Franca Crippa

Ambivalent stereotypes link to peace, conflict, and inequality across 38 nations

Significance Stereotypes reflect a society’s inequality and conflict, providing a diagnostic map of intergroup relations. This stereotype map’s fundamental dimensions depict each group’s warmth (friendly, sincere) and competence (capable, skilled). Some societies cluster groups as high on both (positive “us”) vs. low on both (negative “them”). Other societies, including the United States, have us-them clusters but add ambivalent ones (high on one dimension, low on the other). This cross-national study shows peace-conflict predicts ambivalence. Extremely peaceful and conflictual nations both display unambivalent us-them patterns, whereas intermediate peace-conflict predicts high ambivalence. Replicating previous work, higher inequality predicts more ambivalent stereotype clusters. Inequality and intermediate peace-conflict each use ambivalent stereotypes, explaining complicated intergroup relations and maintaining social system stability. A cross-national study, 49 samples in 38 nations (n = 4,344), investigates whether national peace and conflict reflect ambivalent warmth and competence stereotypes: High-conflict societies (Pakistan) may need clearcut, unambivalent group images distinguishing friends from foes. Highly peaceful countries (Denmark) also may need less ambivalence because most groups occupy the shared national identity, with only a few outcasts. Finally, nations with intermediate conflict (United States) may need ambivalence to justify more complex intergroup-system stability. Using the Global Peace Index to measure conflict, a curvilinear (quadratic) relationship between ambivalence and conflict highlights how both extremely peaceful and extremely conflictual countries display lower stereotype ambivalence, whereas countries intermediate on peace-conflict present higher ambivalence. These data also replicated a linear inequality–ambivalence relationship.

Departures from the formal of actual students' university careers: an application of non-homogeneous fuzzy Markov chains

As in most higher education (HE) systems, the Italian university organisation draws paths of credit progression in formal curricula, which aim at framing the acquisition of knowledge and competencies within each specific major. The resulting yearly syllabi therefore develop in a sequence of examinations that are to be successfully passed, and formal administrative registration allows access to the following academic year. In general, there is a divergence between formal and actual career progression because each university student can proceed at her/his own pace, sketching her/his own trajectories, free to depart from the formal progression. Even if applied to various HE settings, Markov chain models do not fit the aforementioned situation. A methodological extension has been introduced, whereby progression levels are considered as fuzzy states. Markov chains with fuzzy states identify the latter with specified academic years and express each students situation as a relational link to present and past academic attainments. This link is operationalised by means of a membership function, which is here discussed with reference to the Italian HE system.

Moderators of the feature-positive effect in abstract hypothesis-evaluation tasks.

Three studies using abstract materials tested possible moderators of the feature-positive effect in hypothesis evaluation whereby people use the presence of features more than their absence to judge which of 2 competing hypotheses is more likely. Drawing on a distinction made in visual perception research, we tested whether the feature-positive effect emerges both when using nonsubstitutive features, which can be removed without replacement by other features, and substitutive features, the absence of which implies the presence of other features (e.g., the colour red, the absence of which entails the presence of another colour). Furthermore, we tested whether presenting to participants both the clue occurrence probabilities (which are needed to consider clue presence) and their complements (which are needed to gauge the impact of the absent clues) decreased the feature-positive effect. The results showed that regardless of the type of feature (i.e., nonsubstitutive vs. substitutive), participants provided more responses consistent with an evaluation of the subset of present clues compared to all other kinds of responses. However, the use of substitutive features combined with an explicit presentation format of probabilistic information had a debiasing effect. Furthermore, the use of substitutive features negated participant sensitivity to the rarity of clues, whereby the feature-positive effect decreased when there was one absent clue and two present clues for problems in which the exclusive consideration of the presence of features did not suggest the correct response.

Measuring Latent Variables in Space and/or Time: A Gender Statistics Exercise

This paper concerns a Multivariate Latent Markov Model recently introduced in the literature for estimating latent traits in social sciences. Based on its ability of simultaneously dealing with longitudinal and spacial data, the model is proposed when the latent response variable is expected to have a time and space dynamic of its own, as an innovative alternative to popular methodologies such as the construction of composite indicators and structural equation modeling. The potentials of the proposed model and the added value with respect to the traditional weighted composition methodology, are illustrated via an empirical Gender Statistics exercise, focused on gender gap as the latent status to be measured and based on supranational o cial statistics for 30 European countries in the period 2010–2015.

The Coxian phase-type distribution as a contribution to the multilevel model of in-hospital mortality

ABSTRACT This paper presents multilevel models that utilize the Coxian phase-type distribution in order to be able to include a survival component in the model. The approach is demonstrated by modeling patient length of stay and in-hospital mortality in geriatric wards in Italy. The multilevel model is used to provide a means of controlling for the existence of possible intra-ward correlations, which may make patients within a hospital more alike in terms of experienced outcome than patients coming from different hospitals, everything else being equal. Within this multilevel model we introduce the use of the Coxian phase-type distribution to create a covariate that represents patient length of stay or stage (of hospital care). Results demonstrate that the use of the multilevel model for representing the in-patient mortality is successful and further enhanced by the inclusion of the Coxian phase-type distribution variable (stage covariate).

Hospitalization in Geriatric Wards: Considerations on the Evaluation of the Outcome

Abstract Within a defined law framework, the Italian central health system dictates the standards for hospitalization to local care units, which are in turn allowed to establish their own effectiveness criteria. The appropriateness of the hospitalization decision is therefore predetermined at patients admission, whereas its effectiveness relies on the ex post patient well-being as a result of the complex system of reciprocal relations between patients and healthcare agents at the ward level. We consider the outcomes in geriatric wards referring to the national health system, with respect both to patients traits at the individual level and wards/hospital settings. The risk that models the healthcare outcome is accordingly adjusted for covariates at the different levels of analysis (Goldstein & Spiegelhalter, 1996), thus allowing to differentiate among outcomes in terms of the hospitalization structure and, when appropriate, of territorial aggregation.