232451 research outputs found
Sort by
Making degreeness count in QCA:Problems with fuzzy sets and a linguistic solution
No full textQCA prefers fuzzy sets because they capture all empirical variation. However, fuzziness plays no role in minimizing the truth table, which follows threshold (crisp) logic also in fsQCA. Fuzzy set values only matter for calculating set-relationships, where they present a problem. Fuzzy set-relationships are calculated over all cases, whereas crisp set-relationships only consider cases that provide corroborating or contradicting within-case (i.e. crisp) causal evidence. When developing causal explanations, fsQCA thus dialogues incompatible within-case and cross-case causal evidence. In response, this paper suggests capturing degreeness with linguistic hedges – e.g. somewhat, moderately, considerably and very – but calculating crisp set-relationships. This avoids incompatible within-case and cross-case causal evidence. The crossover point sits between intensifying and diluting hedges, but researchers can make degreeness count by setting lower or higher crossover points to investigate (i) necessity in degree, (ii) explanatory power of individual conditions and (iii) differences between solutions for higher and lower crossover points
Optimality conditions for penalized sparse PCA
No full textThis paper establishes the theoretical foundations of an alternating optimization scheme for penalized sparse principal component analysis (PCA) focusing on variance maximization. We provide a theoretical foundation for the optimality of solutions derived from this widely used algorithm, addressing a gap in the current literature where empirical results often lack theoretical support. We show the algorithm’s success when the dataset’s covariance matrix is positive definite. Additionally, we characterize sparsity-inducing penalties and examine the use of various ones, including the L1-norm, SCAD, and L0-norm. We conduct numerical experiments to evaluate standard metrics, such as explained variance, number of iterations, and computational time
Bystanders’ perceptions on online hate speech:Investigating the effects of perpetrators’ justifications and the bystander’s role on bystanders’ attitude and prosocial intervention intentions
No full textOn social media, users are exposed to online hate speech (OHS), which is a type of speech that attacks a person or a group based on a group characteristic, e.g., gender identity or sexual orientation. Not every bystander evaluates OHS as offensive and/or feels the need to intervene, which can lead to the continuation of OHS and damaging consequences for victims. The goal of the present study was to understand attitudinal and behavioral components of bystanders’ perceptions on OHS by investigating content-related, contextual, and personal characteristics. More precisely, the effects of the presence or absence of online moral disengagement strategies or moral excuses in OHS messages (e.g., “I’m posting this because it doesn't hurt if I share my opinion online”) and the bystander’s role (pure bystander or vicarious victim) on bystanders’ attitudes and behaviors were tested, while controlling for previous experience with OHS and connectedness with the target group. To this aim, a repeated measures experiment (5x2x2 mixed design) was conducted among 633 adults aged 18–25. The results indicated no difference in bystanders’ perceived offensiveness of OHS and intention to intervene when exposed to OHS containing a moral excuse compared to OHS without. When bystanders were vicarious victims (being exposed to OHS targeting an individual with whom the bystander shares the targeted group characteristic), OHS was perceived as more offensive and bystanders had a higher intention to intervene with prosocial bystander behavior, compared to when bystanders did not share the group characteristic. Theoretical and practical implications are discussed
Tilburg is a real university town
No full textIs Tilburg een stad met een universiteit of een universiteitsstad? Intussen is vast te stellen dat Tilburg zich met recht een universiteitsstad kan noemen
Do we need a world climate court?
No full textWe are at a critical juncture in the history of international law, as international courts and dispute settlement bodies grapple with the unfolding climate crisis. This article theorises a World Climate Court as a way of evaluating existing institutions which are being called upon to handle climate-related cases. By discussing the potential composition, jurisdiction and remedial regimes of a World Climate Court, we argue that existing international courts are less than deally equipped for dealing with climate change cases. As a counterpoint, we suggest a World Climate Court composed of international law experts with broad legal expertise and supported by climate scientists. The article argues that a specialised court with a broad mandate to assess the international legal impacts of climate change could offer a structural and redistributive approach to remedies, and decide on climate cases in a more expeditious manner
Examining the effect of a firm’s AI specialization on the technology firms it acquires: A real options perspective
No full textThe digital revolution is transforming global business practices. As organizations increasingly embed artificial intelligence (AI) within their operations, they face unprecedented uncertainty regarding future technological trajectories and competitive landscapes. To maintain competitiveness in the emerging technological space, they need to promptly acquire advanced knowledge to enrich their technological portfolio. Drawing on real options theory (ROT), our study integrates AI-based acquisitions with internal AI development. We posit that a firm’s AI specialization represents the accumulation of critical technological knowledge and creates a portfolio of strategic options. These options can subsequently be exercised as AI acquisitions to secure complementary external capabilities. Moreover, the efficacy of these options is contingent on distinct uncertainties, including technical, modal, and complementarity uncertainties, captured by target R&D intensity, target self-fluidity, and product market overlap, respectively. Using a sample of US public firms that carried out acquisitions from 2004 to 2015, we find support for our hypotheses