Author: Trinidad María del Carmen Morán | Email: firstname.lastname@example.org
Trinidad Morán 1°, Diego Golombek 1°, Leandro Casiraghi 1°, Ignacio Spiousas 1°
1° Laboratorio Interdisciplinario del Tiempo (LITERA), Universidad de San Andrés.
The problem of assessing how and when a certain large population sleeps is of great interest to researchers in sleep health. Current methods of tracking sleep are either very precise but costly (i.e. actigraphy through wearables), or easily scalable but of very low precision (i.e. subjective sleep reports). A recently proposed alternative relies on collecting big volumes of longitudinal data from mobile applications and social networks activity that can be analyzed to estimate sleep periods. These sources of information enable researchers to readily access a significant volume of data obtained under natural conditions, and recent studies suggest that, due to their extensive nature, these data facilitate the study of sleep phenomena on a large scale with reasonable reliability. In this pilot study, we propose a model that estimates the sleep events from data of interactions with Android smartphones obtained by their users through Google Takeout, and present preliminary data from a validation study. We also present a set of tools to organize, share and process the estimated sleep parameters using the R language.