Search

Survey data are usually of mixed type (quantitative, multistate categorical, and/or binary variables). Multidimensional scaling (MDS) is one of the most extended methodologies to visualize the profile structure of the data. Since the past 60s, MDS methods have been introduced in the literature, initially in publications in the psychometrics area. Nevertheless, sensitivity and robustness of MDS configurations have been topics scarcely addressed in the specialized literature. In this work, we are interested in the construction of robust profiles for mixed-type data using a proper MDS configuration. To this end, we propose to compare different MDS configurations (coming from different metrics) through a combination of sensitivity and robust analysis. In particular, as an alternative to classical Gower's metric, we propose a robust joint metric combining different distance matrices, avoiding redundant information, via related metric scaling. The search for robustness and identification of outliers is done through a distance-based procedure related to geometric variability notions. In this sense, we propose a statistic for detecting multivariate outliers in the context of mixed-type data and evaluate its performance through a simulation study. Finally, we apply these techniques to a real data set provided by the largest humanitarian organization involved in social programs in Spain, where we are able to find in a robust way the most relevant factors defining the profiles of people that were under risk of being socially excluded in the beginning of the 2008 economic crisis.

Objective: To analyze the clinical characteristics and outcomes of patients with COVID-19-related acute respiratory failure on the basis of their vaccination status at the time of ICU admission. Methods: We conducted a retrospective observational study using a prospective database of patients admitted to the ICU of a university hospital in the city of Murcia, in Spain, between January 1, 2021 and September 1, 2022. Clinical, analytical, and sociodemographic data were collected and analyzed on the basis of patient vaccination status. We adjusted for confounding variables using propensity score matching and calculated adjusted ORs and 95% CIs. Results: A total of 276 patients were included in the study. Of those, 8.3% were fully vaccinated, 12% were partially vaccinated, and 79.7% were unvaccinated. Although fully vaccinated patients had more comorbidities, partially vaccinated patients had higher disease severity. The proportion of patients with severe acute respiratory failure was higher in the unvaccinated group, followed by the partially vaccinated group. No significant differences were found among the different groups regarding complications, duration of ventilatory support, or length of ICU/hospital stay. In the sample selected by propensity score matching, the number of patients with severe complications and the in-hospital mortality rate were higher in unvaccinated patients, but the differences were not significant. Conclusions: This study failed to show a significant improvement in outcomes in critically ill COVID-19 patients vaccinated against SARS-CoV-2. However, the CIs were wide and the mortality point estimates favored patients who received at least one dose of COVID-19 vaccine. [ABSTRACT FROM AUTHOR]

The frequency of producers opting to abandon agricultural land has become increasingly, highlighting the significance of this phenomenon due to its environmental, landscape, and socio-economic impacts. The decisions of producers to abandon or maintain/improve their farms depend on individual and contextual factors. The aims of this research are twofold. Firstly, to evaluate the influence of the neighbours on the winegrowers' decisions, using spatial analysis. Secondly, to clarify the specific importance of each of the individual and contextual drivers in farmers' decisions to improve their farms, to keep them unchanged or to abandon them, using multilevel models. The results obtained for the case study of vineyards in Spain, reveal a strong agglomeration phenomenon in farmers' decisions indicating that producers make land use decisions influenced by what their neighbours do. A multilevel analysis identifies that individual factors are determinant and that the influence of contextual factors is conditioned by the innovation process at farm level. Individual drivers, such as size, innovation, Protected Designations of Origin and irrigation influence vineyard area, with irrigation having the greatest overall influence, and is expected to be decisive in climate change projections. The Protected Designations of Origin are driving forces that dynamize the territory and achieve productive concentrations, encouraging winegrowers to replant, but they are not enough to halt abandonment. The elements that slow down the abandonment of plots are irrigation and the combination of innovation and context variables, mainly the combination of modernised plots in the municipalities with trading options. [Display omitted] • Farmers' land use decisions are made considering individual and contextual drivers. • Specific importance of each driver is obtained by multinomial multilevel model. • Spatial analysis estimates that producer decision is not due to random patterns. • Irrigation is the most important individual driver, decisive in climate projections. • Abandonment is mitigated by a combination of innovation and commercial options. [ABSTRACT FROM AUTHOR]