Search

The 13th RISIS Policymakers Session, a crucial event focused on monitoring and analyzing research careers to guide policy making in the European Research Area, took place in Brussels on June 8th, 2023 in HYBRID FORMAT. The event aimed to bring together policymakers, researchers, and professionals from various fields to discuss the challenges and opportunities in understanding and supporting research careers. By exploring frameworks and strategies for data collection, the session provided valuable insights into shaping effective policies that foster research excellence. The Policymakers Session explored various aspects related to research careers and changing frameworks. It started with an introduction by Emanuela Reale, Director of IRCrES, CNR & Member of RISIS Governing Board, providing an overview of the RISIS projects and infrastructure. The first session, titled “Research Careers and Changing Frameworks”, featured presentations by Carolina Cañibano from CSIC, who discussed the conceptual framework for studying science research careers, and Dario Capezzuto from the European Commission, DG RTD, R&I actors and research careers who delved into the scenario of research careers in Europe. The second session, “Monitoring Research Careers,” covered topics such as data collection strategies and their combination in RISIS2, presented by Stephan Stahlschmidt from DZHW, and the development of the Doctoral Degree and Career Pilot Dataset (DDC), presented by Eric Iversen from NIFU. Tudor Ion from the European Commission, DG RTD, and Fernando Galindo-Rueda from the OECD discussed monitoring research careers through a Research and Innovation Careers Observatory. In the afternoon a presentations on “The DocEnhance Career-Tracking Survey of Doctorate Holders” by Julia Boman from the European Science Foundation and “The German National Academics Panel Study (NACAPS)” by Sarah Widany from DZHW. The event ended with a final open debate., Video of discussion: https://www.youtube.com/watch?v=gOrro7qlZOU

Identifying Open Access (OA) publications might seem a trivial task while practical efforts prove otherwise. In this project, we wanted to assign OA tags to publications in KB database. We queried KB in-house database up to 2017 (including Web of Science (WOS) and Scopus) for all articles and reviews. We then matched the corresponding DOIs to three sources of OA information: Unpaywall, Crossref and Bielefeld list of gold OA journals. This allowed us to define the OA status for publications. We found close to 14 million publications (articles and reviews between 2000 and 2016) from WOS (69.75% of all) and close to 18 million from Scopus (68.67% of all) with an equivalent DOI in Unpaywall. We matched KB publications database with Crossref data (from April 2018) and found 53 distinct licence URLs, which define in many cases the legally binding access status of publications. We found that more than half a million publications have more than one licence record in Crossref (in contrast to near 8 million with only one record and more than 6 million without a licence URL). We evaluated if these licences were open or closed access. We also matched respective journal ISSNs with DOAJ and ROAD databases and presented a categorization of publications to Gold, Hidden Gold, Hybrid and Delayed OA accounting for uncertainty due to missing licence information via a new sub-category Probable Hybrid OA. We validate our findings via manual checks and a crosscheck of OA information from the aforementioned varying sources. While the manual check on a sample of publications revealed a small but noticeable degree of apparently incorrect meta-information on publication’s OA status, the contrast of OA information from the diverse OA information sources highlights the partially unsteady base for an OA monitoring based on open data.

The distribution of treatment effects extends the prevailing focus on average treatment effects to the tails of the outcome variable and quantile treatment effects denote the predominant technique to compute those effects in the presence of a confounding mechanism. The underlying quantile regression is based on a L1{loss function and we propose the technique of expectile treatment effects, which relies on expectile regression with its L2{loss function. It is shown, that apart from the extreme tail ends expectile treatment effects provide more effcient estimates and these theoretical results are broadened by a simulation and subsequent analysis of the classic LaLonde data. Whereas quantile and expectile treatment effects perform comparably on extreme tail locations, the variance of the expectile variant amounts in our simulation on all other locations to less than 80% of its quantile equivalent and under favourable conditions to less than 2=3. In the LaLonde data expectile treatment effects reduce the variance by more than a quarter, while at the same time smoothing the treatment effects considerably.

Principal component analysis denotes a popular algorithmic technique to dimension reduction and factor extraction. Spatial variants have been proposed to account for the particularities of spatial data, namely spatial heterogeneity and spatial autocorrelation, and we present a novel approach which transfers principal component analysis into the spatio-temporal realm. Our approach, named stPCA, allows for dimension reduction in the attribute space while striving to preserve much of the data's variance and maintaining the data's original structure in the spatio-temporal domain. Additionally to spatial autocorrelation stPCA exploits any serial correlation present in the data and consequently takes advantage of all particular features of spatial-temporal data. A simulation study underlines the superior performance of stPCA if compared to the original PCA or its spatial variants and an application on indicators of economic deprivation and urbanism demonstrates its suitability for practical use.

The distribution of treatment e ects extends the prevailing focus on average treatment e ects to the tails of the outcome variable and quantile treatment effects denote the predominant technique to compute those effects in the presence of a confounding mechanism. The underlying quantile regression is based on a L1-loss function and we propose the technique of expectile treatment effects, which relies on expectile regression with its L2-loss function. It is shown, that apart from the extreme tail ends expectile treatment effects provide more efficient estimates and these theoretical results are broadened by a simulation and subsequent analysis of the classic LaLonde data. Whereas quantile and expectile treatment effects perform comparably on extreme tail locations, the variance of the expectile variant amounts in our simulation on all other locations to less than 80% of its quantile equivalent and under favourable conditions to less than 2/3. In the LaLonde data expectile treatment effects reduce the variance by more than a quarter, while at the same time smoothing the treatment e ects considerably.

Informal care has drawn much attention among scholars and policymakers as it concerns an essential but hard to evaluate resource of welfare. Albeit several studies addressed the monetary value of informal care, differences in the relationship between caregivers and recipients have often been ignored. We report on a profound and formerly unobserved distinction between care in the household and non-household care for a family member or in a voluntary framework. According to our results caregivers within the household perceive care as a burden and a positive shadow price arises. By contrast in the family but non-household context - and especially in the voluntary case - care is (at least partly) understood as an enriching experience which extends well-being and leads to negative shadow prices. This distinction calls a marketized view of informal care into question and may contribute to explaining the limitations of monetary incentive policies to encourage informal care.

Informal care has drawn much attention among scholars and policymakers as it concerns an essential but hard to evaluate resource of welfare. Albeit several studies addressed the monetary value of informal care, di erences in the relationship between caregivers and recipients have often been ignored. We report on a profound and formerly unobserved distinction between care in the household and non-household care for a family member or in a voluntary framework. According to our results caregivers within the household perceive care as a burden and a positive shadow price arises. By contrast in the family but non-household context - and especially in the voluntary case - care is (at least partly) understood as an enriching experience which extends well-being and leads to negative shadow prices. This distinction calls a marketized view of informal care into question and may contribute to explaining the limitations of monetary incentive policies to encourage informal care.

We present a twofold analysis in the domain of sex-related homicides. Police profilers often help in criminal investigation of high profile cases, but empirical evidence in the domain is incomplete. We therefore at first apply a structural learning approach and secondly, try explicitly to predict the age of an unknown offender from information obtained from the crime scene. We apply graphical modelling to obtain a factorisation of the probability function which governs the domain. This factorisation allows us to infer dependencies and independencies between the variables and therefore describes the domain. We apply several structure learning algorithms for Bayesian Networks and combine them to a final graphical model. In the second part, we compare several prediction techniques concerning their error rate in predicting the offender's age. The graphical model broadly presents a distinction between an offender and a situation driven crime. A situation driven crime may be characterised by an offender lacking preparation and typically attacking a known victim in familiar surroundings. The offender tends to apply blunt force to gain control over the victim and does not show a high level of forensic awareness. In contrast offender driven crimes may be identified by the high level of forensic awareness demonstrated by the offender and the sophisticated measures applied to control the victim. Furthermore the graphical model indicates that these offenders are more likely to attack an unknown victim in unfamiliar surroundings and prepare their attack. Applying several prediction techniques to the date results in a significant decrease in the root mean square error, if compared with a simple baseline model. However the actual performance of the best model, namely the lasso, is still not applicable in criminal investigation, as it its average error of 8 years is too high.

We present a statistical investigation on the domain of sex-related homicides. As general sociological and psychological theory on this specific type of crime is incomplete or even lacking, a data-driven approach is implemented. In detail, graphical modelling is applied to learn the dependency structure and several structure learning algorithms are combined to yield a skeleton corresponding to distinct Bayesian Networks. This graph is subsequently analysed and presents a distinction between an offender and a situation driven crime.