The use of different data sources in the analysis of co-authorship networks and scientific performance

Scientific collaboration is usually derived from archival co-authorship data. Several data sources may be examined, but they all have advantages and disadvantages, especially when a specific discipline or community is of interest. The aim of this paper is to explore the effect of the use of three data sources – Web of Science, Current Index to Statistics and nationally funded research projects – on the analysis of co-authorship networks among Italian academic statisticians. Results provide evidence of our hypotheses on distinct collaboration patterns among statisticians, as well as distinct effects of scientist network positions on scientific performance, by both Statistics subfield and data source.     

DO BETTER PAID POLITICIANS PERFORM BETTER? DISENTANGLING INCENTIVES FROM SELECTION

The wage paid to politicians affects both the choice of citizens to run for office and the performance of those who are appointed. First, if skilled individuals shy away from politics because of higher opportunities in the private sector, an increase in politicians’ pay may change their mind. Second, if the re‐election prospects of incumbents depend on their in‐office deeds, a higher wage may foster performance. We use data on all Italian municipal governments from 1993 to 2001 and test these hypotheses in a quasi‐experimental setup. In Italy, the wage of mayors depends on population size and sharply rises at different thresholds. We apply a regression discontinuity design to the only threshold that uniquely identifies a wage increase: 5,000 inhabitants. Exploiting the existence of a two‐term limit, we further disentangle the composition from the incentive component of the effect of the wage on performance. Our results show that a higher wage attracts more‐educated candidates, and that better‐paid politicians size down the government machinery by improving efficiency. Importantly, most of this effect is driven by the selection of competent politicians, rather than by the incentive to be re‐elected.     

Testing for Heterogeneity with Categorical Data: Permutation Solution vs. Bootstrap Method

In this article the problem of comparing distributional heterogeneities for categorical variables is addressed. Specifically, the one-sided testing problem for heterogeneity comparisons is considered. For such a problem a bootstrap method is proposed and compared with a permutation method already present in literature. The power behavior of the two methods is compared through a Monte Carlo simulation study. The results of two real applications are shown.     

Reliability and Validity of the Pathological Gambling Adaptation of the Yale-Brown Obsessive-Compulsive Scale (PG-YBOCS)

The Yale Brown Obsessive Compulsive Scale adapted for Pathological Gambling (PG-YBOCS) was developed to measure the severity and change in severity of pathological gambling symptoms. The PG-YBOCS is a 10-item clinician-administered questionnaire that measures the severity of PG over a recent time interval (usually within the past one/two week(s)). In order to assess and validate the scale, it was administered to 337 subjects: 188 pathological gamblers and 149 healthy controls. Internal consistency and correlations between individual items and total score were assessed for various permutations of the sample. Other scales were administered to assess convergent, discriminant and content validity. Sensitivity to change was evaluated in treatment studies with fluovoxamine, lithium, and valproate. Each item was frequently endorsed across a range of severity. Good inter-rater reliability and internal consistency were obtained. The PG-YBOCS showed high validity and reliability for total score, item-total correlations, and for each subscale (Thoughts/Urges and Behavior). PG-YBOCS scores correlated with global severity and South Oaks Gambling Screen (SOGS) scores. The scale was also sensitive to change in pathological gambling severity. PG-YBOCS thus appears to be a reliable and valid measure of pathological gambling severity, and can be regarded as an important tool for clinicians and researchers treating pathological gamblers.     

Presenting Results from Model Based Studies to Decision-Makers: Can Sensitivity Analysis be a Defogging Agent?

The motivation of the present work is to provide an auxiliary tool for the decision-maker (DM) faced with predictive model uncertainty. The tool is especially suited for the allocation of R&Dresources. When taking decisions under uncertainties, making use of the output from mathematical or computational models, the DM might be helped if the uncertainty in model predictions be decomposed in a quantitative-rather than qualitativefashion, apportioning uncertainty according to source. This would allow optimal use of resources to reduce the imprecision in the prediction. For complex models, such a decomposition of the uncertainty into constituent elements could be impractical as such, due to the large number of parameters involved. If instead parameters could be grouped into logical subsets, then the analysis could be more useful, also because the decision maker might likely have different perceptions (and degrees of acceptance) for different kinds of uncertainty. For instance, the decomposition in groups could involve one subset of factors for each constituent module of the model; or one set for the weights, and one for the factors in a multicriteria analysis; or phenomenological parameters of the model vs. factors driving the model configuratiodstructure aggregation level, etc.); finally, one might imagine that a partition of the uncertainty could be sought between stochastic (or aleatory) and subjective (or epistemic) uncertainty. The present note shows how to compute rigorous decomposition of the output's variance with grouped parameters, and how this approach may be beneficial for the efficiency and transparency of the analysis.     

A note on multiple testing for composite null hypotheses

Multiple hypothesis testing literature has recently experienced a growing development with particular attention to the control of the false discovery rate (FDR) based on p-values. While these are not the only methods to deal with multiplicity, inference with small samples and large sets of hypotheses depends on the specific choice of the p-value used to control the FDR in the presence of nuisance parameters. In this paper we propose to use the partial posterior predictive p-value [Bayarri, M.J., Berger, J.O., 2000. p-values for composite null models. J. Amer. Statist. Assoc. 95, 1127–1142] that overcomes this difficulty. This choice is motivated by theoretical considerations and examples. Finally, an application to a controlled microarray experiment is presented.     

Global imbalances, exchange rates adjustment and the crisis: Implications from network analysis

The persistence of global imbalances remains a potential source of instability for the international economic system, which ought to be tackled in order to exit from the current crisis in a better position. This paper shows that the complexity of the global network of trade imbalances has increased over time in terms of the number of countries involved and of the heterogeneity of their conditions. These findings imply that orchestrating a global exchange rate adjustment plan, as done in the 1980s to tackle a similar problem, is certainly more difficult than it was 20 years ago. We also show, however, that the uncoordinated movements in bilateral exchange rates occurred since the early 2000s did not go in the direction of reducing bilateral imbalances. Rebalancing the world economy, thus, seems to require a rich set of coordinated policy actions and economic changes, involving adjustments in both exchange rates and real variables, which are heterogeneous at the country level. We argue that, provided some conditions are satisfied, the variegated effects of the ongoing financial crisis, and the policy responses this has induced, may help the global and bilateral rebalancing process.     

Different Predictors of Quality of Life in Urban Environment

The present study aimed at comparing the effects of different sets of predictors on quality of life in an urban environment. We used secondary data collected by means of a self-report questionnaire on a sample of 343 residents of a big Italian city. The questionnaire included a multidimensional scale elaborated by the World Health Organization (WHOQoL brief scale) assessing quality of life in terms of four different evaluations concerning distinct aspects of life: physical health; psychological status; social relationships; environment. Four different types of predictors were considered: (1) socio-demographic characteristics; (2) quality of social relations (perceived social support); (3) place attachment; (4) healthy lifestyle. To test the influence of different groups of predictors on the dimensions of WHOQoL we performed four hierarchical regression analyses. Several significant influences were found. In particular the results pointed out the great role of perceived social support and place attachment in promoting quality of life. That result suggests the importance of community interventions in urban environment.