This study evaluated the ability of observers to assess the body condition of Holstein cows using a body condition score (BCS). Four observers independently assigned a BCS (five-point scale with .25 increments) and described the appearance of seven body regions in 225 Holstein cows. The areas described included the thurl region, ischial and ileal tuberosities, ilio-sacral and ischial-coccygeal ligaments, transverse processes of the lumbar vertebrae, and spinous processes of the lumbar vertebrae. An absolute BCS was determined for each cow based on the modal BCS of all observers. If no modal BCS existed, the mean score was used. Statistical analysis of principal components was used to examine the relationship between body region descriptions and the absolute BCS. Four principal component vectors explained 83.6% of the variation in the body region correlation matrix. The first principal component vector accounted for 55% of the variation and was uniformly correlated with all body regions. Analysis of variance of the first principal component vector as the dependent variable and absolute BCS as the class variable indicated that body condition could be separated into .25 units between 2.5 and 4.0. Below 2.5 and above 4.0, body condition could only be separated by .5 units. Distinct changes in specific body regions were associated with changes in the absolute BCS. Observers agreed with the absolute score 58.1% of the time, deviating by .25 units 32.6% of the time. A BCS can be given to a cow based on principal descriptors of specific body regions between 2.5 and 4.0 by .25 units. The study found that descriptions of body regions were highly correlated across all absolute BCS scores. Principal component analysis was used to identify principal descriptors that could be used to describe a specific BCS. The results showed that the first principal component vector was the most important descriptor, as it accounted for 55% of the variation in the data. The study concluded that descriptions of seven body regions were sufficient to categorize cows by .25-unit increments from BCS 2.25 to 4.00. The BCS could be simplified using principal descriptors for unique changes at each BCS class. Observers agreed 58 to 67% of the time when independently scoring cattle body condition; 21 to 34% of the time, observers differed by ±0.25 units.This study evaluated the ability of observers to assess the body condition of Holstein cows using a body condition score (BCS). Four observers independently assigned a BCS (five-point scale with .25 increments) and described the appearance of seven body regions in 225 Holstein cows. The areas described included the thurl region, ischial and ileal tuberosities, ilio-sacral and ischial-coccygeal ligaments, transverse processes of the lumbar vertebrae, and spinous processes of the lumbar vertebrae. An absolute BCS was determined for each cow based on the modal BCS of all observers. If no modal BCS existed, the mean score was used. Statistical analysis of principal components was used to examine the relationship between body region descriptions and the absolute BCS. Four principal component vectors explained 83.6% of the variation in the body region correlation matrix. The first principal component vector accounted for 55% of the variation and was uniformly correlated with all body regions. Analysis of variance of the first principal component vector as the dependent variable and absolute BCS as the class variable indicated that body condition could be separated into .25 units between 2.5 and 4.0. Below 2.5 and above 4.0, body condition could only be separated by .5 units. Distinct changes in specific body regions were associated with changes in the absolute BCS. Observers agreed with the absolute score 58.1% of the time, deviating by .25 units 32.6% of the time. A BCS can be given to a cow based on principal descriptors of specific body regions between 2.5 and 4.0 by .25 units. The study found that descriptions of body regions were highly correlated across all absolute BCS scores. Principal component analysis was used to identify principal descriptors that could be used to describe a specific BCS. The results showed that the first principal component vector was the most important descriptor, as it accounted for 55% of the variation in the data. The study concluded that descriptions of seven body regions were sufficient to categorize cows by .25-unit increments from BCS 2.25 to 4.00. The BCS could be simplified using principal descriptors for unique changes at each BCS class. Observers agreed 58 to 67% of the time when independently scoring cattle body condition; 21 to 34% of the time, observers differed by ±0.25 units.