The positional distributions of fatty acids (FAs) in milk fat containing short- and medium-chain FAs were analyzed by sn-1(3)-selective transesterification of triacylglycerols (TAGs) with ethanol using immobilized Candida antarctica lipase B (CALB), in a collaborative study conducted by 10 laboratories. The mean C4:0, C6:0, and C8:0 FA contents, when analyzed as propyl esters (PEs) using gas chromatography (GC) with a DB-23 capillary column, were found to be 3.0, 2.0, and, 1.3 area%, respectively. Their reproducibility standard deviations were 0.33, 0.18, and 0.19, respectively. The mean C4:0, C6:0, and C8:0 contents at the sn-2 position were 0.3, 0.4, and 1.0 area%, respectively. Their reproducibility standard deviations were 0.17, 0.11, and 0.19, respectively. The reproducibility standard deviations of C4:0, C6:0, and C8:0 FAs at the sn-2 position were either the same as or smaller than those for milk fat, although the FA contents at the sn-2 position were smaller than those in the milk fat. Therefore, it was concluded that the CALB method for estimating the regiospecific distribution is applicable to TAGs containing short- and medium-chain FAs. When estimating the short-chain (SC) FA contents in fats and oils by GC, it is better to analyze SCFAs as PEs or butyl esters, and not as methyl esters, in order to prevent loss of SCFAs during the experimental procedure because of their volatility and water solubility. This study also revealed that the stationary phase of the GC capillary column affected the flame ionization detector (FID) response of SCFAs. The theoretical FID correction factor (MWFA / active carbon number / atomic weight of carbon) fitted well with the actual FID responses of C4:0-C12:0 FAs when they were analyzed as PEs using a DB-23 column; however, this was not the case when the GC analysis was performed using wax-type columns.