Is Crowdsourcing a Reliable Method for Mass Data Acquisition? The Case of COVID-19 Spread in Greece During Spring 2020

Varvara Antoniou,Emmanuel Vassilakis,Maria Hatzaki

doi:10.3390/ijgi9100605

Varvara Antoniou, Emmanuel Vassilakis + Show 1 more

Open Access

https://doi.org/10.3390/ijgi9100605

Copy DOI

Abstract

We present a GIS-based crowdsourcing application that was launched soon after the first COVID-19 cases had been recorded in Greece, motivated by the need for fast, location-wise data acquisition regarding COVID-19 disease spread during spring 2020, due to limited testing. A single question was posted through a web App, to which the anonymous participants subjectively answered whether or not they had experienced any COVID-19 disease symptoms. Our main goal was to locate geographical areas with increased number of people feeling the symptoms and to determine any temporal changes in the statistics of the survey entries. It was found that the application was rapidly disseminated to the entire Greek territory via social media, having, thus, a great public reception. The higher percentages of participants experiencing symptoms coincided geographically with the highly populated urban areas, having also increased numbers of confirmed cases, while temporal variations were detected that accorded with the restrictions of activities. This application demonstrates that health systems can use crowdsourcing applications that assure anonymity, as an alternative to tracing apps, to identify possible hot spots and to reach and warn the public within a short time interval, increasing at the same time their situational awareness. However, a continuous reminder for participation should be scheduled.

Highlights

Several applications have been introduced in recent years based on the usability of Geographic Information Systems (GIS) in synergy with a wide range of spatial tools that seem to provide solutions for public concerns, such as public health data analysis
A single closed-ended question was posted via a web app formulated as “Have you experienced COVID-19 symptoms?”, with the participants having to select among three predefined options, i.e., no symptoms (No), Not Sure and Yes
From the 212,715 entries, 207,762 were onshore and 4953 were offshore. The latter were excluded from the statistical approach, many of them might have been responses from people either travelling by boat or some responders were using low-precision GPS mobile devices

Summary

Introduction

Several applications have been introduced in recent years based on the usability of Geographic Information Systems (GIS) in synergy with a wide range of spatial tools that seem to provide solutions for public concerns, such as public health data analysis. Referenced information regarding the location of cases, potential sources and likely directions of epidemic spreading in order to define dispersion patterns and rates, can be provided efficiently via digital mapping [2]. It has been in the last few decades or so that GIS platforms became more user friendly, gradually migrating from desktop to the web, making it much more efficient and faster in reaching out answers or information products. The development of an increasingly high number of geoinformation algorithms and, especially, the establishment of online software operation through high-speed network infrastructures, gives the opportunity of real-time big data acquisition and processing, breaking new ground for innovative applications

Objectives

Methods

Results

Conclusion