Improved Transferability of Data‐Driven Damage Models Through Sample Selection Bias Correction

Damage models for natural hazards are used for decision making on reducing and transferring risk. The damage estimates from these models depend on many variables and their complex sometimes nonlinear relationships with the damage. In recent years, data‐driven modeling techniques have been used to capture those relationships. The available data to build such models are often limited. Therefore, in practice it is usually necessary to transfer models to a different context. In this article, we show that this implies the samples used to build the model are often not fully representative for the situation where they need to be applied on, which leads to a “sample selection bias.” In this article, we enhance data‐driven damage models by applying methods, not previously applied to damage modeling, to correct for this bias before the machine learning (ML) models are trained. We demonstrate this with case studies on flooding in Europe, and typhoon wind damage in the Philippines. Two sample selection bias correction methods from the ML literature are applied and one of these methods is also adjusted to our problem. These three methods are combined with stochastic generation of synthetic damage data. We demonstrate that for both case studies, the sample selection bias correction techniques reduce model errors, especially for the mean bias error this reduction can be larger than 30%. The novel combination with stochastic data generation seems to enhance these techniques. This shows that sample selection bias correction methods are beneficial for damage model transfer.     

Leader-member exchange as a linking pin in the idiosyncratic deals - Performance relationship in workgroups

In the current investigation, idiosyncratic deals (i-deals; individualized work arrangements) are modeled as differentiated resources that shape leader-member exchange (LMX) relationships in workgroups. We integrate literature on leader-member exchange (LMX) with research on i-deals to argue that employee evaluations of i-deals received from the grantor –typically the leader- enhance employee perceptions of LMX, which in turn become instrumental in generating positive performance outcomes. Furthermore, because workgroup characteristics have potential implications on the relationship between a deal grantor and the deal recipient, drawing upon social identity theory of leadership, we reason that the i-deals-LMX relationship is affected by the overall value congruence among the group members. Cross-level moderated mediation analyses on multi source data obtained from 289 employees nested in 60 workgroups showed that the mediational role of LMX in the i-deals to performance outcomes relationship was weaker in high value congruence groups.     

Constructing Trustworthy Historical Narratives: Criteria,Principles and Techniques

Organizational scholars increasingly recognize the value of employing historical research. Yet the fields of history and organization studies struggle to reconcile. In this paper, the authors contend that a closer connection between these two fields is possible if organizational historians bring their role in the construction of historical narratives to the fore and open up their research decisions for discussion. They provide guidelines to support this endeavor, drawing on four criteria that are prevalent within interpretive organization studies for developing the trustworthiness of research: credibility; confirmability; dependability; and transferability. In contrast to the traditional use of trustworthiness criteria to evaluate the quality of research, the authors advance the criteria to encourage historians to generate more transparent narratives. Such transparency allows others to comprehend and comment on the construction of narratives, thereby building trust and understanding. Each criterion is converted into a set of guiding principles to enhance the trustworthiness of historical research, pairing each principle with a practical technique gleaned from a range of disciplines within the social sciences to provide practical guidance.     

Has Opposition to Immigration Increased in the United States after the Economic Crisis? An Experimental Approach

We employ two population‐level experiments to accurately measure opposition to immigration before and after the economic crisis of 2008. Our design explicitly addresses social desirability bias, which is the tendency to give responses that are seen favorably by others and can lead to substantial underreporting of opposition to immigration. We find that overt opposition to immigration, expressed as support for a closed border, increases slightly after the crisis. However, once we account for social desirability bias, no significant increase remains. We conclude that the observed increase in anti‐immigration sentiment in the post‐crisis United States is attributable to greater expression of opposition rather than any underlying change in attitudes.     

Negotiating the Intersection of Racial Oppression and Heteronormativity

This qualitative study examines the narratives of seven heterosexual, cisgender individuals who identify both as persons of color and as positively disposed toward LGB and transgender-identified persons. Using psycho-discursive qualitative methodology, the authors will present the narrative strategies taken up by these positively disposed cisgender heterosexuals of color as they attempt to position themselves as supportive of LGB and transgender persons while negotiating the discourse of heteronormativity. The three narrative strategies have been titled differentiation, empathy, and coherence. In addition to mapping the three narrative strategies, the authors also explore why informants may choose certain strategies over others and argue that the intersection of social identities must be considered when attempting to understand social oppression. Implications for the human services fields will be discussed.     

A new long-term survival model with dispersion induced by discrete frailty

Lifetime Data Analysis - Frailty models are generally used to model heterogeneity between the individuals. The distribution of the frailty variable is often assumed to be continuous. However, there...     

The matching extension problem in general graphs is co-NP-complete

A simple connected graph G with 2n vertices is said to be k-extendable for an integer k with \(0<k<n\) if G contains a perfect matching and every matching of cardinality k in G is a subset of some perfect matching. Lakhal and Litzler (Inf Process Lett 65(1):11–16, 1998) discovered a polynomial algorithm that decides whether a bipartite graph is k-extendable. For general graphs, however, it has been an open problem whether there exists a polynomial algorithm. The new result presented in this paper is that the extendability problem is co-NP-complete.