Social Sciences and Digital Humanities Archive – SODHA
Not a member yet
133 research outputs found
Sort by
The Tip Off
No full textThis dataset contains 26 episodes from the independent podcast "The Tip Off", which is produced and hosted by investigative journalist Maeve McClenaghan. The episodes were collected in 2025 and were selected through random and purposive sampling. They were analyzed via multimodal discourse analysis to examine how emotions were present in the discourses and sound design of the podcast
Report belgicaweb - Data collection & description of process
No full textFriedel Geeraert, Christina Vandendyck. (December 2025). Deliverable 1.3.1 Data collection & description of process. Technical report about the methodology and architecture implemented for the BelgicaWeb project's data collection chain and the complete workflow from website to archive
ARTPRESSE - A Curated Dataset Investigating Fine Arts in Belgian Family Magazines 1929-1936
No full textThis dataset contains the files related to a survey investigating the fine arts in seventeen titles of Belgian family magazines between 1929 and 1936 (FR/NL). The survey took place within the framework of the Artpresse project (2020-2025), which aimed at digitizing and studying a corpus of interwar mass-market periodicals from the perspective of their cultural content, leveraging the periodical collections and digital tools of the KBR. Funded by the Federal Science Policy of Belgium (BELSPO) through its BRAIN-be 2.0 program, Artpresse is a research and digitization project led by the Digitization Department of the Royal Library of Belgium (KBR) in collaboration with the Department of Literary Theory and Cultural Studies of the University of Leuven and the Service d'Histoire de l'Art de l'Époque contemporaine of the University of Liège. Focusing on the fine arts scene in Belgium from the 1929 stock market crash to 1936, this survey captures a critical selection of content drawn from magazine titles representative of the ideological and technological trends of Belgium’s interwar illustrated press. From the broader Artpresse corpus (comprising over 700,000 digitized pages on the Belgica Periodicals portal), a specific sub-corpus of 150,592 pages across 4,369 issues was selected for analysis. Within this scope, the dataset comprises 1,542 items identified or interpreted as fine arts content. This structure links specific editorial units with the individual figures identified within them, enabling the systematic documentation of 1,217 distinct artists. The dataset also serves as the foundation for an illustrated and indexed catalog, integrating image files and transcriptions as the primary material for an art historical doctoral research on the representation of arts in a large digitized corpus of illustrated press
ArchMapper | Tool for describing and linking archival series to a retention schedule
No full textArchMapper is a tool that can be used to link files and folders to archival series defined in a retention schedule (selection list) and/or describe them on an ad-hoc basis. The allocated links and descriptions and a summary can be saved in a file fully compliant with the State Archives's template for transfer lists
GENDISC Data
No full textThe data was generated for the 'Gender Discrimination in the Tarim Basin' project (GENDISC), based on the Kharosthi documents found at the Niya site located in the Xinjiang Uyghur Autonomous Region of China. It provides details of find locations, royal titles, the types of complaints people made, and the subject of legal cases. The documents date to between the 2nd and 4th centuries CE
SIGMA (longitudinal cohort study on adolescent mental health)
No full textA longitudinal study that combined questionnaires and ecological momentary assessment data to assess the relationships between social risk factors, psychological mechanisms and psychopathology in an initial sample of 1,913 adolescents
aFLEXIthymia - Stimulus selection
No full textThis dataset contains trial-level data from the following study :
Mas, M., & Luminet, O. (2025). Alexithymia Moderates Salience Effects in Emotional Facial Expression Perception and Recognition. International Journal of Psychology, 60(2), e70037. https://doi.org/10.1002/ijop.7003
Stemtest 2024 / Test Electoral 2024
No full textThis deposit accompanies the datasets related to the development of Stemtest 2024 / Test Electoral 2024, a voting advice application developed in the context of the 2024 general elections of June 2024. Stemtest/Test Electoral was commissioned by VRT, RTBF, Mediahuis and IPM.
In total, the project comprised 12 distinct voting advice applications (VAAs):
2 ‘generic’ VAAs that span the three policy levels (Federal, Regional and European) (referred to as ‘VAA_GEN_NL’ for the Dutch version, ‘VAA_GEN_FR’ for the French version);
2 ‘youth’ VAAs with particular focus on topics pertaining to younger voters (also spanning the three policy levels (referred to as ‘VAA_YOUTH_NL’ and ‘VAA_YOUTH_FR’);
2 federal VAAs containing only statements on federal policy competences (‘VAA_FED_NL’ and ‘VAA_FED_FR’);
4 regional VAAs containing only statements pertaining to the regional competences (VAA_VLA for Flanders, VAA_WAL for Wallonia, VAA_BXL_NL and VAA_BXL_FR for Brussels);
2 European VAAs containing only statements pertaining to the European competences (VAA_EU_NL and VAA_EU_FR).
The data associated with the deposit are:
Methods note: outlines the process and outlines the relevant documents pertaining to the development of the 2024 VAA.
Final statement dataset: finalized dataset in Stata, SPSS, CSV and Excel format, containing all 348 policy statements, party positions and arguments, assignment to each of the 12 VAAs, statement weights, left-right coding, issue allocations, etc.
Benchmark surveys: 8 datasets in Stata and SPSS format containing the benchmark survey data used in the calibration. The benchmark names refer to the associated VAA for which they were used. Note that due to budget constraints, the following VAAs were not benchmarked: VAA_BXL_NL, VAA_BXL_FR, VAA_EU_NL and VAA_EU_FR.
Calibration files: this stata do file (Calibration.do) uses multiple raw input files (located in the subfolder ‘raw data’) to generate 12 excel files that KPSoft needed to set up the calibration tool.
Test files: this stata do file (TestFiles.do) updates the test files for BitsOfLove (BoL).
Media reporting files: the script “MediaReporting.do” generates excels with basic analyses of the Benchmark surveys.
Log data reports: the reports based on the log data of VAA usage were prepared in early 2025, and can be found under the subfolder ‘LogDataReports’.</li
ProvEnhance - Collection Risk Assessment of Artworks Acquired by the Royal Museums of Fine Arts of Belgium, 1933-1960
No full textAs part of the research project “ProvEnhance. Enhancing the provenance data of the collections of the Royal Museums of Fine Arts of Belgium (RMFAB) since 1933 – Scientific study, digital valorization and narrative in context”, the collections of the Royal Museums of Fine Arts of Belgium are examined as a case study to advance Belgian Nazi-era provenance research, aiming to progress it in alignment with the Washington Principles. Therefore, a preliminary risk assessment typically performed in the realm of Nazi-era provenance research, referred to as the “First Check” was conducted. The objective was to evaluate the necessity and priorities for additional research within the specified project corpus to aid in revealing possible historical injustices and to identify the rightful owners if needed. As a result, this dataset brings together the already available provenance information for the artworks included in the project corpus, offering a base for further exploration of object histories and the museum’s network involved in the acquisitions. Although the compiled data is not exhaustive to date, it provides a basis for future in depth-research. Publishing this dataset aims at rendering the histories of the objects, the biographies of former owners, and the underlying practices of the RMFAB accessible and transparent – all while situating them in their historical context. (2025-12-15
Replication Data for: 4000 AI tools - 50 categories - TasticAI
No full textIntroducing an extensive and openly accessible dataset custom-tailored for AI researchers and data scientists, this repository encompasses an impressive array of over 4,000 AI tools, meticulously categorized into more than 50 distinct classifications, generously made available by TasticAI to facilitate diverse applications such as research, benchmarking, and market surveys.
Overview of the Dataset:
Nomenclature of AI Tools: Each AI tool has been systematically assigned a unique name to facilitate facile reference.
Concise Descriptions: Succinct one-line descriptions provide an overview of the intended purpose and functional attributes of each AI tool.
Comprehensive Categorization of AI Tools: With careful consideration, these tools have been thoughtfully grouped into more than 50 specific categories, accommodating a wide spectrum of research interests.
Visual Aids: Complementing textual information, visual representations are included alongside each AI tool, expediting the process of identification.
Web Links: Direct hyperlinks to associated tool websites or documentation have been thoughtfully integrated, streamlining access to detailed information.
Utilization and Advantages: This dataset serves a multitude of purposes:
Research Facilitation: Researchers can promptly discern pertinent AI tools, significantly expediting literature reviews and enabling comparative analyses.
Performance Benchmarking: Rigorous evaluation and comparative assessments are made feasible within and across specific tool categories.
Market Insight: This dataset offers valuable insights into the landscape of AI tools, enabling informed decision-making for market analysts.
Educational Endeavors: Ideal for pedagogical purposes, educators and students can leverage this dataset to enrich their understanding of AI tools and their multifaceted applications.<br