Volume 17 - 2022
Volume 16 - 2021
Volume 15 - 2020
Volume 14 - 2019
Volume 13 - 2018
Volume 12 - 2017
Volume 11 - 2016
Volume 10 - 2015
Volume 9 - 2014
Volume 8 - 2013
Volume 7 - 2012
Volume 6 - 2011
Volume 5 - 2010
Volume 4 - 2009
Volume 3 - 2008
Volume 2 - 2007
Volume 1 - 2006
Journal Home >
Volume 6, Issue 4 - December 30, 2011
JAQM Volume 6, Issue 4 - December 30, 2011
Editorial Letter (p )
Banking Services Evaluation: a Dynamic Analysis (p )
Today, the most important asset for a bank is its customer and therefore, the main targets to achieve by management are: knowledge of his needs, anticipation of his concerns and to distinguish itself in his eyes. The awareness that a satisfied customer is a highly profitable asset effort to provide a satisfactory service to the customer by diversifying its services.
This paper aims to analyze customer evaluation evolution of the main attributes of banking services to catch differences among the clusters and time lags through a dynamic factorial model.
We propose an empirical study: the management of a national bank with a spread network throughout Italy wanted to analyze its reduced competitiveness in retail services, probably due to low customer satisfaction. The survey aims to analyze weaknesses in its retail services, propose possible recovery actions and measure their effectiveness across different “waves” (time lags).
Measuring Risk Profile with a Multidimensional Rasch Analysis (p )
Sergio Ortobelli LOZZA
In this paper we propose an evaluation of investors’ risk profiles such as to meet the minimal requirements that Italian financial institutions must satisfy by law (d. lgs. 164, 2007). Thus we investigate all aspects specific to so-called risk profiles: an investor’s knowledge and his financial experience (concerning financial instruments and their use); financial objectives, a personal predisposition to risk /earn and the temporal horizon.
The methodology used in financial literature with regard to risk profiles is essentially based on simplistic statistical analyses that often fail to consider possible psychological aspects. In order to account for investor preferences and psychological attitudes, we suggest to use an item response theory model. We first assume a unidimensional model, belonging to the family of Rasch models and then, as an alternative approach, a Generalized Multidimensional Rasch model. In particular, the objective is to assess the value of a questionnaire whose items describe different characteristics of the main latent variable risk profile.
Under the assumption of a multidimensional measurement model, given the multivariate position of each investor with respect to identified latent traits we can represent his position with respect to possible investments proposed by a bank and we can identify different situations that respect the investor’s risk profile and best characterize typical investor choices.
Logistic Regression Response Functions with Main and Interaction Effects in the Conjoint Analysis (p )
Amedeo DE LUCA,
In the Conjoint Analysis (COA) model proposed here - an extension of the traditional COA - the polytomous response variable (i.e. evaluation of the overall desirability of alternative product profiles) is described by a sequence of binary variables. To link the categories of overall evaluation to the factor levels, we adopt - at the aggregate level - a multivariate logistic regression model, based on a main and two-factor interaction effects experimental design.
The model provides several overall desirability functions (aggregated part-worths sets), as many as the overall ordered categories are, unlike the traditional metric and non metric COA, which gives only one response function.
We provide an application of the model and an interpretation of the main and interactive effects.
Using Structural Equation and Item Response Models to Assess Relationship between Latent Traits (p )
We deepen the two main approaches to the problem of measurement error in social sciences, the Structural Equation Models (SEM) and the Item Response Theory Models (IRM), comparing two different estimation procedures.
The One-step procedure (related to SEM) requires that researcher specifies a complete model of both measurement aspects (single link between the latent variable and its indicators) and structural aspects (links between different latent variables), with the model parameters estimated simultaneously. In the Two-step procedure (related to IRM), we first estimate the measures (one for each construct), then we will assess, through a regression model, the relationships between these measures and the latent variables that they represent.
Our aim is to define a Two-step method that, using information obtained in the first step about the measurement error, presents low levels of bias and loss of efficiency, as close as possible to that of One-step method.
Statistical Models to Measure Corporate Reputation (p )
Reputation can be defined as how an entity (private or public) is perceived by each of its stakeholder groups and reputation risk as the risk that an event will negatively influence stakeholder perceptions. Since reputation involves intangible assets (public opinion, perception, reliability, merit), it is not simple to define and consequently to measure and to monitor the correlated risk. In this contribution we propose statistical models based on ordinal data aimed at measuring effectively reputation. The proposed models are applied to real data on Italian public companies taken from financial media corpora.
A Two-Level Structural Equation Model for Evaluating the External Effectiveness of PhD (p )
In recent years the number of PhDs in Italy has significantly grown and purposes of PhD courses have expanded from the traditional ones. The analysis of the contribution of PhD title for job placement and employment condition of PhDs is an important tool for evaluating the quality and the effectiveness of PhD courses. For this reason, knowledge of the employment status and career of PhDs becomes essential and can help to reduce the gap between academia and labour market. The aim of this paper is to estimate a two-level structural equation model with latent variables to assess the external effectiveness of PhD. The analysis is performed using data from the research "Current situation and employment prospects of PhDs", commissioned by National Committee for the Evaluation of the University System (CNVSU) to the Department of Statistics "G. Parenti" of the University of Florence. The proposed measure of "external effectiveness" is a latent variable obtained by evaluating the level of satisfaction with the employment status of PhDs who achieved the title in 2008. The opinion was expressed one year after obtaining PhD on a ten ordered point scale. External effectiveness indicators used are Consistency with studies, Utilization of the acquired skills and Compliance with the cultural interests.
Differential Variability of Test Scores among Schools: A Multilevel Analysis of the Fifth-Grade Invalsi Test using Heteroscedastic Random Effects (p )
The performance of a school system can be evaluated through the learning levels of the pupils, usually summarized by school mean scores. The variability of the mean scores among schools is rarely studied in detail, though it is a crucial issue especially in primary schools: in fact, a high variability among schools raises doubts on the capacity of the system to guarantee equal educational opportunities. To investigate the patterns of variability in Italy, we analyse data from INVALSI, the Italian national institute for the evaluation of the school system, which regularly carries out standardized tests to assess the learning levels of the pupils at various grades. We consider the mathematics test administered to fifth-grade pupils at the end of the 2008/2009 year, along with a pupil's questionnaire for measuring socio-economic factors. The analysis is performed using a random intercept linear model on the Rasch score of the mathematics test, with pupil-level errors depending on gender and school-level errors depending on the geographical area. The model includes several demographic and socio-economic explanatory variables and some compositional variables obtained as school means of pupil variables. The results show a considerable increase in the residual variance among schools when going from North to South, pointing out a serious issue of fairness in Southern Italy. The situation is mitigated by the finding that a substantial part of the residual variance among schools is due to a few schools with exceptionally positive results.
Impact of Educational Test Features on Item Difficulties by the Linear Logistic Test Model (p )
Carlo DI CHIACCHIO,
The aim of the paper is to investigate the effect of item and person properties on item difficulties using the Linear Logistic Test Model (LLTM) and its extensions. The data under investigation are the Italy mathematics data from the Program for International Student Assessment (PISA) 2006. The information regarding the geographical macro-area (North-Italy versus South-Italy) has been used in the application as person property. Furthermore, the comparison on item properties effects between North-Italy versus South-Italy is performed fitting the LLTM for each geographical macro-area.
Measures for Ph.D. Evaluation: the Recruitment Process (p )
In the last years the quality of Higher Education (HE) system and its evaluation have been key issues of the political and scientific debate on education policies all over Europe. In the wide landscape that involves the entire HE system we draw attention on the third level of its organization, i.e. the Ph.D. In particular, this paper discusses the necessity of monitoring the recruitment process of Ph.D. system because it represents a fundamental aspect of the Ph.D. system as a whole. We introduce a set of concepts related to the recruitment process and then we make them computable with synthetic indicators. The study provides an empirical analysis based on doctoral schools of four academic years at the University of Siena. Proposed indicators are finally used for detecting weakness and strength of each Ph.D. school.
Descriptive Analysis of Student Ratings (p )
Let X be a statistical variable representing student ratings of University teaching. It is natural to assume for X an ordinal scale consisting of k categories (in ascending order of satisfaction). At first glance, student ratings can be summarized by a location index (such as the mode or the median of X) associated with a convenient measure of ordinal dispersion. For instance, the median of X may be associated with the dispersion index of Leti, resulting in a synthesis that takes into account the ordinal nature of data and also communicates information in an effective way. More generally, there are many indexes (such as the ordinal entropy) that can be properly employed to measure the ordinal dispersion. On the other hand, student ratings are often converted into scores and treated as a quantitative variable. More generally, it is possible to measure student satisfaction by means of a real-valued function defined on the standard simplex and satisfying some appropriate conditions. Finally, such a measure of satisfaction can be associated with a suitable measure of variability.
The Italian Judicial Offices Productivity in Almost 130 Years of Cognition Civil Procedures (p )
In Italy, one of the hinge points on which the concept of State leans - the Justice - has slipped in a deep crisis more and more since remarkable difficulties in its internal reorganization are accompanied to the natural process of review in the civil society in evolution. The more evident external aspects of such crisis are translated in the slowness of the judicial mechanism, in the high cost of its antiquated procedures and in the difference of the sentences for degrees of judgment.
To comfort or to contradict this or that thesis, also, sometimes statistic data are brought in contrast from each other, because of what they define is not well specified. Wanting to give clarity, it is first of all necessary to delimit this analysis to the procedure of cognition, essential unit of the civil trial activity, for an objective knowledge of the phenomenon from the quantitative point of view, to be able to supply stable terms of reference for a better interpretation of the facts and a more serious search of the causes and the effects, reaching a suitable territorial distribution of the enquirer personnel, judging or not.
The analysis of some statistic indicators (i.e., the procedures duration, the index of disposal, the percentage variation of pending) derived by the data related to supervened, exhaustions and pending allows to estimate the productivity of the judicial offices in comparison to the justice demand. In the centennial oscillation of the civil procedures of cognition (and particularly of the relative quotients for 100.000 inhabitants), both in first degree and appeals, a growth is established, especially in the last twenty-thirty years, of supervened and exhausted procedures, and still more of those leaning that among the 1991/2000 decade and the average value of the last seven years go over the doubling. The average life of the civil procedures in every degree of judgment that on the contrary has gone growing since 1881 to today, even though with occasional lowering events. The civil procedures of cognition have reached by now the average duration of 3.000 days, and this means around eight years of waiting for the definitive sentence.
Safety at Work in Europe: An Efficiency Analysis (p )
Nowadays workplace accidents are more and more recognised as a social problem that has undesirable consequences on both human and organisations. As a result, there has been increasing concern in improving working conditions and in reducing occupational accidents in the European Union. In this context, this paper examines safety performance of fifteen European countries in four economic sectors – manufacturing, construction, distribution trades and transportation – by applying a frontier analysis method, Data Envelopment Analysis (DEA). A linear programming framework is therefore used to construct both constant and variable returns to scale (CRS and VRS, respectively) production frontiers which allow measurement of relative efficiency with respect to the number of workplace accidents.