Additive manufacturing (AM) brings significant freedom in design, yet it can get hard to produce the same part at identical dimensional tolerances; this is also known as the reproducibility problem. Reproducibility, the ability to produce the same part under similar conditions, is one of the major challenges in AM as reproducibility plays an important role in the replacement of worn-out or damaged parts in an assembly. The objective of this paper is to identify the impacts of two most common factors (i.e., layer thickness and printing speed) on the dimensional accuracy of additively manufactured parts through a designed experiment. A full-factorial experimental design involving these factors at three levels is implemented to investigate them. We printed a dog bone testing specimen by using Poly Lactic Acid (PLA) polymer and Fused Filament Fabrication (FFF) technology. The dimensional properties of the parts are then measured to statistically compare the variability in each level to derive significant factors and their levels. The results show that printing speed has a significant effect on deviation in length but has no effect on deviation in height. Also, layer thickness and interaction between layer thickness and printing speed can cause significant variation in height.