Grammatical agreement, a property of inflectional languages such as Spanish, consists of a formal association of different syntactic constituents. Even though it is a categorical morphosyntactic phenomenon, in Peruvian Spanish non-agreement occurs in different communicative contexts, due to certain social factors that will be analyzed in this paper. Based on a variationist perspective, we intend to identify and analyze non-concordant (NC) utterances extracted from the PRESEEA-LIMA corpus (directed by the author) in several linguistic contexts related to the constituents of noun phrases and verb phrases. An overview of the quantitative results will be provided in this paper. Moreover, we will approach data from both a heuristic and a hermeneutic point of view, through a reinterpretation of the role of quantitative value in the determination of linguistic variation.