The objective of this study was to estimate observer accuracy and repeatability of body condition scoring sows when scorers have different levels of prior experience. Three groups of participants (n = 10) for this study were identified as having no (NE, n = 3), some (SE, n = 4), and extensive (EE, n = 3) prior experience evaluating conformation or body condition in livestock species. Two persons having extensive prior experience with body condition scoring served as instructors (TR) during the training sessions. Twenty-five of a total 150 sows were utilized in the participant training session, and the remaining sows (n = 125) were utilized during the independent scoring process. Sows utilized in the scoring process were objectively categorized into a 5- and 9-point body condition score (BCS5 and BCS9, respectively) using last rib backfat estimates. Participants were in poor agreement with BCSbackfat as overall Kappa values were 0.23 on the BCS5 and 0.13 on the BCS9 scales. While the trainers consistently averaged the largest measures of intra- and interobserver agreement with BCSbackfat, other participants primarily in the EE and SE groups achieved similar levels of agreement. Participant BCS5 and BCS9 deviation evaluations from BCSbackfat, revealed a tendency for participants to overestimate BCS in some sows and underestimate BCS in others. While the trainers consistently averaged the largest measures of intra- and interobserver agreement with BCSbackfat, other participants primarily in the EE and SE groups achieved similar levels of agreement.