The complex and uncertain environment of the humanitarian response to crises can lead to data bias, which can affect decision-making. Evidence of data bias in crisis information management (CIM) remains scattered despite its potentially significant impact on crisis response. To understand what biases emerge in complex crises and how they affect CIM, we conducted a combined interview and document analysis study. Focusing on the largest humanitarian crisis in the world, i.e., the conflict in Yemen, we conducted 25 interviews with managers and analysts of response organizations, and assessed 47 reports and datasets created by response organizations in Yemen. We find evidence of a cycle of bias reinforcement through which bias cascades between field, headquarters and donor levels of crisis response. Researchers, as well as practitioners, need to consider these underlying biases and reinforcement loops because they influence what data can be collected when, by whom, from whom, and how the data is shared and used. To the CIM literature, we contribute an in-depth understanding of how four types of data bias emerge in crises: political, accessibility, topical, and sampling bias.@en

The United Nations estimates that hundreds of millions of people worldwide are affected by complex crises. Examples are the protracted conflict in Yemen, climate change-induced displacement, and the COVID-19 pandemic. These crises have severe implications for societies. To mitigate crises’ effects, crisis response organizations strive to make data-driven decisions. However, these crises are complex: they involve many actors with different mandates and objectives that face uncertain information as well as decision urgencies. These issues can lead to systematic errors within collected crisis data, i.e., data bias, and challenge decision-makers cognitive information processing capacities by inducing cognitive bias....@en

A crisis requires the affected population, governments or non-profit organizations, as well as crisis experts, to make urgent and sometimes life-critical decisions. With the urgency and uncertainty they create, crises are particularly amenable to inducing cognitive biases that influence decisionmaking. However, there is limited empirical evidence regarding the impact of cognitive biases on estimation, judgment, and decision-making tasks in crises. Possible biases occurring in crises are: (1) to be influenced by how information is framed (i.e., framing effect), (2) to overly rely on information that confirms rather than opposes preliminary assumptions (i.e., confirmation bias), (3) to rely heavily on a skewed informational cue when making estimations (i.e., anchoring bias), and (4) to see the own decision-making as less biased than decision-making of others (i.e., bias blind spot). We investigate these four cognitive biases using three online survey experiments targeting crisis-affected people of the general public (n = 460, mTurk workers), governmental and non-profit workers (n = 50, mTurk workers), and crisis experts (n = 21, purposefully sampled). Our findings show that crisis experts are the least biased group but are still significantly affected by anchoring, framing, and bias blind spot. Crisis-affected people from the general public showed the strongest susceptibility to all four biases studied. The findings have implications for future research on crisis information systems (IS) design. As crisis response is increasingly facilitated through IS, we propose debiasing functions that account for biased user behavior in crises. @en

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.@en

With ongoing research, increased information sharing and knowledge exchange, humanitarian organizations have an increasing amount of evidence at their disposal to support their decisions. Nevertheless, effectively building decisions on the increasing amount of insights and information remains challenging. At the individual, organizational, and environmental levels, various factors influence the use of evidence in the decision-making process. This research examined these factors and specifically their influence in a case-study on humanitarian organizations and their WASH interventions in Uganda. Interviewees reported several factors that impede the implementation of evidence-based decision making. Revealing that, despite advancements in the past years, evidence-based information itself is relatively small, contradictory, and non-repeatable. Moreover, the information is often not connected or in a format that can be acted upon. Most importantly, however, are the human aspects and organizational settings that limit access to and use of supporting data, information, and evidence. This research shows the importance of considering these factors, in addition to invest in creating knowledge and technologies to support evidence-based decision-making. @en

Humanitarian organizations are increasingly challenged by the amount of data available to drive their decisions. Useful data can come from many sources, exists in different formats, and merging it into a basis for analysis and planning often exceeds organizations’ capacities and resources. At the same time, affected communities’ participation in decision making processes is often hindered by a lack of information and data literacy capacities within the communities. We describe a participatory disaster risk analysis project in the central Philippines where the community and a humanitarian NGO worked towards a joint understanding of disaster risks and coping capacities through data integration and IT-supported analysis. We present findings from workshops, focus group discussions and semi-structured interviews, showing the reciprocal effects of the collaborative work. While the community valued the systematically gathered and structured evidence that supported their own risk perceptions and advocacy efforts, the humanitarian NGO revisited established work practices for data collection for analysis and planning.@en

The effectiveness of machine learning algorithms depends on the quality and amount of data and the operationalization and interpretation by the human analyst. In humanitarian response, data is often lacking or overburdening, thus ambiguous, and the time-scarce, volatile, insecure environments of humanitarian activities are likely to inflict cognitive biases. This paper proposes to research the effects of data ambiguity and cognitive biases on the interpretability of machine learning algorithms in humanitarian decision making.@en

In the aftermath of disasters, information is of the essence for humanitarian decision makers in the field. Their concrete information needs is highly context-influenced and often they find themselves unable to access the right information at the right time. We propose a novel ICT-based approach to address these information needs more accurately. First, we select a group of in-field decision makers and collect their concrete information needs in the disaster aftermath. We then review to what extent existing data and tools can already address these needs. We conclude that existing solutions fall short in meeting important information needs of the selected group. We describe the design of an information system prototype to address these gaps more accurately. We combine data of the International Aid Transparency Initiative and the Humanitarian Data Exchange to form the data-backend of our system. We describe our implementation approach and evaluation plan. @en

From November 2017 to April 2018, HumTech Lab at TU Delft partnered with Cordaid to work on an applied research project and address data and analysis challenges in humanitarian resilience and response activities. Central to the project were the questions: what information challenges do humanitarian field workers face? How can communities be empowered through data and information? And how can novel data initiatives like IATI create direct practical value for communities and humanitarian field workers? The project was started by HumTech Lab with a review of the state of research and practice around information needs in humanitarian contexts. A major source were the findings of the 2013 report on humanitarian decision makers information needs1. Existing data sets and software platforms, incl. those dealing with IATI, were examined with respect to their ability to address those information needs. The matching process between information needs and available data and tools uncovered gaps where data and tools do not meet information needs and formed the first set of requirements for a novel software prototype that would select and combine certain data for a more tailored information supply.@en