When the analysis of digital data reaches its limits, methods that focus on personal observation can be useful. In situations such as coronavirus pandemics, information that is difficult to obtain from digital trace data can be extracted by a method called human social perception. Professor Frauke Kreuter of Ludwig Maximilian University (LMU) in Munich is now using this method in conjunction with the global Covid Trends & Impact Survey to predict the course of a pandemic.
While today’s social science researchers have access to historically unparalleled amounts of data, many aspects of modern social development have proven difficult to predict. National elections and the ongoing coronavirus pandemic are highly visible examples of current systems being challenged to provide accurate predictions.
“Don’t focus too much on the analysis of digital trace data. It’s a mistake to ignore the fact that people have sensory abilities. These are difficult to capture with data from traces of behavior. It can be particularly useful in the field, “says Frauke Kreuter, a professor of statistics and data science in social and human sciences at LMU. Many social phenomena develop very rapidly and often in unexpected directions. This makes it impossible to collect enough up-to-date passive data to allow researchers to track the trajectory of such trends.
The individual as a human sensor plays a central role
In a research paper just published in the journal Nature, Kreuter and her international co-authors say that interviews with individual informants still play an important role, even in an era when big data produces the biggest headlines. Is shown. “We never lose sight of the fact that research can serve as a very valuable source of supplementary information,” she says.
Individual interviewees can ask not only about their situation, but also about themselves and the local environment, but the latter possibility has long been overlooked. ” Now, in this current situation, it is useful to take advantage of the fact that an individual can be seen as a potentially useful “social sensor” about what is happening around him.
Development of coronavirus pandemic as a test case
As a statistician, Frauke Kreuter chose to adopt this approach with his colleagues in the current global Covid Trends & Impact Survey, which seeks to predict the pandemic trajectory of the colonovirus. The survey began in April 2020 and has interviewed more than 55 million people worldwide. One of the questions asks the interviewee if he or she personally knows someone in the local community with COVID-19 symptoms.
This is especially noticeable because the answer to this question turned out to be a powerful predictor of further development of the pandemic. Preliminary results of the study are already displayed as a preprint of MedRxiv. “A key prerequisite for using an individual as a sensor in a social science study is that the selected informant must compose a controlled representative sample of the population of interest.”
social experiment by Livio Acerbo #greengroundit #thisisnotapost #thisisart