scalar (covered in detail in §3.5) just rescales the vector’s length without changing its direction, so \(\hat{A}\) points the same way as \(\vec{A}\) but has length \(1\). Any vector can be written as the product of its magnitude and its direction:
\[\vec{A} = A\,\hat{A}. \tag{3.6}\]
Example 3.3: Unit vector from components
Given \(\vec{A} = \langle -4\,\mathrm{m}, 3\,\mathrm{m}\rangle\), find the unit vector \(\hat{A}\).
Solution
Step 1: Find the magnitude. By equation (3.3),
\[A = \sqrt{(-4)^2 + 3^2}\,\mathrm{m} = \sqrt{16 + 9}\,\mathrm{m} = \sqrt{25}\,\mathrm{m} = 5\,\mathrm{m}.\]
Keep the parentheses around \(-4\) when you square it: \((-4)^2 = (-4)(-4) = +16\). Many calculators read \(-4^2\) without parentheses as \(-(4^2) = -16\), which would give the square root of a negative number here.
Step 2: Divide by the magnitude. Dividing a vector by a scalar divides each component by that scalar (equation (3.11) with \(c = 1/A\)):
\[\hat{A} = \frac{\vec{A}}{A} = \left\langle \frac{-4\,\mathrm{m}}{5\,\mathrm{m}},\; \frac{3\,\mathrm{m}}{5\,\mathrm{m}} \right\rangle = \langle -0.8,\, 0.6 \rangle.\]
Every component shrank by the same factor of \(5\), so the direction is unchanged and the length went from \(5\) to \(1\). The meters cancel in each fraction, so \(\hat{A}\) is dimensionless.
Check. \(\sqrt{(-0.8)^2 + (0.6)^2} = \sqrt{0.64 + 0.36} = \sqrt{1.00} = 1\), so \(\hat{A}\) has magnitude \(1\) as required.
Example 3.4: Unit vector from a compass direction
A force \(\vec{C}\) has magnitude \(450\,\mathrm{N}\) and points \(27^\circ\) north of west. Find \(\hat{C}\).
Solution
Take east as \(+x\) and north as \(+y\). This time we will convert to a standard-position angle: measuring counterclockwise from \(+x\), the direction \(27^\circ\) north of west is at \(\theta = 180^\circ - 27^\circ = 153^\circ\). The components of \(\vec{C}\) are then
\[C_x = (450)\cos 153^\circ \approx -401\,\mathrm{N}, \qquad C_y = (450)\sin 153^\circ \approx +204\,\mathrm{N},\]
and the signs come out right automatically, as promised. Then
\[\hat{C} = \frac{\vec{C}}{C} = \frac{\langle -401, 204\rangle\,\mathrm{N}}{450\,\mathrm{N}} \approx \langle -0.891,\, 0.454\rangle.\]
(Carry the unrounded components into this division; rounding to \(204\,\mathrm{N}\) first would give \(0.453\) instead of \(0.454\). As a rule, round only at the very end of a calculation.) Again, the units of newtons cancel: a unit vector is dimensionless.
As a check, notice that \(\hat{C} = \langle \cos 153^\circ, \sin 153^\circ\rangle\). That is no accident: dividing by the magnitude strips away the “how much” and leaves only the “which way,” and for a two-dimensional vector the direction is completely captured by the cosine and sine of its standard-position angle.
The standard basis vectors
Particularly important are the unit vectors that point along the coordinate axes. We define
\[\hat{\imath} = \langle 1, 0, 0\rangle, \qquad \hat{\jmath} = \langle 0, 1, 0\rangle, \qquad \hat{k} = \langle 0, 0, 1\rangle, \tag{3.7}\]
called the standard basis vectors of three-dimensional space (Figure 3.5). In two dimensions, only \(\hat\imath\) and \(\hat\jmath\) are needed. These three mutually perpendicular unit vectors give us a third, very common way of writing vectors:
\[\vec{A} = A_x\,\hat\imath + A_y\,\hat\jmath + A_z\,\hat{k}. \tag{3.8}\]
So for example \(\vec{A} = \langle 4, -3\rangle\) may equivalently be written as \(\vec{A} = 4\hat\imath - 3\hat\jmath\). The three notations (magnitude and angle, ordered tuple of components, and basis-vector expansion) are completely interchangeable; we choose whichever is most convenient for the problem at hand.
Example 3.5: Three notations for one vector
A displacement vector is \(\vec{C} = \langle 3\,\mathrm{m}, -4\,\mathrm{m}\rangle\). Write it in basis-vector form and in magnitude–angle form.
Solution
The basis-vector form just reads the components off as coefficients of \(\hat\imath\) and \(\hat\jmath\) (equation (3.8)):
\[\vec{C} = (3\hat\imath - 4\hat\jmath)\,\mathrm{m}.\]
The minus sign belongs to the \(y\)-component: \(+3\hat\imath + (-4)\hat\jmath\) is written more compactly as \(3\hat\imath - 4\hat\jmath\). Writing the unit once, outside the parentheses, is cleaner than attaching “m” to each term.
For the magnitude–angle form, use equation (3.3), again keeping the negative component inside parentheses when it is squared:
\[C = \sqrt{3^2 + (-4)^2}\,\mathrm{m} = \sqrt{9 + 16}\,\mathrm{m} = \sqrt{25}\,\mathrm{m} = 5\,\mathrm{m}.\]
For the direction, equation (3.4) gives \(\tan\theta = C_y/C_x = (-4)/3\). Before trusting a calculator, decide the quadrant from the signs: \(C_x > 0\) and \(C_y < 0\), so the vector points right and down, into the fourth quadrant. A calculator gives \(\tan^{-1}(-4/3) \approx -53.1^\circ\), and a negative angle means clockwise from \(+x\), which is indeed a fourth-quadrant direction, so no correction is needed:
\[\vec{C} = 5\,\mathrm{m} \text{ at } 53.1^\circ \text{ below the } +x\text{-axis.}\]
All three descriptions (\(\langle 3, -4\rangle\) m, \((3\hat\imath - 4\hat\jmath)\) m, and “\(5\,\mathrm{m}\) at \(-53.1^\circ\)”) refer to the same arrow. (The standard-position angle would be \(360^\circ - 53.1^\circ = 306.9^\circ\), which names the same direction.)
Example 3.6: A three-dimensional vector
Write \(\vec{D} = \langle 1\,\mathrm{m}, -2\,\mathrm{m}, 3\,\mathrm{m}\rangle\) in basis-vector form and find its magnitude.
Solution
\[\vec{D} = (\hat\imath - 2\hat\jmath + 3\hat{k})\,\mathrm{m},\]
and by the three-dimensional form of equation (3.3),
\[D = \sqrt{1^2 + (-2)^2 + 3^2}\,\mathrm{m} = \sqrt{14}\,\mathrm{m} \approx 3.74\,\mathrm{m}.\]
Note that we do not attempt a magnitude–angle form here. In three dimensions a single angle is not enough to specify a direction, which is exactly why the unit vector \(\hat{D} = \vec{D}/D\) is the tool of choice.
Checkpoint 3.2
Is \(\hat\imath + \hat\jmath\) a unit vector? Justify your answer with a calculation.
Answers to Checkpoints
Checkpoint 3.2. No. Its magnitude is \(|\hat\imath + \hat\jmath| = \sqrt{1^2 + 1^2} = \sqrt{2} \approx 1.41\), not \(1\). Adding two unit vectors does not, in general, give a unit vector. The unit vector pointing in the same direction is \((\hat\imath + \hat\jmath)/\sqrt{2}\).
This page titled 3.3: Unit Vectors was last modified on Thu, 12 Nov 2026 04:26:43 GMT and is shared under a CC BY 4.0 license and was authored, remixed, and/or curated by Michael Ragsdale.
The book could not be loaded just now.