In this paper, we present the results of a community-based research project directed at the Latin American community of Turin, which is mainly composed of first-generation Peruvian migrants.

Within the project “DiverSITa: Diversity in spoken Italian”, oral data from four different communities (Peruvian, Chinese, Moroccan, and Romanian) in Turin are currently being collected to build a corpus of spoken Italian that includes the varieties of multilingual individuals. Since the data collection involves interviews and spontaneous conversations, the deep engagement of the researchers with the community is highly beneficial. Getting to know the community from the inside and establishing a strong relationship with the people allows for the collection of better-quality data and the acquisition of ethnographic information that enriches the researcher’s knowledge during the data analysis. Hence, we adopted a community-based approach to data collection (Rice 2010; Bischoff & Jany 2018; Ganassin & Holmes 2020) by listening to the needs of the interlocutors and combining them with our specific research objectives.

The Peruvian case-study represents a good example of such a methodology in that we carried out a collaborative activity with a local association that provides assistance to Latin American people who have recently moved to Italy. We organized a cooperative five-meeting project with individuals from 14 to 20 years old, whose goal was to produce a mockup tourist guide of places and activities that best represent their community in Turin. This enabled us to both respond to the association’s educational necessities and to collect data for our corpus by recording the interactions occurring during the meetings and using the interviews recorded during the task.

In this presentation, we will present the corpus under development, describe the activity organized in the Peruvian community to collect our data, and provide evidence of the multilingual practices observed in the community.