Scuola di specializzazione in “Metodologia della ricerca”, Università Pontificia Salesiana.

You can read here the programme of the event (Italian)

PLS Virtual Summer School for Students PSV3 (Italian)
Seconda edizione, 6-10 settembre 2021.

La fallacia dello scommettitore è spesso legata ad un’idea che tutti noi abbiamo della casualità e che si basa su una serie di regole estetiche che secondo noi il caso dovrebbe rispettare per potersi definire tali. Un esempio emblematico al riguardo è la funzione “shuffle” (riproduzione casuali dei brani) che le prime versioni degli iPod di Apple introdussero sul mercato. Quando si selezionava questa opzione, poteva accadere (e spesso accadeva) che venissero riprodotte l’una dopo l’altra più canzoni dello stesso artista e/o album. La reazione dell’utente di fronte a questa ripetizione dipendeva dal particolare brano che veniva ripetuto. In alcuni casi, infatti, quando il risultava gradito, si tendeva ad attribuire una sorta di intelligenza al dispositivo, ipotizzando che lo stesso fosse in grado di “imparare” il gusto musicale del proprietario. Nei casi in cui invece il brano selezionato per la ripetizione non era tra quelli preferiti, non mancavano le critiche rivolte al costruttore. Proprio in seguito a tali critiche, la Apple modificò l’iniziale estrattore casuale dei brani sostituendolo con un estrattore “quasi casuale”, che selezionava casualmente ogni canzone ma solo tra quelle di un differente autore e album rispetto a quella ascoltata precedentemente. La frase di Steve Jobs al riguardo è emblematica: “«We’re making it (the shuffle) less random to make it feel more random», perchè ci permette di intuire che spesso è più facile adattare il caso alle nostre regole estetiche che convincerci di come funziona veramente.

Lo stesso avviene per molti meccanismi distorsivi (bias cognitivi) legati al ragionamento umano in condizioni di incertezza, molti dei quali sono alla base delle falsi convizioni legate al gioco d’azzardo da parte degli scommettitori. Alcuni di questi meccanismi sono stati studiati e descritti da Amos Tversky e Daniel Kahneman nel filone della letteratura nota come economia comportamentale. Analoghi studi, con particolari applicazioni in campi assicurativo, furono sviluppati da Bruno de Finetti nel 1940, con la formalizzazione del teorema della rovina del giocatore.

PDF version of the abstract (Italian)

Webinar d’apertura del convegno ASA2021 - Associazione per la Statistica Applicata
L’Università italiana ai tempi del Covid: dalla gestione dell’emergenza un’opportunità di crescita nel modo di fare didattica?

You can read here the programme of the event

63rd ISI World Statistics Congress 2021, 11-16 July 2021
Session: IPS Non-Live 123 - Statistical methods for assessing the effectiveness of e-learning during the COVID-19 era.

Universities around the world moved the teaching activities online to face with Covid-19 emergency. Even if such a strategy has made it possible to continue offering teaching activities, the switch to online teaching is expected to exacerbate existing educational inequalities and penalize more vulnerable students. Indeed, since less advantaged students have difficulty accessing relevant learning digital resources, family social and economic conditions could impact the e-learning experience. It goes without saying that less advantaged students could experience more likely experiment problems for proper equipment and a comfortable learning environment. This paper aims to analyze the impact of distance learning activities on students’ families regarding the arrangement of the learning environment and the costs to set it up. The study exploits data collected at the University of Naples Federico II. The same data offer the opportunity to reflect on students’ performance due to the closure of the universities.

You can read here the paper

Riunione Scientifica SIS 2021 - Società italiana di statistica
Satellite event: Covid-19: the urgent call for a unified statistical and demographic challenge.

LVII Riunione Scientifica SIEDS - Società italiana di economia demografia e statistica
Session: Composite indicators.

The covid-19 pandemic has disrupted people’s lives and forced governments to impose great sacrifices on their citizens. One of the areas where the pandemic has imposed the greatest sacrifices is certainly education. To avoid the complete suspension of educational activities, it was necessary to adopt a completely different type of teaching than that traditionally used. Schools and Universities around the world have responded to the crisis by moving teaching activities online. While the enormous effort made to quickly make the transition to these new ways of teaching was certainly appreciated, the transition certainly had a major impact on the lives of students and teachers and their families.

The aim of this paper is to evaluate the impact of e-learning on the different types of students, in terms of preparation, characteristics and social background. The challenge is to explore if the switch from on-site to online learning caused by the emergence is exacerbating existing educational inequalities penalising more vulnerable students. The risk is that the social and economic conditions of families have a major influence on the e-learning experience because less advantaged students are less likely to have access to relevant learning digital resources (e.g. laptop/computer, broadband internet connection) and less likely to have a suitable home learning environment (e.g. a quiet place to study or their own desk).

The study is based on the analysis of data collected at the University of Naples Federico II in June 2020. More than 19,000 students took part in a survey, carried out to monitor distance learning activities and perceptions. The paper exploits a factorial method to obtain a composite indicator measuring the family impact of distance learning. Then, the family impact is analysed, trying to understand if it takes different forms and intensity depending on the students’characteristics, the availability of computer equipment and the type of teaching used. Finally, quantile regression allows to differentiate the study of effects for different levels of family impact. Some considerations on the preferred teaching method for the future and on the effects that the closure of universities has had on the performance of students are also enclosed.

The results, although in many cases expected, allow to quantify, and visualising heterogeneity in the conditions and characteristics of students. For example, the study quantifies the difference in the family impact among students who predominantly experienced a quiet e-learning experience without changing family habits (they already had all the equipment available for their exclusive use) and students who were forced to share both the workstation and the device with family members engaged in smart working or other learning activities. Moreover, quantile regression detects the different effects of socio-demographics and IT equipment in case of low, medium, or high family impact.

[PDF version of the abstract]/files/seminars/2021-sieds/2021-sieds-davino_vistocco-abstract.pdf)

Pre-conference on “Evaluation of Educational Systems” of ASA - Applied Statistics Associations.

The Covid-19 emergency has forced universities around the world to transfer teaching activities online. Even if online teaching allowed to carry out the planned teaching activities, it is necessary, in retrospect, to evaluate the impact of this teaching method on the different types of students, in terms of preparation, characteristics and social background. The switch from offline to online learning caused by Covid-19 is expected to exacerbate existing educational inequalities penalising more vulnerable students. The social and economic conditions of families have a major influence on the e-learning experience because less advantaged students are less likely to have access to relevant learning digital resources (e.g. laptop/computer, broadband internet connection) and less likely to have a suitable home learning environment (e.g. a quiet place to study or their own desk) (Di Pietro et al., 2020). Furthermore, according to the 2020 European Commission’s annual report on the levels of digitalisation achieved by the various member states, Italy ranks 25th among the 28 EU Member States.

The aim of this paper is to analyse whether and how the distance learning activities impacted on the students’ families both in terms of the organisation of spaces and daily rhythms and from an economic point of view, having required additional expenses. The study is based on the analysis of data collected at the University of Naples Federico II in June 2020. More than 19,000 students took part in a survey, carried out to monitor distance learning activities and perceptions. The paper is organised into two sections. In the first, a factorial method is exploited to obtain a composite indicator measuring the family impact of distance learning. Then, we try to explain if the family impact takes different forms and intensity depending on the students’ characteristics, the availability of computer equipment and the type of teaching used. Finally, quantile regression allow to differentiate the study of effects for different levels of family impact. Some considerations on the distance learning experience in terms of family impact and the evaluation on the preferred teaching method for the future are also enclosed.

You can read here the paper

PSYCHOBIT 2020 - Second Symposium on Psychology-Based Technologies
Conference webpage

Over the years, several studies have shown the relevance of one-to-one compared to one-to-many tutoring, shedding light on the need for technology-based platforms to assist traditional learning methodologies. Therefore, in recent years, tutoring systems that collect and analyse responses during the user interaction for an automated assessment and profiling were developed as a new standard to improve the learning outcome. In this framework, the tutoring system Adaptive LEArning system for Statistics (ALEAS) is aimed at providing an adaptive assessment of undergraduate students’ statistical abilities enrolled in social and human sciences courses. ALEAS is developed in the contest of the ERASMUS+ Project (KA+ 2018-1-IT02-KA203-048519). The article describes the ALEAS workflow; in particular, it focuses on the students’ categorisation according to their abilities. The student follows a learning process defined according to the Knowledge Space Theory, and she/he is classified at the end of each learning unit. The proposed classification method is based on the multidimensional latent class item response theory, where the dimensions are defined according to the Dublin learning dimensions. In this work, results from a simulation study support our approach’s effectiveness and encourage its future use with students.

• You can read here the paper

ASA Conference 2019 - Statistics for Health and Well-being
Conference of the Applied Statistics Association.

Infertility is a major psychosocial crisis as well as being a medical problem. The factors that predict psychosocial consequences of infertility may vary in different gender, education level, socio-economic status. The primary purpose of this study was to investigate the relationship between sociodemographic characteristics and levels of depression and anxiety in infertile couples by exploring the role of each partner and of the related perceived levels of depression and of quality of dyadic adjustment.

This paper analyses these components by means of a mixture model for ordinal rating responses, allowing for uncertainty in answering (Piccolo, 2003). In responding to rating questions as the latent components which lead the perception of depression and/or anxiety, an individual may give answers either according to his/her feeling or to his/her level of indecision, typically motivated by a response style. Since ignoring this uncertainty may lead to misleading results, we define the distribution of the ordinal responses via a mixture model which weights both components in answering. The study allows also to model the actor/partner interdependence in case of categorical dyadic data by presenting an alternative approach with respect to the current used methods (see Kenny et al. (2006), among others).

The effectiveness of the model is attested through the analysis of a cross-sectional study of 206 infertile couples interviewed from 2014 to 2016. A gynecologist evaluated participants for demographic and medical data and then they were visited by a psychologist to perform questionnaire scales which were the Dyadic Adjustment Scale, the Edinburgh Depression Scale and the State-Trait Anxiety Inventory for the evaluation of the perceived levels of psychological disease (Zurlo et al, 2018, 2017).

• PDF version of the abstract
• Read here the paper

ASA Conference 2019 - Statistics for Health and Well-being
Conference of the Applied Statistics Association.

The Italian system of indicators on Equitable and Sustainable Well-being (Benessere Equo e Sostenibile - BES) proposed by the National Institute of Statistics represents a well-established reference database in the national and international debate on the research on alternative well- being measures. The main strengths of this set are represented by the broad coverage of all the components of this complex concept and the availability of information not only at the aggregate level but also at the provincial level (NUTS3 level) (Istat, 2019; Taralli et al. 2015). In this framework, it is possible to consider not only the levels of well-being but also the differences in their distribution thus highlighting differences in the territories.

The paper proposes an advancement of work elaborated in Davino et al. (2018), where a hierarchical model was used to study the relationships among components of the BES. The proposed hierarchical model allows us to synthesize individual indicators into single indexes, in order to construct composite indicators at a global and a partial level. Partial Least Squares path modeling (Lohmöller J.B., 1989) and a recent method, called Quantile Composite-based path modeling (Davino and Vinzi, 2016), were used respectively to estimate average effects in the network of relationships among variables and to explore whether the magnitude of these effects changes across different parts of the variables distributions. The present contribution aims to deepen the study taking into account that living conditions are quite different according to an unobserved or observed heterogeneity (for example according to the geographic area of the province).

• PDF version of the abstract
• Read here the paper

ASA Conference 2019 - Statistics for Health and Well-being
Conference of the Applied Statistics Association.

Academic well-being and performance are important tasks to achieve in all school grades, and are negatively affected by stress and anxiety. Actually, students feel discomfort with regard to specific subjects, like Math and Statistics. For this reason, mathematical and statistical anxiety have been widely studied (among the others, see Primi, Donati & Chiesi, 2016). Statistics is a mandatory course in the curriculum in most of humanities programs. Many authors showed that these students consider Statistics as a burden and exhibit higher levels of statistical anxiety. Statistical anxiety can be defined as “the feeling of anxiety encountered when attending a Statistics course or doing statistical analyses” (Cruise, Cash & Bolton, 1985, p.92). These students are made weary by anything related to Mathematics and believe that Statistics is not important for their degree programs and careers (Primi, Donati & Chiesi, 2016). Moreover, Statistics is viewed as an unpleasant and difficult subject, making students feel uncomfortable and leading them to believe that they are not able to achieve the task that is being requested from them. Several studies on this topic classified the statistical anxiety antecedents. They are typically divided in situational factors (e.g. math skills, previous statistical experience), dispositional factors (e.g. attitude toward Statistics, self-concept and self-efficacy) and demographic factors (e.g. gender, age). The most common and widely used psychometrics tools to assess statistical anxiety are STARS (Cruise, Cash & Bolton, 1985) and SAS (Vigil-Colet, Lorenzo-Seva & Condon, 2008).

In this work we present a preliminary analysis based on data collected within the ALEAS (Adaptive LEArning in Statistics) ERASMUS+ project (https://aleas-project.eu/wordpress/) about statistical anxiety in undergraduate students enrolled in the Psychology course at Federico II University of Naples.

Path analysis (Duncan, 1966) was carried out to study the interplay between statistical anxiety and a set of considered variables. Our results show that math background affects the attitude towards Statistics and the statistical anxiety. Moreover, statistical anxiety also depends on other variables such as math comprehension, gender, high school final mark, and past experience in Statistics.

• PDF version of the abstract
• Read here the paper

INTED 2019 - 13th International Technology, Education and Development Conference
Valencia, Spain, 11-13 March

Massive Open Online Courses differ from traditional forms of learning, because of their free access, abundance of resources and of user interaction tools available. Notwithstanding these features, some learners reach their achievement while others do not. In learning analytics literature there are many contributing relating to how Massive Open Online Courses learners’ behaviours determine performance. The paper proposes a study to face the conceptualisation, operativisation and measurement of learning and engagement, two determinists of learners’ performance. Moreover, considering the multidimensional and complex nature of Massive Open Online Courses learning assessment, a Partial Least Squares model is also proposed.

• Read here the paper

Cladag 2019 - 12th Scientific Meeting Classification and Data Analysis Group
Cassino, September 11 – 13, 2019.

Quantile Composed-based Path Modeling complements the classical PLS Path Modeling. The latter is widely used to model relationships among latent variables and between the manifest variables and their corresponding latent variables. Since it essentially exploits classical least square regressions, PLS Path Modeling focuses on the effect the predictors exert on the conditional means of the different outcome variables involved in models. Quantile Composed-based Path Modeling extends the analysis to the whole conditional distributions of the outcomes. This paper proposes a procedure to estimate the conditional quantiles for the manifest variables of the outcome blocks. Starting from the information related to a grid of conditional quantiles, it is possible to define the most accurate model for each health indicator and the best predictive model for each Italian province. The proposed method is shown in action both on artificial and real data. The real data concerns the prediction of health indicators.

• Read here the paper

Cladag 2019 - 12th Scientific Meeting Classification and Data Analysis Group
Cassino, September 11 – 13, 2019.

The aim of the paper is to propose a new approach to preference mapping by exploiting quantile regression. The proposal consists into a multi-steps procedure combining principal component analysis, least squares and quantile regression. Results of the procedure on a case study show how the classical preference map can be enriched by information on the variability along the direction of the most preferred products. Such an additional information is obtained by the use of quantile regression.

• Read here the paper

IES 2019 - Statistical Methods for Service Quality Evaluation
9th International Conference IES 2019 - Innovation & Society
Statistical evaluation systems at 360°: techniques, technologies and new frontiers