https://hdl.handle.net/10807/218 Fri, 11 Sep 2026 16:03:12 GMT 2026-09-11T16:03:12Z https://hdl.handle.net/10807/346188 Titolo: Identifying corruption risk profiles in public procurement over emergency periods: a latent class - Item Response Theory approach Autori: Del Sarto, Simone; Gnaldi, Michela; Salvini, Niccolo Abstract: Frequent crises require adapting quality assessment systems for public and private services to ensure timeliness, equity, and sustainability while addressing corruption risks. Public procurement, a key component of crisis response, is particularly vulnerable to corruption due to regulatory relaxations, increased spending, and dynamic market conditions. This work relies on an approach that integrates traditional red flag detection with a supervised machine learning approach to identify corruption risks based on historical data and statistical testing. Using a big data source on Italian tenders and a multidimensional Item Response Theory model, we analyze public procurement during the Covid-19 crisis. Awarded companies are classified into subgroups based on their corruption risk profiles, accounting for latent and multidimensional risk factors. The findings contribute to improving corruption risk assessment in crisis-driven procurement systems, supporting more effective and transparent governance. Wed, 01 Jan 2025 00:00:00 GMT https://hdl.handle.net/10807/346188 2025-01-01T00:00:00Z https://hdl.handle.net/10807/346187 Titolo: Assessing and Improving Data Quality in Open Spatial Data: a case study with ANAC Data Autori: Salvini, Niccolo; Nardelli, Vincenzo Abstract: In this paper, we focus on assessing data quality in the context of open anti-corruption data, using data from the National Italian Anti-Corruption Authority (ANAC). The open data movement promotes governments to publish data sets to enhance transparency and accountability, which has been particularly beneficial in combating corruption. Nonetheless, open data is not exempt from challenges, one of which is data quality. We investigate missingness of the data to determine if it is missing at random, not at random, or completely at random. We then present a data quality algorithm, specifically designed for ANAC data, to sanitize errors. To further investigate the phenomena of missingness, we enriched the dataset with socio economic indicators at municipality level coming from official sources (such as ISTAT and Ministry of Economy and Finance, etc.). Finally, we use modelling to determine the factors that contributed to the missingness. An important addition to the classical modeling approaches widely used in literature is to assess if the missingness depends also on the geolocalization of the municipality. This is carried on testing whether it exists an autocorrelation on residuals not explained by classical methods. Our results indicate that addressing missing data with the proposed methodology can lead to more accurate and reliable data for anti-corruption assessments. This study contributes to literature on data quality assessment and provides insights into the challenges and potential solutions to missing data in open data initiatives exploiting spatial statistics techniques. Sun, 01 Jan 2023 00:00:00 GMT https://hdl.handle.net/10807/346187 2023-01-01T00:00:00Z https://hdl.handle.net/10807/346184 Titolo: Food Prices and Household Vulnerability in Italy: Insights from Web-Scraped Data Autori: Benedetti, Ilaria; Laureti, Tiziana; Salvini, Niccolo; Palumbo, Luigi Abstract: This paper investigates food price dynamics and inflation inequality in Italy by leveraging web-scraped data from major retailers. In the aftermath of the COVID-19 pandemic, surging food prices have disproportionately affected lower-income households, which allocate a greater share of their expenditure to essential goods. Drawing on over 5 million high-frequency price observations from 2021 to 2023, we examine within-category price variations across regions and time for staple food items, with a focus on Milk and Olive Oil. Using a modified version of the Time-Interaction-Region Product Dummy (TiRPD) model, we construct sub-national price indices across different price deciles, enabling a detailed assessment of inflation heterogeneity. Our results reveal that the lowest-cost items within each category experienced faster price growth compared to higher-priced counterparts—a dynamic known as “cheapflation”—thus limiting the ability of consumers to mitigate inflation by substituting down in quality. These patterns are consistent across product categories and regions. The findings underscore the need for timely and targeted policy interventions to mitigate the regressive effects of inflation and to protect vulnerable households. Wed, 01 Jan 2025 00:00:00 GMT https://hdl.handle.net/10807/346184 2025-01-01T00:00:00Z https://hdl.handle.net/10807/346183 Titolo: Can You Afford to Work? Transport Costs vs Income in Tuscany's Local Labour Systems Autori: Salvini, Niccolo; Secondi, Luca Abstract: This study investigates transport poverty in Tuscany’s Local Labor Systems (LLS) through the development of a Daily Transport Poverty Index (DTPI), which combines monetary and temporal commuting burdens. Leveraging real-time data from Google Maps and Michelin APIs, along with census commuting flows and income data, we evaluate accessibility for both private and public transport. Results show significant spatial disparities: while central LLS benefit from relatively efficient public transport, peripheral systems face a double burden–longer travel times and higher costs–when not using a private vehicle. The DTPI captures these imbalances, highlighting critical areas where public transport fails to provide a viable alternative. It is confirmed that transport poverty represents a structural barrier to labour market access, particularly for low-income workers in peripheral areas. The proposed index can support targeted policy interventions aimed at improving equity in regional mobility systems. Wed, 01 Jan 2025 00:00:00 GMT https://hdl.handle.net/10807/346183 2025-01-01T00:00:00Z