No abstract

Humanitarian crises, such as the 2014 West Africa Ebola epidemic, challenge information management and thereby threaten the digital resilience of the responding organizations. Crisis information management (CIM) is characterised by the urgency to respond despite the uncertainty of the situation. Coupled with high stakes, limited resources and a high cognitive load, crises are prone to induce biases in the data and the cognitive processes of analysts and decision-makers. When biases remain undetected and untreated in CIM, they may lead to decisions based on biased information, increasing the risk of an inefficient response. Literature suggests that crisis response needs to address the initial uncertainty and possible biases by adapting to new and better information as it becomes available. However, we know little about whether adaptive approaches mitigate the interplay of data and cognitive biases. We investigated this question in an exploratory, three-stage experiment on epidemic response. Our participants were experienced practitioners in the fields of crisis decision-making and information analysis. We found that analysts fail to successfully debias data, even when biases are detected, and that this failure can be attributed to undervaluing debiasing efforts in favor of rapid results. This failure leads to the development of biased information products that are conveyed to decision-makers, who consequently make decisions based on biased information. Confirmation bias reinforces the reliance on conclusions reached with biased data, leading to a vicious cycle, in which biased assumptions remain uncorrected. We suggest mindful debiasing as a possible counter-strategy against these bias effects in CIM.

Social media data is heavily used to analyze and evaluate situations in times of disasters, and derive decisions for action from it. In these critical situations, it is not surprising that privacy is often considered a secondary problem. In order to prevent subsequent abuse, theft or public exposure of collected datasets, however, protecting the privacy of social media users is crucial. Avoiding unnecessary data retention is an important question that is currently largely unsolved. There are a number of technical approaches available, but their deployment in disaster management is either impractical or requires special adaption, limiting its utility. In this case study, we explore the deployment of a cardinality estimation algorithm called HyperLogLog into disaster management processes. It is particularly suited for this field, because it allows to stream data in a format that cannot be used for purposes other than the originally intended. We develop and conduct a focus group discussion with teams of social media analysts. We identify challenges and opportunities of working with such a privacy-enhanced social media data format and compare the process with conventional techniques. Our findings show that, with the exception of training scenarios, deploying HyperLogLog in the data acquisition process will not distract the data analysis process. Instead, several benefits, such as improved working with huge datasets, may contribute to a more widespread use and adoption of the presented technique, which provides a basis for a better integration of privacy considerations in disaster management.

Virtual Operations Support Teams are groups of institutionalized digital volunteers in the field of disaster management who conduct Social Media Analytics tasks for decision-makers in Emergency Operation Centers (EOCs) during hazard situations such as floods. Through interagency integration into EOC structures, the volunteers provide analytical support using advanced tools and monitoring various social media platforms. The goal of VOSTs is to increase decision-makers’ situational awareness through need-oriented analysis and to improve decision-making by providing actionable information in a time-critical work context. In this case study, the data collected during the 2021 flood in Wuppertal, Germany by 22 VOST analysts was processed and analyzed. It was found that information from eight social media platforms could be classified into 23 distinct categories. The analysts’ prioritizations indicate differences in the formats of information and platforms. Disaster-related posts that pose a threat to the affected population’s health and safety (e.g., requests for help or false information) were more commonly prioritized than other posts. Image-heavy content was also rated higher than text-heavy data. A subsequent survey of EOC decision-makers examined the impact of VOST information on situational awareness during this flood. It also asked how actionable information impacted decisions. We found that VOST information contributes to expanded situational awareness of decision-makers and ensures people-centered risk and crisis communication. Based on the results from this case study, we discuss the need for future research in the area of integrating VOST analysts in decision-making processes in the field of time-critical disaster management.

No abstract