The main objetive of this studt was to evaluate a statistical methodology for the characterization of guinea pig farms (Cavia porcellus). The metodology is based on the teory of principal components, wich analyses the structural dependency of sets of multivariate data. In order to properly evaluate the procedure, information for surveys conducted en Cajamarca, Peru in 1987 y 1988, was used. The analyses were carried out with the SPSS/PC package. The results of the study proved that the original nine relevant variables could be sustituted by four principal componets which explained 65% of the total variation as compared to the original variables. The first principal component expresses the livestock prooductivity and farm income derived from the monthly sale of guinea pigs. The second component reflects the behaviour of the head of the family according to age and schooling. The third comp...