I was working on the following question about directional derivatives.
The directional derivative of a function $\phi(x, y, z)$ at the point $(2, 0, 3)$ in the direction towards $(3, -2, 3)$ is $1.789$, in the direction towards $(2, 4, 4)$ it is $0.243$, whilst in the direction towards $(4, -1, 2)$ it is zero. By showing that the three first order partial derivatives of $\phi$ with respect to $x,$ $y$ and $z$ are $2.0$, $-1.0$ and $5.0$ respectively, verify that the value of the directional derivative of $\phi$ at $(2, 0, 3)$ in the direction towards $(0, 2, 14)$ is $4.314$.
Now I used the following approach. We know that the directional derivative of a scalar field $\phi$ in the direction of a unit vector $\hat{\boldsymbol{\mathrm{s}}}$ is given by
$$\frac{\partial \phi}{\partial s}=\hat{\boldsymbol{\mathrm{s}}}\cdot\nabla\phi$$
Let the first direction given be $\boldsymbol{\mathrm{s}}_1=(3,-2,3)$. Then $\hat{\boldsymbol{\mathrm{s}}}=\frac{1}{\sqrt {22}}(3,-2,3)$. Substituting into the result above gives
$$\frac{\partial \phi}{\partial s_1}=\frac{1}{\sqrt{22}}\begin{pmatrix}3\\-2\\3\end{pmatrix}\cdot\nabla\phi=\frac{3}{\sqrt{22}}\frac{\partial \phi}{\partial x}-\frac{2}{\sqrt{22}}\frac{\partial \phi}{\partial y}+\frac{3}{\sqrt{22}}\frac{\partial \phi}{\partial z}=1.789$$
However if we substitute the given results $\partial\phi/\partial x=2$, $\partial\phi/\partial y=-1$ and $\partial\phi/\partial z=5$, equality does not hold; nor does it hold if we repeat the procedure for the other two directional derivatives.
Is there a flaw in my approach? Perhaps the results in the question are mistaken? I'd appreciate any feedback.