Location-based sensors confirm that the public opted to ‘lockdown’ before policy interventions came into force.
A new research study has confirmed that German citizens practiced self-imposed social distancing ahead of policy interventions during the COVID-19 pandemic. Using data from 100 sensors located across 49 metropolitan areas in Germany, researchers found that the numbers of pedestrians had fallen by 85% in 2020 compared with the previous, pre-pandemic year, before formal lockdowns were introduced by the Government.
The study was initially designed to assess how suitable location-based sensors would be for tracking changes in patterns of behaviour in real time, and whether they could be reliably used as an additional source of data by policymakers for crisis management and policy evaluation.
By monitoring the levels of pedestrian activity during the pandemic, researchers could gain an understanding the effectiveness of policy interventions such as lockdown periods, school/university closures, closing workspaces, physical distancing, and closing (non-essential) businesses, in reducing transmission of the virus.
The authors commented that, “Real-time open sensor data in the context of policy interventions (with varying degrees of stringency) are a valuable and underused data source to help tackle future crises.”
There is no doubt that a complex pandemic situation requires fast detection of risk and assessment of the effectiveness of policy interventions. Online surveys and mobile phone data, which have typically been used during the COVID-19 pandemic to evaluate the effects of interventions are less effective for a number of reasons – online surveys don’t provide real time data and are subject to people providing honest answers, and there are limitations to accessing mobile data due to confidentiality and privacy regulations that surround the access and use of sensitive data.
By using data from location sensors that measure pedestrian flows, researchers were able to gain objective measurements of population movement whilst ensuring the privacy of individuals who could not be identified. Analysis of the data produced two very interesting insights.
Firstly, the sensors detected 85% reduced pedestrian flow counts that were self-imposed at the onset of the pandemic, and before socially restrictive policy interventions were introduced.
Secondly, prior to relaxing policy restrictions by the Government during the pandemic, there was already an increase in the numbers of people out on the streets.
The ability to measure human behaviour in real-time is clearly a valuable tool to help solve the social challenges of crises such as the COVID-19 pandemic, both nationally and internationally. The pandemic has revealed new challenges, opportunities, and potential for improvement in crisis management in the future, such as climate change and energy provision, and the ability to track and assess behavioural and social changes is essential for policymakers to tackle future events.
Jonas Klingwort, Statistics Netherlands (CBS), Dataservice, Research & Innovation, Heerlen 6412 EX, The Netherlands
Sofie De Broe, Sciensano, Strategy and external positioning, Brussels 1050, Belgium; Maastricht University, Data Analytics and Digitalisation, Maastricht 6211 LM, The Netherlands
Sven Alexander Brocker, University of Duisburg-Essen, Faculty of Social Sciences, Institute for Sociology, Duisburg 47057, Germany