Introduction The R2D dataset collection was developed as part of the digitisation project 'Razzia van Rotterdam Digitaal' (2025-2026). The project aimed to digitally preserve, transcribe, and enrich a post-war research archive compiled by historian Ben Sijes between 1946 and 1951. This collection formed the empirical basis for Sijes’ study

De razzia van Rotterdam, 10-11 november 1944 (Nijhoff, 1951).

Sijes’ research examined the large-scale razzias carried out by the Nazi occupation regime in and around Rotterdam in November 1944, during which approximately 52,000 men were forcibly removed from their homes and subjected to forced labour in the Netherlands and Germany.

The aim of the 2025-2026 digitisation project was to make this historically significant material sustainably accessible and reusable for both researchers and a broader audience. By applying contemporary digitisation and data enrichment methods, the project enables new forms of research, preservation, and presentation.

As part of this work, the questionnaires used by Sijes were not only digitised as images but also processed into a structured, machine-readable dataset. Using automated layout recognition and text recognition techniques, the project produced structured transcriptions of the questionnaire forms, linking the original survey questions to the transcribed handwritten responses of participants. This structured representation makes it possible to analyse the questionnaire data systematically and to connect individual responses.

Creator The dataset collection was created by the NIOD Institute for War, Holocaust, and Genocide Studies (hereafter: NIOD). Founded in 1945, NIOD serves as the national institute for the documentation and study of war and occupation in the Netherlands. The institute functions both as an archival repository and as a research institute, conducting interdisciplinary research on war, mass violence, and genocide.

NIOD manages approximately 400 archives and collections, together comprising around 2,500 linear meters of material related to twentieth-century war and mass violence. Through digitisation projects such as 'Razzia van Rotterdam Digitaal', NIOD aims to ensure the long-term accessibility and reuse of its collections for research, education, and public engagement.

Razzia van Rotterdam Collection Sijes’ research collection, Collectie Onderzoekingen 258: Razzia van Rotterdam, consists of nearly six linear meters of documentation gathered between 1946 and 1951. The collection includes completed questionnaires from respondents of diverse social backgrounds, personal letters, correspondence with governmental and private institutions, and written reports of interviews and personal conversations.

Sijes personally conducted interviews, distributed hundreds of questionnaires, and maintained correspondence with a wide range of contemporaries, including resistance members, mayors, civil servants, police officers, company employees, and victims of the razzia and their relatives. He considered these interviews and questionnaires an important supplement to diaries and letters.

Because the collection was assembled shortly after the events of November 1944, it provides both direct insight into individual wartime experiences and valuable evidence of early post-war research practices at the Rijksinstituut voor Oorlogsdocumentatie (RIOD), the predecessor of NIOD.

Contents

The R2D dataset collection consists of three main components.

1. Persons index with demographical data

This file contains tabular data with personal and demographic information on individuals surveyed or interviewed for Ben Sijes’ research project (1946-1951). The dataset was originally compiled during earlier research based on NIOD’s archival collection 'Collectie Onderzoekingen 258: Razzia van Rotterdam' and later enriched with additional information.

During the 2025–2026 'Razzia van Rotterdam Digitaal (R2D)' project, the dataset was cleaned, manually checked, and expanded. Where possible, links were added to HTR transcriptions from NIOD’s digitisation project and to corresponding personal entry pages on Oorlogsbronnen.nl.

Contents: • Readme file: PersonsIndex_R2D_2026_Readme (.txt) • Persons index table: PersonsIndex_R2D_2026 (.csv)

1. Transcriptions of correspondence, interview reports, and questionnaires (1946-1951)

The folder 'Transcriptions of correspondence, interview reports, and questionnaires' contains automatically generated transcriptions of historical handwritten and typescript documents. The transcriptions were created using Handwritten Text Recognition (HTR) and Optical Character Recognition (OCR) and are provided in Alto XML format, preserving both text and layout of the original records.

Contents: • Readme file: TextTranscriptions_R2D_2026_Readme (.txt) • 29.624 Transcriptions in Alto XML format organized, in 346 different folders (.xml)

1. Structured transcriptions of questionnaires (1946-1951)

This folder contains machine-generated structured transcriptions of the questionnaire forms used by Sijes, including both blank and completed forms. The questionnaires were processed using automated layout and text recognition to identify elements such as question numbers, answer fields, and document structure.

The results are provided in JSON format, making the data machine-readable and suitable for further computational processing and/or analysis. Each file contains information about the questionnaire structure and links questions to the transcribed handwritten responses of participants. Contents: • Readme file QuestionnaireTranscriptions_R2D_2026_Readme (.txt) • 11.944 JSON files containing questionnaire structure and transcribed text in 180 folders (.json)

Access NIOD seeks to promote broad access to and responsible use of the data it collects and publishes for academic research purposes. Access to this dataset is provided free of charge to individuals or legal entities. Use of this dataset is subject to the present Terms and Conditions. These terms apply not only to the dataset in its original form but also to any data derived from it, including representations such as tables, charts, or other non-textual formats.

If you wish to use this dataset, you must request permission via the ‘Access File’ button or by contacting: onderzoeksdata@niod.knaw.nl, as copyright, GDPR, or other legal restrictions may apply.

Terms and Conditions 1. The dataset may be used exclusively for non-commercial academic research purposes. 2. The dataset may contain: a. personal data relating to living persons or individuals who may still be alive; and/or b. material protected under copyright law or other applicable legislation. Such data and materials are provided solely for personal study and research. They may not be published, shared, distributed, or otherwise made publicly available without the explicit consent of the relevant data subjects and/or rights holders. 3. NIOD accepts no liability for any unlawful or improper use of the dataset, including violations of privacy rights or copyright legislation by the user.

NIOD is committed to safeguarding the privacy of living persons and to identifying copyright holders wherever possible. Further information can be found in NIOD’s privacy policy.

If you have concerns regarding privacy, such as requests for the removal of names or other personal data, or if you believe you hold rights to materials included in this dataset, please contact onderzoeksdata@niod.knaw.nl.