On the validity of crowdsourced data

This chapter demonstrates the validity of crowdsourced data by comparing the crowdsourced data from the VinKo project with traditionally collected data from the AThEME project. Both datasets target non-standard language varieties of the South Tyrol, Trentino, and Veneto regions in north-eastern Italy. Three different morphosyntactic phenomena are discussed, each relating to a particular language variety, providing evidence that the crowdsourced data is of comparable quality to the traditionally gathered data and has the added advantage of yielding a larger overall dataset covering a denser location network.