Purpose: This study was performed to evaluate the effect of changing the orientation of a reconstructed image on the accuracy of linear measurements using cone-beam computed tomography (CBCT). Materials and Methods: Forty-two titanium pins were inserted in seven dry sheep mandibles. The length of these pins was measured using a digital caliper with readability of 0.01 mm. Mandibles were radiographed using a CBCT device. When the CBCT images were reconstructed, the orientation of slices was adjusted to parallel (i.e., $0^{circ}$), $+10^{circ}$, $+12^{circ}$, $-12^{circ}$, and $-10^{circ}$ with respect to the occlusal plane. The length of the pins was measured by three radiologists, and the accuracy of these measurements was reported using descriptive statistics and one-way analysis of variance (ANOVA); p<0.05 was considered statistically significant. Results: The differences in radiographic measurements ranged from -0.64 to +0.06 at the orientation of $-12^{circ}$, -0.66 to -0.11 at $-10^{circ}$, -0.51 to +0.19 at $0^{circ}$, -0.64 to +0.08 at $+10^{circ}$, and -0.64 to +0.1 at $+12^{circ}$. The mean absolute values of the errors were greater at negative orientations than at the parallel position or at positive orientations. The observers underestimated most of the variables by 0.5-0.1 mm (83.6%). In the second set of observations, the reproducibility at all orientations was greater than 0.9. Conclusion: Changing the slice orientation in the range of $-12^{circ}$ to $+12^{circ}$ reduced the accuracy of linear measurements obtained using CBCT. However, the error value was smaller than 0.5 mm and was, therefore, clinically acceptable.