Search

This dissertation, composed of three essays, aims at providing an evaluation of selected public policies in the field of addictions, focusing on drugs and gambling. Evaluation is increasingly recognised as essential for effective policymaking, helping to empirically verify that policies and programmes have the desired effect, provide value for money and do not have negative unintended consequences. In the case of socially undesirable behaviours, policy evaluation is frequently challenged by the limited data available. To tackle this challenge, the analyses proposed in the present work make use of novel datasets which combine epidemiological, financial, political and administrative data. On the one hand, addictive behaviours constitute an enormous source of revenue for organized crime and for national governments, in the case of legal substances or behaviours such as gambling. On the other hand, they represent a cost for public authorities, e.g., for public health and law enforcement reasons. For these reasons, over the years several policy approaches have been implemented to contrast the externalities produced by the spread of addictive behaviours, however very limited work has been performed to evaluate their effects. The present dissertation aims at contributing to the literature seeking to provide evidence on the possible intended and unintended effects of public policies on addictions, focusing on selected policies aimed at regulating cannabis consumption and gambling. The first chapter is aimed at analysing the effects that the drug policy reforms implemented in Europe in the last 15 years had on youth cannabis consumption behaviours. In fact, cannabis is the most used illicit drug and accounts for the largest share of the European illicit drug market. To tackle this longstanding problem, a broad set of reforms to national cannabis control policies have been implemented, but their effectiveness in achieving the desired results is still unclear. This essay analyses the association between selected categories of cannabis policy reforms and changes in cannabis availability and useamong adolescents. Data from 20 European countries across 15 years were drawn from a novel database of the European school Survey Project on Alcohol and other Drugs. The influence of policy changes on reported access to cannabis and different frequencies of use was estimated through an identification strategy based on a Differences-in-Differences model, which application is allowed by the fact that only thirteen out of the twenty countries included in the analysis implemented relevant reforms. The findings suggest that selected categories of reforms influence the availability and prevalence of cannabis among adolescents. In particular, some forms of restrictive interventions reduce the general prevalence of use and more liberal reforms seem to cause an increase in the share of students approaching cannabis. However, none of the analysed reforms seems to have been able to influence the consumption habits of frequent users. The second chapter deepens the analysis of the effects of drug policy changes by focusing of the reform that in the period 2006-2014 criminalised the possession of cannabis for personal use in Italy. The chapter addresses a key question in the ongoing policy debate over cannabis? legal status as to whether removing or increasing penalties attached to cannabis use and possession leads not only to changes in the proportion of cannabis users, but also to changes in the dynamics of cannabis use. Empirically investigating this aspect is a growing field of study. This paper aims at extending previous research by studying the effect of increasing penalties on initiation into cannabis use. This is done in the Italian context where, despite the relevant drug policy reforms implemented over the last 30 years, very little econometric studies have been performed to evaluate their effects. In order to identify the effect of criminalisation on the transition into cannabis use, the paper exploits the policy change that in the period from 2006 to 2014 increased the penalties attached to cannabis possession in Italy. For this, a unique dataset pooling seven waves (2001-2017) of the Italian Population Survey on Alcohol and other Drugs (IPSAD) is used. The empirical investigation is based on a Complementary Log-Log regression model to analyse the starting rate, that is, the transition rate from non-use to use. The results suggest that the introduction of higher punishments for cannabis possession has a significant, although limited, effect in reducing the uptake of cannabis. However the research highlights that this relative achievement need to be evaluated against the backdrop of the social consequences of criminalisation. The third chapter is aimed at assessing the market effects of limiting the supply of gambling products. This is done by exploiting the effects of the local policies that imposed restrictions on the supply of offline entertainment machines implemented in the years 2017-2018 by some municipalities in one of the Northern regions of Italy. The market effects of reducing the allowed functioning time of offline entertainment machines are assessed both on the level of the gross gambling turnover and on its composition in terms of possible shifts towards online gambling products. Although the policy should have applied to the entire regional territory, some municipalities did not approve the necessary implementing acts, whilst others imposed functioning time restrictions with varying level of intensity. This variation is exploited using a regression-based Differences-in-Differences (DiD) design with continuous treatment with a unique dataset recording yearly gambling turnover by type of game in the years 2016, 2017, and 2018 matched with data on municipal acts collected for the study and socio-demographic variables. The results suggest that the implemented policies actually reduced the gambling consumption through offline entertainment machines. However, no significant reduction of the total gambling consumption was observed. In fact, focusing on the composition of gambling turnover there is evidence of a shift toward online gambling products. These results support the argument that limiting the physical availability of entertainment machines is an effective policy for reducing the amount of resources invested by consumers in this type of game. At the same time attention should be paid to the possible unintended effects on the shift towards other gambling products, which by definition cannot be subject to similar restrictions.

This Thesis uses administrative data in order to investigate some crucial issues about Italian Student Population such as internal mobility with a particular focus to South-North Migration; freshmen performance in STEM area and School-to-Work Transition. The analysis are based on two novel datasets and they benefit of three different modelling techniques in order to appropriately explore the three research questions.

This paper evaluates multidimensional poverty in European countries introducing two main novelties compared with the previous literature: first, the dimensions of poverty are selected on the basis of the shared values included in the Charter of Fundamental Rights of the European Union; second, the whole space of feasible weights is used to summarise the multidimensional information, in order to remain agnostic about the importance given to the different deprivations. Using data from four waves of EU-SILC, the methodological innovations introduced here have allowed to produce a family of measures that capture the individual probability of being multidimensionally poor. Individual probabilities are then used to analyse the within and between distribution of multidimensional poverty in ten countries. Finally, they get combined with the generalised Lorenz dominance techniques in order to derive socially preferred distributions with the minimum load of value judgments. The novel methods proposed in this analysis allow to move from a dual definition of poverty, where poor and non-poor individuals are classified in a mutually exclusive context, to a continuous measure of deprivation, which allows to capture both the extensive and intensive margin of multidimensional poverty.

Neighbourhood effects have been defined by Oakes (2004) as the independent causal effects of neighbourhood on a given number of health or social outcome(s). The aim of this thesis is to estimate the neighbourhood effect on old population in Turin with a propensity score approach. To achieve this goal, we need to work on adapting propensity score techniques to work well in a framework with many treatments (with ten or more treatments). Data used in the thesis come from the Turin Longitudinal Study (SLT), described in chapter 3. Our main goal is to understand if the observed differences in health outcomes across neighbourhoods can be causally attributed to neighbourhoods' as opposed to their different composition, i.e. to the fact that individuals with different risks factors live in different areas. In order to adjust for confounders and simulate an experimental approach, we focused on propensity score techniques that are briefly described in chapter 2. The first part of the analysis focuses on the performance evaluation of an inverse probability of treatment approach (IPTW) in a 10-treatment framework (chapter 4) and its application on real data on two different health outcomes: hospitalized fractures and mental health (chapter 5). In the second part of the thesis we propose a novel method that consists on a matching based on partially ordered sets (poset) (chapter 6). The Matching on Poset based Average Rank for Multiple Treatment (MARMoT), tested with some simulations, has revealed to be really useful to estimate neighbourhood effect, reducing bias of estimates because of the initial improvement of covariates' balance between groups.

The purpose of this deliverable is to present the PERCEIVE study’s quantitative dataset, drawn from secondary, freely available sources. The general idea of task 2.1 is to develop a database of consistent, comparable and timely available data aimed at collecting, standardizing and analyzing data which supports all the qualitative-quantitative analyses and WPs of the project. This task is connected with Work Package 1, and is to be integrated with the original data collected in this project in task 1.3 (EU citizen survey) on perceptions of structural and cohesion funds and reasons why these funds may or may not be supported

BACKGROUND According to recent literature the increasing women's labour market participation is only the first part of the so called gender revolution, while a second part is now unfolding, with an increased participation of men in family life with special attention to childcare. OBJECTIVE The aim of this paper is to explore fathers' involvement in parenting tasks within different contexts in terms of gender regimes, family policies, and workplace culture. The idea is to evidence individual factors that may enable/challenge the capability of fathers to stay with children and care for them, and to suggest opportune father-friendly policies. METHODS Time with children is compared among a sample of fathers in Time Use survey in France (2009-2010), Italy (2008-09), Sweden (2000-2001) and the UK (2000). Three different measures of father involvement are examined: the total time father spend with their children, the time they spend alone with them, and their engagement in childcare activities. RESULTS Results show that distinct micro-level factors contribute in determining the three levels of father's commitment analysed. Few cross-countries differences emerge. Fathers' involvement is mainly determined by their work-related features, by their children characteristics, and by their partner's working schedules. Weekday and weekend differences are observed. The quantum of father engagement strongly depends on the countries' institutional context: it is the highest in Sweden and the lowest in Italy. CONTRIBUTION This comparative study shows the methodological importance of considering different measures of father involvement to understand how micro-level factors influence the time fathers spend with their children in different institutional context.

Despite a long tradition in the study of graphs and relational data, for decades the analysis of complex networks was limited by difficulties in data collection and computational burdens. The advent of new technologies in life sciences, as well as in our daily life, has suddenly shed light on the many interconnections that our world features, from friendships and collaborations between individuals or organizations, to functional couplings between cellular molecules. This has highly facilitated the collection of relational data, fostering an unprecedented interest in network science. Understanding relations encoded in complex networks, however, still represents a challenging task, and statistical methods that can help to summarize and simplify complex networks are needed. In this thesis we show that often one can gain a deep insight of a network by focusing their attention on communities, i.e. on clusters of nodes, and on the relations that exist between them. We begin by presenting NEAT, a network-based test that allows to assess relations between gene sets in a gene interaction network. NEAT extends traditional gene enrichment analysis tests by incorporating information on interactions between genes and it overcomes some limitations of existing network enrichment analysis approaches. Then, we propose two extended stochastic blockmodels that allow to infer the relations that exist between communities from relations between pairs of individuals in a social network. We advocate the use of penalized inference to estimate these models, with the aim of deriving a sparse reduced graph between communities. Application of these models to bill cosponsorship networks in the Italian Chamber of Deputies allows us to reconstruct the pattern of collaborations between Italian political parties from 2001 to 2015. Finally, we propose a novel clustering strategy for sequences of graphs, based on mixtures of generalized linear models. We show that the proposed clustering method not only is capable to retrieve subpopulations of networks within a cross-sectional or longitudinal sequence of networks, but it also allows to directly characterize them by considering each of the components that form the mixture model.

Given a big set of data with several variables, the aim is the evaluation of each unit with a method that produces a synthetic measure describing a complex and non-observable concept; this goal is achieved respecting the characteristics of the variables, specially the measurement scale. The information gathered with partially ordered sets (poset) reflects this aim, because posets depends only on the order relations among the observations, and allows to handle ordinal and dichotomous variables fairly. In this setting, the vector of variables observed on a unit is handled as a unique entity called profile and not as a group of different variables that need to be aggregated. Starting from recent developments in poset theory, this thesis is organized in three parts. The first proposes to obtain a unique indicator combining the values given by the severity measures for evaluation, derived from the fuzzy identification method. The second contribution is the HOGS (Height Of Groups by Sampling) procedure, which is aimed to estimate the average rank of groups of units of a big population. HOGS is a step forward the statistical estimation of the average rank of a profile; furthermore it allows the estimation of the effect of external variables on the synthetic measure. The last results are two new R functions: the first computes the approximated average rank for large data sets overcoming the usual sample sizes considered by the software usually used until now, the second implements the information given by the frequency of profiles in the computation of approximated average rank, making its use more profitable for social sciences.

Measuring individuals' preferences for goods and services has recently obtained considerable attention in both public and private contexts. Individuals' judgments are used for many different purposes, including setting social policies and forecating the acceptance of a new product in the market. While preference modeling is a long-studying problem, modern applications, related to the web, make it an actual topic. Respondents are called to express their preferences among a set of alternatives and collected data can be represented in various kinds of matrices. This thesis is focused on some popular methods to estimate either scores or ranks of a set of alternatives by analyzing a generalized tournament matrix. The proposed methods are compared via simulation and some special situations are investigated to detect their reliability. Our aim is to compare methods that assume parametric hypotheses on data distribution with methods that do not require such hypotheses. When respondents do not compare directly two alternatives, the matrix representing their preferences may show one or more missing values. We propose a method to estimate the missing entries of a generalized tournament matrix based on the minimization of the sum of its singular values, i.e. the nuclear norm. We perform some simulation studies to investigate the nuclear norm minimization effectiveness.

Item response theory parameters have to be estimated, and because of the estimation process, they do have uncertainty in them. In most large-scale testing programs, the parameters are stored in item banks, and automated test assembly algorithms are applied to assemble operational test forms. These algorithms treat item parameters as fixed values, and uncertainty is not taken into account. As a consequence, resulting tests might be off target or less informative than expected. In this article, the process of parameter estimation is described to provide insight into the causes of uncertainty in the item parameters. The consequences of uncertainty are studied. Besides, an alternative automated test assembly algorithm is presented that is robust against uncertainties in the data. Several numerical examples demonstrate the performance of the robust test assembly algorithm, and illustrate the consequences of not taking this uncertainty into account. Finally, some recommendations about the use of robust test assembly and some directions for further research are given.

The aim of this thesis is to apply multilevel regression model in context of household surveys. Hierarchical structure in this type of data is characterized by many small groups. In last years comparative and multilevel analysis in the field of perceived health have grown in size. The purpose of this thesis is to develop a multilevel analysis with three level of hierarchy for Physical Component Summary outcome to: evaluate magnitude of within and between variance at each level (individual, household and municipality); explore which covariates affect on perceived physical health at each level; compare model-based and design-based approach in order to establish informativeness of sampling design; estimate a quantile regression for hierarchical data. The target population are the Italian residents aged 18 years and older. Our study shows a high degree of homogeneity within level 1 units belonging from the same group, with an intraclass correlation of 27% in a level-2 null model. Almost all variance is explained by level 1 covariates. In fact, in our model the explanatory variables having more impact on the outcome are disability, unable to work, age and chronic diseases (18 pathologies). An additional analysis are performed by using novel procedure of analysis :"Linear Quantile Mixed Model", named "Multilevel Linear Quantile Regression", estimate. This give us the possibility to describe more generally the conditional distribution of the response through the estimation of its quantiles, while accounting for the dependence among the observations. This has represented a great advantage of our models with respect to classic multilevel regression. The median regression with random effects reveals to be more efficient than the mean regression in representation of the outcome central tendency. A more detailed analysis of the conditional distribution of the response on other quantiles highlighted a differential effect of some covariate along the distribution., Lo scopo di questa tesi è quello di applicare il modello di regressione multilivello nel contesto di indagini sulle famiglie. La struttura gerarchica in questo tipo di dati è caratterizzato da numerosi piccoli gruppi. Negli ultimi anni analisi comparative e multilivello sullo stato di salute percepito sono aumentate molto. L’obiettivo di questa tesi è di applicare un'analisi multilivello a tre livelli per la variabile risposta Physical Component Summary allo scopo di: valutare entità all'interno e tra varianza ad ogni livello (individuale, familiare e comune); indagare quali covariate influiscono sulla percezione dello stato di salute fisica ogni livello; confrontare le analisi model-based e di design-based al fine di stabilire se i pesi campionari siano informativiti per il modello di interesse; stimare una regressione quantile per i dati gerarchici. La popolazione target sono i residenti italiani di età compresa tra 18 anni. Il nostro studio rileva un’elevata omogeneità tra le unità di livello 1 e una correlazione intraclasse del 27% nel modello nullo a 2livelli. Quasi tutta la varianza è spiegata dalle covariate di livello. Nel nostro modello le variabili esplicative hanno un impatto maggiore sulla variabile risposta sono la disabilità, inabilità al lavoro, l’età e le malattie croniche (18 patologie). Un'ulteriore analisi viene eseguita utilizzando una nuova procedura di analisi: "Regressione lineare quantile multilivello”. Questa analisi ci dà la possibilità di descrivere più in generale la distribuzione condizionata della variabile risposta attraverso la stima dei suoi quantili. Questo ha portato un grande vantaggio nei nostri modelli rispetto al classico modello di regressione multilivello. La regressione mediana con effetti casuali si rivela più efficiente del regressione media nella rappresentazione della tendenza centrale. Un'analisi più dettagliata della distribuzione condizionata della variabile risposta in corrispondenza di altri quantili ha evidenziato che certe covariate hanno un effetto diverso lungo la distribuzione.

2009 - 2010 Il tema della valutazione emerge periodicamente nella discussione politica italiana. L’azione del valutare, che rappresenta ormai un’operazione ricorrente nella vita quotidiana, diviene una pratica consolidata anche in seno alle istituzioni pubbliche, indispensabile per costruire un giudizio sul funzionamento delle politiche stesse. La pratica valutativa si rivela, però, difficile da applicare in contesti complessi e dinamici come quelli che caratterizzano gli interventi nel sociale, dove la complessità attiene alla eterogeneità e pluralità di attori coinvolti e alla multiproblematicità dei bisogni territoriali. Quando la riuscita di una politica di intervento dipende non solo dalle capacità di coordinamento dall’alto, cioè di chi programma gli interventi sociali e offre i servizi per rispondere ai bisogni di una comunità, ma anche dalla volontà e dalla partecipazione dal basso, cioè di chi fruisce degli interventi, risulta chiaro quanto un processo di valutazione diventi complesso. In tali situazioni, le tecniche della Social Network Analysis (di seguito analisi delle reti sociali) risultano particolarmente adatte a rilevare, studiare ed interpretare le interazioni di tutti gli attori coinvolti in uno o più interventi di politica sociale. Tali tecniche di analisi vengono utilizzate sempre più spesso nella ricerca valutativa, in quanto si presuppone che ci possa essere una relazione fra le caratteristiche della rete, costituita dagli attori sociali coinvolti nell’attuazione di un programma, e l’efficacia del programma stesso. [a cura dell'autore] IX n.s.

In testing, item response theory models are widely used in order to estimate item parameters and individual abilities. However, even unidimensional models require a considerable sample size so that all parameters can be estimated precisely. The introduction of empirical prior information about candidates and items might reduce the number of candidates needed for parameter estimation. Using data for IQ measurement, this work shows how empirical information about items can be used effectively for item calibration and in adaptive testing. First, we propose multivariate regression trees to predict the item parameters based on a set of covariates related to the item-solving process. Afterwards, we compare the item parameter estimation when tree-fitted values are included in the estimation or when they are ignored. Model estimation is fully Bayesian, and is conducted via Markov chain Monte Carlo methods. The results are two-fold: (a) in item calibration, it is shown that the introduction of prior information is effective with short test lengths and small sample sizes and (b) in adaptive testing, it is demonstrated that the use of the tree-fitted values instead of the estimated parameters leads to a moderate increase in the test length, but provides a considerable saving of resources.

In this paper, empirical prior information about the candidate is applied in computerized adaptive testing (CAT). The main objective of CAT is to improve efficiency of test administration. In this paper, it is shown how the inclusion of background variables both in the initialization and the ability estimation is able to improve the accuracy of ability estimates. In particular, a Gibbs sampler scheme is proposed in the phases of interim and final ability estimation. By using both simulated and real data, it is demonstrated that the method produces more accurate ability estimates, especially for short tests and when reproducing boundary abilities. This implies that operational problems of CAT related to weak measurement precision under particular conditions, can be reduced as well. In the empirical example, the methods were applied to CAT for intelligence testing in the area of personnel selection. Other promising applications would be in the medical world, where testing efficiency is of paramount importance as well.

Il lavoro di ricerca qui proposto si inserisce nel filone delle ricerche sulla mobilità intragenerazionale appena esposto. Questo tema è ampiamente affrontato negli studi che si occupano dell’analisi del mercato del lavoro e dei suoi cambiamenti avvenuti nel corso del secolo scorso. Tuttavia, la gran parte della letteratura si concentra sull’ingresso o sull’uscita delle donne dal mondo lavorativo tralasciando lo studio delle loro effettive possibilità di sperimentare un cambiamento di posizione occupazionale. Rispetto agli studi già condotti in questo ambito, riteniamo possa essere interessante aggiungere all’analisi delle carriere femminili anche altri aspetti. Un primo ambito di analisi è l’influenza della posizione occupazionale dei mariti e dei padri sulla carriera delle donne. Un altro ambito riguarda l’influenza di due eventi familiari di particolare rilevanza, quali il matrimonio e la nascita di un figlio, sull’abbandono del mondo del lavoro e sulla chance di cambiamento della posizione occupazionale delle donne. Con il primo interrogativo di ricerca, ci proponiamo di indagare se, nel corso del ventesimo secolo, si siano modificati i modelli di mobilità di carriera in Italia. La seconda domanda di ricerca cercherà invece di capire come gli eventi familiari - matrimonio e nascita di un figlio - influiscano sulla partecipazione delle donne al mercato del lavoro. Infine, con la terza domanda di ricerca, ci chiediamo come gli eventi sopra esposti influenzino le opportunità di carriera delle stesse. Per rispondere agli interrogativi indicati, prenderemo in considerazione lo studio sia dei fattori inerenti la dimensione individuale, sia di quelli specifici riguardanti l’ambito strutturale. L’analisi di questa ultima sfera permetterà di porre attenzione sui cambiamenti che hanno interessato, a partire dagli anni cinquanta del secolo scorso, il sistema produttivo italiano e le modifiche riguardanti i meccanismi di regolamentazione nei rapporti di impiego. Per quanto riguarda, invece, la dimensione individuale, ci si propone di studiare in profondità alcune caratteristiche sociodemografiche dei soggetti, come il genere, la coorte di nascita e il titolo di studio: quelle che gli studiosi di questo ambito ritengono fondamentali per capire come si strutturano le disuguaglianze. Per raggiungere il nostro obiettivo è stata presa in considerazione l'Indagine Longitudinale delle Famiglie Italiane (ILFI 1997-2005) e sono stati utilizzati una pluralità di strumenti statistici, quali tavole di mobilità, tecniche di event history analysis, modelli log-lineari e modelli di regressione logistica. Rispetto al primo interrogativo di ricerca, il quadro che emerge dalle nostre analisi mostra che nel nostro paese sono presenti contenuti tassi di mobilità di carriera e che nel corso delle coorti sia possibile rilevare solo una lieve variazione rispetto alle opportunità di cambiamento di classe per gli individui. Prendendo in considerazione il secondo interrogativo proposto, abbiamo potuto constatare come siano soprattutto le donne maggiormente istruite a rimanere nel mercato del lavoro, anche in occasione del matrimonio e della nascita di un figlio. La possibilità di lavorare solo per una parte della giornata (usufruendo, quindi, di contratti part-time) sembra leggermente favorire la conciliazione tra famiglia e lavoro. All'opposto, abbiamo riscontrato una maggiore propensione all'abbandono nei casi in cui le garanzie legislative del lavoro vengono a mancare. La posizione lavorativa del partner ha mostrato di avere un importante ruolo. Così, il vantaggio relativo del marito sulla moglie porta quest'ultima ad abbandonare il mercato del lavoro per occuparsi, con molte probabilità, al lavoro domestico. Infine è apparso evidente che la nascita del primo figlio ha una forte spinta a uscire dal mercato.

Nella presente analisi si è avuta l’eccezionale disponibilità di dati longitudinali su molti individui (6000studenti frequentanti le scuole superiori bolognesi). Per ottenere un modello che al meglio spiegasse e riproducesse l’andamento degli esiti scolastici (promozione alla classe successiva) tenendo conto del percorso scolastico nel tempo, si è scelto il modello a curva latente condizionato. La variabile risposta è combinazione lineare dell’esito di fine anno (Promosso/Non promosso); riassume, per ogni studente/anno scolastico, classe frequentata ed esito finale. Le variabili esplicative sono state selezionate tra le informazioni disponibili per gli individui. Vengono presentati alcuni dati aggregati, poi descritti i dati individuali che entreranno nel modello, evidenziando la composizione degli studenti. La prima fase è la stima del modello logistico, con specificazione delle criticità, che hanno indotto alla scelta successiva del modello dipendente dal tempo. Dopo la descrizione della metodologia principale utilizzata, la teoria del conditionalLCM, e la selezione degli indicatori di fitting, viene delineata la procedura di stima, e raggiunto il modello ottimale. Le variabili significative per spiegare l’andamento delle promozioni individuali nel tempo risultano: cittadinanza (italiani con risultati significativamente migliori degli stranieri), sesso (ragazze con percorso scolastico mediamente migliore dei ragazzi: la differenza risulta però significativa soltanto negli istituti tecnici e professionali), tipologia di scuola frequentata (studenti del liceo con risultati significativamente migliori di chi frequenta altri tipi di istituto). I risultati risultano fortemente dipendenti dai dati impiegati, specie riguardo al limite territoriale. Precedenti analisi evidenziavano una forte differenziazione dei risultati scolastici tra studenti del nord e del sud Italia, oltre che tra studenti dei comuni maggiormente popolati e studenti dei comuni di provincia. Sarebbe interessante disporre di dati individuali analoghi a quelli qui utilizzati, ma riferiti all’intero territorio nazionale, oppure ad un zona maggiormente vasta dell’Italia, onde saggiare l’entità dell’influenza sul percorso scolastico,ed in particolare sulla regolarità, della differenza territoriale.

In this paper, we investigate the pattern of longevity during the last 15 years in Emilia Romagna, a North-Eastern region of Italy, at a municipality level. We consider a specific index of extreme longevity based on people aged 95 and over in two different periods (1995-1999 and 2005-2009). Spatio-temporal modeling is used to tackle at both periods the random variations in the occurrence of people 95+, due to the increasingly rareness of such events, especially in small areas. This method exploits the spatial proximity and the consequent interaction of the geographical areas to smooth the observations, as well as to control for the effects of a set of regressors. As a result, clusters of areas characterized by high and low indexes of longevity are well identified and the temporal evolution of the phenomenon can be depicted. In a parallel analysis, we consider the past levels of mortality on the same cohort of individuals reaching 95 years and over in the second period and when they were aged 80-89 and 90-99. Within this longitudinal framework, the longevity outcome is modeled by a spatial regression. The area-specific structures of mortality are included as regressors, whose effects represent the causal link between the occurrence of people 95+ and the causes of death in the same cohort.

Il lavoro si propone di ripercorrere alcune metodologie generali di analisi dei test per la valutazione degli apprendimenti, discutendo i risultati ottenuti in due casi di studio riguardanti le prove preparate dal Servizio Nazionale di Valutazione (SNV) dell’INVALSI per la classe seconda della scuola primaria. In particolare, viene descritto il processo di analisi dei pre-test attraverso l’utilizzo congiunto degli indicatori derivanti dalla Classical Test Theory e dei modelli di Item Response Theory.

The paper deals with the investigation of gender differences in performances in mathematics for Italian students at the end of lower secondary school in the Emilia-Romagna region. The study is based on a new large-scale assessment test developed and administered by the National Evaluation Institute for the School System (INVALSI). Given the evidence in the literature which favours males, performances of female and male students are compared using different approaches. In particular, scores proposed by educational experts based on item subgroups were considered, while a model-based approach was used within item response theory to estimate ability scores. The results revealed a significant advantage to males compared with females in overall performance, while no meaningful differences were observed with respect to item domain and item type. Following this, an interpretable item map was developed crossing expert opinions with IRT abilities, and plausible proficiency levels were defined. According to the map-based student classification, a relatively lower percentage of females fell into the highest proficiency groups with respect to males.

Il rapporto presenta i risultati dell'analisi statistiche effettuate sulle risposte date dagli studenti al'Esame di Stato della scuola secondaria di primo grado, classe terza, anno 2007-2008. I dati forniti dall'Invalsi riguardano le risposte fornite alle prove standardizzate di italiano e matematica.

In this paper, empirical prior information is introduced in computer adaptive testing. Despite its increasing use, the method suffers from a weak measurement precision, especially under particular conditions. Therefore, it is shown how the inclusion of background variables both in the initialization and the ability estimation is able to improve the accuracy of ability estimates. In particular, a Gibbs sampler scheme is proposed in the phases of interim and final ability estimation. By using simulated data, it is demonstrated that the method produces more accurate ability estimates, especially for short tests and when reproducing boundary abilities.

When the goal of a study is to assess and compare the evolution of old mortality, the researchers often face with problems of defects in data, such as loss of accuracy and numerical inconsistency of information. In the present work, an adaptive procedure for the issue of survival estimations is proposed as a combination of several methods: spline interpolation to split grouped data into age-specific quantities; use of mortality models for estimating the risk of death at advanced ages; methods for choosing the fitting age range based on the analysis of life-table aging rates. The procedure is applied to Emilia-Romagna data in order to construct complete period life tables corresponding to the occurrence of each population census from 1871 to 2001, and, finally, to compare the measures of old-age mortality by time and sex.

This article reports findings from a survey addressed to measure university students’ quality of life in Cagliari. The aim to build up a synthetic indicator of students’ ‘quality of life at the university’ has been pursued by adopting an ad hoc modeling approach to scale ordered items (Item Response Models) which belongs to the family of the Generalized Linear and non Linear Mixed Models . The sensibility of the results has been deeply analyzed by setting up several models with different characteristics. A comparison study with other scaling methods has been made.

Nel 2004 il Sistema bibliotecario, in collaborazione con l'Osservatorio statistico, ha effettuato una rilevazione sulla soddisfazione degli utenti tramite questionario-intervista. L'indagine è stata condotta nelle otto biblioteche centralizzate della sede di Bologna e in quattro delle maggiori biblioteche di dipartimento, una per ognuna delle aree disciplinari dello SBA (biosfera, scientifico-tecnica, sociale e umanistica). Allo scopo di raccogliere le valutazioni di un campione rappresentativo dell'utenza delle dodici biblioteche nel medesimo periodo di calendario, la rilevazione è stata interamente condotta nella settimana compresa fra il 22 ed il 26 novembre 2004. Il questionario di indagine, predisposto dallo SBA, è stato somministrato e compilato direttamente dai rilevatori (intervista face to face). Complessivamente, sono state effettuate nei 5 giorni della rilevazione oltre 3.600 interviste, a fronte di un'utenza giornaliera che nelle biblioteche interessate è stimata prossima alle 2.400 unità. I primi mesi del 2005 sono stati dedicati alla scannerizzazione delle schede raccolte, al successivo controllo e correzione dei risultati della lettura ottica, all'elaborazione ed alla produzione della documentazione statistica. I risultati complessivi sono stati illustrati nel corso di una conferenza stampa il 20 maggio e sono raccolti in questo rapporto, mentre quelli relativi ad ogni singola biblioteca sono stati inviati ai responsabili della struttura.

On the 20th of April 2005 there was a presentation on the statistical report of graduates at the Chamber of Commerce, Milan titled "An analysis on the resumes of graduates". This statistical report of the graduates is the result of a project (named STELLA an acronym which stands for "Statistiche sul TEma Laureati & Lavoro in Archivio on-line") whose aim was to obtain a complete picture of the profile of the graduates in relation to their positions in the job market.

The article contains a broad description of the tourist trend in the Emilia-Romagna region during the years 2001-2003. Official Istat data are used to place the regional tourism in the national outline, while data from provincial and regional administration are handled to find the specific characteristics of the phenomenon in the region. The analysis is based on structural indicators (hotels and complementary structures capacity) and flux indicators (arrivals, presences and mean time of stay) with particular reference to guest’s nationality. National and international tourist fluxes are analysed considering the kind of structure used, the kind of tourist resort, destinations and seasonality. -------------------------------------------------------------------------- --------------------------- Il lavoro descrive l’andamento dell’offerta e della domanda nel settore turistico dal 2001 al 2003 all’interno della regione Emilia-Romagna, con particolare accento alle modifiche più rilevanti avvenute in questo triennio. Dopo aver collocato il turismo regionale nel quadro nazionale e rilevato i principali provvedimenti legislativi regionali in materia di turismo, si propone inizialmente un’analisi territoriale disaggregata a livello provinciale. Un secondo punto di vista concerne le origini dei flussi turistici italiano ed estero, e quindi le nazionalità dei clienti ospitati nelle strutture ricettive dell’Emilia-Romagna in relazione alla scelta dell’esercizio utilizzato, al tipo di località turistica visitata, alle diverse destinazioni prescelte e alla stagionalità dei flussi. tourism, demand, offer, arrivals, presences, mean time of permanence

In this work we analyse ERSU’s quality of services about residence, food and cultural services or for playtime. Customer satisfaction approach is used to measure quality.

Household vulnerability relates to the incapacity of a household to preserve its welfare because of negative events. In Bosnia and Herzegovina, vulnerability is a central issue in the policy debate: many households are at risk of poverty due to fragile livelihood systems or high exposure to shocks. Since no panel data are available in Bosnia and Herzegovina, a micro-simulation methodology is adopted on the basis of a household consumption model based on quintile regression. Shocks are simulated in the labour market and the methodology consents the identification of the typologies of households which suffer from severe welfare losses or that are plunged into poverty after the shocks. The results show that the vulnerable households identified change when different shocks or a diverse definition of vulnerability are taken into consideration. At the same time, the characteristics of households vulnerable to poverty are shown to be different with respect to those of the poor.

We investigate the timing of adoption of product and process innovation using a differential game where firms may invest in both activities. We consider horizontal product innovation that reduces product substitutability, and process innovation that reduces marginal cost. First, we demonstrate that the incentive for cost-reducing investment is relatively higher than the incentive to increase product differentiation. Second, depending on initial conditions: (i) firms activate both types of investment from the very outset to the steady state; (ii) firms initially invest in only one R&D activity and then reach the steady state either carrying out only this activity or carrying out both; or (iii) firms do not invest at all in either type of innovation. Comparing R&D investments under Cournot and Bertrand behavior shows that quantity competition entails lower R&D incentives than price competition in both directions.

A critical variable in determining university students' performances is to get at least one positive result in a first year final examinations. The success at a final exam is likely to be connected with the attendance at course lectures. The aim of this paper is to give some descriptive information on the attendance of lectures and final performances with respect to a large set of first year students of Statistics course at the Faculty of Political Sciences at University of Cagliari.

Bio-demographical studies on longevity in Sardinia have resulted in some important findings, such as confirmation of high prevalence and low female/male ratio in validated centenarians (Poulain et al., 2004). Several issues however remain still open. For example, what are the biological mechanisms determining extraordinary life spans in some individuals, and moreover what is the connection between longevity and extrabiological factors such as environment, lifestyle, nutrition. In this paper, using census data, we try to trace the historical relevance of longevity in the Sardinian population in the last two centuries. From two censuses (1844 and 1848), it emerges that in Sardinia, also in the first half of XIX century - despite the high mortality rates - there are several recorded cases of longevity in individuals and population groups. Even if the percentage of people who lived to advanced age is small, the presence of nonagenarians was significant and was in proportion higher when compared to that of centenarians recorded in the past few decades. Starting from these findings, we have tried to compare the geographical distribution of long- lived individuals in the XIX and in XX century. The longevity maps show a certain continuity in a time period of 150 years in the areas characterized by the presence of "very old" individuals in the Island. The areas located include mountainous and hilly internal areas, which have been described already in previous studies as particularly "closed" areas in terms of marital patterns (eso/endogamy). Moreover, it seems that the presence of "very old" individuals in the XIX century was more widespread throughout the territory than today. An interesting hypothesis is that a genetic factor favourable to longevity was present in greater measure in the Sardinian population in the past, and that at the present time this genetic component is only found in the more isolated and endogamous areas. Finally, our research, notes the lower sex ratio value (F/M) among Sardinian long lived people, particularly in some areas of the province of Nuoro and Oristano.

In this work we analyse quality of life of university’s students according to mixed approach (objective-subjective). We propose a new mixed indicator (ISG’) and analyse with a GLM regression model the influence of some important covariates: residence, faculty, sex and academic situation.

Starting from evidence advanced in previous research, the Authors, carry out an analysis on a sample of school pupils. They were asked to indicate what things in their opinion are more or less important in life.

The aim of this paper is to give a simple introduction to the bootstrap techniques showing a basic computer algorithm. The algorithm displays, step by step, how to determinate a bootstrap confidence interval.

The authors, in addition to underlining further possibilities for using the ecclesiastical source Quinque Libri, explore data that is possible to extract from the inventory of the book of confirmations. They do that with reference to four dioceses of Central-South Sardinia. Such an exploration has led the authors to understand how the book of confirmations, for the period before the Unity of Italy, could be a useful support for demo-social surveys as well as a source of information for studies concerning to the knowledge and the activities of Sardinian bishops.

In this paper some factors affecting the spatial and diachronic variability of the infant mortality in Italy are discussed together with the age structure and the seasonal pattern of infant deaths in Sardinia. The first official statistics after unification of Italy, in the nineteenth century, revealed the existence of significant regional differences in the levels of infant and child mortality. Sardinia seems to have been in better conditions since his infant mortality rate appears to be lower compared to all the other Italian regions. This advantage depends crucially on the very low risk for new-borns to die in the first month. However, after that period, between 1 and 12 months, the mortality rate in Sardinia was higher than the one referred to Italy average. A similar pattern of low neonatal and high post-neonatal mortality was observed in some areas of Great Britain (west Cornwall and south-Wales) as well, but it remains difficult to explain.