The aim of this study was to evaluate how tray shape (half or full arch) and impression procedure affect the precision of impressions. Half arch and full arch trays in metal and plastic were utilized in combination with a two stage impression technique. For each technique-tray combination, 10 impressions were made of a master cast with two steel rod representing the mandibular left first premolar and first molar, where a standard steel rod was placed. Each steel abutment had four marks, which served as a reference point. With a universal measuring microscope, the x-, y-, and zcoordinates were computed for each mark on the master cast and impressions. The span between marks and the reference point on the impressions were calculated and compared with those of the master cast. All techniques employed with the full arch metal trays and the plastic trays had distances that were not significantly different from the master cast (P < 0.05), while for the half arch plastic trays, it was different from the distance in master cast but the distances were not significantly different from the master cast (P < 0.05). Half arch plastic trays produced correct impressions. When full arch were employed, the full arch metal tray impressions were dimensionally better than full arch plastic impressions.