The outbreak of the COVID-19 pandemic has dramatically shaped higher education and seen the distinct rise of e-learning as a compulsory element of the modern educational landscape. Accordingly, this study highlights the factors which have influenced how students perceive their academic performance during this emergency changeover to e-learning. The empirical analysis is performed on a sample of 10,092 higher education students from 10 countries across 4 continents during the pandemic’s first wave through an online survey. A structural equation model revealed the quality of e-learning was mainly derived from service quality, the teacher’s active role in the process of online education, and the overall system quality, while the students’ digital competencies and online interactions with their colleagues and teachers were considered to be slightly less important factors. The impact of e-learning quality on the students’ performance was strongly mediated by their satisfaction with e-learning. In general, the model gave quite consistent results across countries, gender, study fields, and levels of study. The findings provide a basis for policy recommendations to support decision-makers incorporate e-learning issues in the current and any new similar circumstances.

The elicitation of an ordinal judgment on multiple alternatives is often required in many psychological and behavioral experiments to investigate preference/choice orientation of a specific population. The Plackett-Luce model is one of the most popular and frequently applied parametric distributions to analyze rankings of a finite set of items. The present work introduces a Bayesian finite mixture of Plackett-Luce models to account for unobserved sample heterogeneity of partially ranked data. We describe an efficient way to incorporate the latent group structure in the data augmentation approach and the derivation of existing maximum likelihood procedures as special instances of the proposed Bayesian method. Inference can be conducted with the combination of the Expectation-Maximization algorithm for maximum a posteriori estimation and the Gibbs sampling iterative procedure. We additionally investigate several Bayesian criteria for selecting the optimal mixture configuration and describe diagnostic tools for assessing the fitness of ranking distributions conditionally and unconditionally on the number of ranked items. The utility of the novel Bayesian parametric Plackett-Luce mixture for characterizing sample heterogeneity is illustrated with several applications to simulated and real preference ranked data. We compare our method with the frequentist approach and a Bayesian nonparametric mixture model both assuming the Plackett-Luce model as a mixture component. Our analysis on real datasets reveals the importance of an accurate diagnostic check for an appropriate in-depth understanding of the heterogenous nature of the partial ranking data.

Abstract. We propose the use of probability models for ranked data as a useful alternative to a quantitative data analysis to investigate the outcome of bioassay experiments, when the preliminary choice of an appropriate normalization method for the raw numerical responses is difficult or subject to criticism. We review standard distance-based and multistage ranking models and in this last context we propose an original generalization of the PlackettLuce model to account for the order of the ranking elicitation process. The usefulness of the novel model is illustrated with its maximum likelihood estimation for a real data set. Specifically, we address the heterogeneous nature of experimental units via model-based clustering and detail the necessary steps for a successful likelihood maximization through a hybrid version of the Expectation-Maximization algorithm. The performance of the mixture model using the new distribution as mixture components is compared with those relative to alternative mixture models for random rankings. A discussion on the interpretation of the identified clusters and a comparison with more standard quantitative approaches are finally provided.

Background: We aimed at evaluating the differential involvement of large myelinated Aβ-, small myelinated Aδ-, and unmyelinated C-fibers in patients with diabetic polyneuropathy and how they contribute to neuropathic pain. Methods: We collected clinical and diagnostic test variables in 133 consecutive patients with diabetic polyneuropathy. All patients underwent Aβ-fiber mediated nerve conduction study, Aδ-fiber mediated laser-evoked potentials and skin biopsy mainly assessing unmyelinated C-fibers. Results: Pure large-fiber and small-fiber polyneuropathy were relatively uncommon; conversely mixed-fiber polyneuropathy was the most common type of diabetic polyneuropathy (74%). The frequency of neuropathic pain was similar in the three different polyneuropathies. Ongoing burning pain and dynamic mechanical allodynia were similarly associated with specific small-fiber related variables. Conclusions: Diabetic polyneuropathy mainly manifests as a mixed-fiber polyneuropathy, simultaneously involving Aβ-, Aδ-, and C-fibers. In most patients, neuropathic pain is distinctly associated with small-fiber damage. The evidence that the frequency of neuropathic pain does not differ across pure large-, pure small-, and mixed-fiber polyneuropathy, raises the possibility that in patients with pure largefiber polyneuropathy nociceptive nerve terminal involvement might be undetected by standard diagnostic techniques.

Lung involvement is related to the natural history of anti-citrullinated proteins antibodies (ACPA)-positive rheumatoid arthritis (RA), both during the pathogenesis of the disease and as a site of disease-related injury. Increasing evidence suggests that there is a subclinical, early lung involvement during the course of the disease, even before the onset of articular manifestations, which can potentially progress to a symptomatic interstitial lung disease. To date, reliable, non-invasive markers of subclinical lung involvement are still lacking in clinical practice. The aim of this study is to evaluate the diagnostic potential of functional assessment and serum biomarkers in the identification of subclinical lung involvement in ACPA-positive subjects. Fifty ACPA-positive subjects with or without confirmed diagnosis of RA (2010 ARC-EULAR criteria) were consecutively enrolled. Each subject underwent clinical evaluation, pulmonary function testing (PFT) with assessment of diffusion lung capacity for carbon monoxide (DLCO), cardiopulmonary exercise testing (CPET), surfactant protein D (SPD) serum levels dosage and high-resolution computed tomography (HRCT) of the chest. The cohort was composed of 21 ACPA-positive subjects without arthritis (ND), 10 early (disease duration < 6 months, treatment-naïve) RA (ERA) and 17 long-standing (disease duration < 36 months, on treatment) RA (LSRA). LSRA patients had a significantly higher frequency of overall HRCT abnormalities compared to the other groups (p = 0.001). SPD serum levels were significantly higher in ACPA-positive subjects compared with healthy controls (158.5 ± 132.3 ng/mL vs 61.27 ± 34.11 ng/mL; p < 0.0001) and showed an increasing trend from ND subjects to LSRD patients (p = 0.004). Patients with HRCT abnormalities showed significantly lower values of DLCO (74.19 ± 13.2% pred. vs 131.7 ± 93% pred.; p = 0.009), evidence of ventilatory inefficiency at CPET and significantly higher SPD serum levels compared with subjects with no HRCT abnormalities (213.5 ± 157.2 ng/mL vs 117.7 ± 157.3 ng/mL; p = 0.018). Abnormal CPET responses and higher SPD levels were also associated with specific radiological findings. Impaired DLCO and increased SPD serum levels were independently associated with the presence of HRCT abnormalities. Subclinical lung abnormalities occur early in RA-associated autoimmunity. The presence of subclinical HRCT abnormalities is associated with several functional abnormalities and increased SPD serum levels of SPD. Functional evaluation through PFT and CPET, together with SPD assessment, may have a diagnostic potential in ACPA-positive subjects, contributing to the identification of those patients to be referred to HRCT scan.