BBC Pesantren
Description
This dataset contains 1,595 retrievable YouTube comments collected from a BBC News Indonesia documentary addressing sexual abuse by pesantren (Islamic boarding school) leaders in Indonesia. The dataset documents audience responses during the first eleven days following the documentary's publication and provides empirical material for examining public discourse, emotional expression, and audience engagement with sensitive social issues on digital media platforms. Each comment is recorded with its comment number, comment type (original comment or reply), publication date, number of likes, pseudonymised account identifier, and comment text. The dataset also includes systematic coding variables covering several analytical dimensions: manually coded emotions, final emotion classifications, discussion topics, communicative response practices, additional textual or contextual flags, and lexical references to specific cases (Jombang and Lombok). A validation sample indicator is included to identify comments selected for coding validation. The dataset is designed to support research in communication studies, media studies, digital sociology, and social sciences, particularly research on online audience reception, digital public discourse, emotional responses to mediated accounts of sexual abuse, and public engagement with institutional accountability. It enables quantitative analysis of comment distributions and audience reactions, as well as qualitative and thematic examination of highly liked comments and extensive reply threads. The data can be used to investigate how audiences express anger, empathy, solidarity, criticism, and other responses to sensitive issues; how discussions develop around institutional, religious, and social contexts; and how engagement patterns reflect different forms of public participation in online discussions. Personal account identifiers have been pseudonymised. The dataset is provided as a supplementary research resource for transparency, methodological documentation, and further scholarly analysis of audience responses to sensitive social issues in Indonesian digital media. Keywords: YouTube comments; audience reception; digital discourse; emotional responses; social media; sexual abuse; pesantren; Indonesia; BBC News Indonesia; online engagement.
Files
Steps to reproduce
The dataset was compiled and coded through the following procedures: 1. **Data Source Identification** Identify the BBC News Indonesia documentary addressing sexual abuse by pesantren leaders, published on YouTube. The documentary serves as the primary source for collecting audience comments. 2. **Comment Collection** Retrieve publicly accessible comments from the documentary during the first eleven days following its publication. A total of 1,595 retrievable comments were collected, including original comments and replies. 3. **Data Compilation and Preparation** Organise the collected comments into a structured dataset using CSV format. Each record represents an individual comment. Relevant metadata, including comment type, publication date, likes, pseudonymised account identifier, and comment text, are retained where available. 4. **Comment Classification and Coding** Code each comment according to the established analytical framework. The coding dimensions include: - Emotion: identification of emotional expressions in the comments. - Topic: classification of the main subject or issue discussed. - Response practice: identification of communicative and discursive response patterns. - Contextual flags: identification of additional relevant textual characteristics. - Case references: identification of lexical references to Jombang, Lombok, or both. 5. **Coding Validation** Select a validation sample of 200 comments to support the assessment of coding consistency and classification reliability. The validation sample is identified through a dedicated variable in the dataset. 6. **Data Organisation and Export** Compile the original comment information and coded variables into a consolidated CSV file. The resulting dataset contains 1,595 records and is provided for further quantitative, qualitative, and mixed-methods analysis. **Reproducibility Note:** Researchers wishing to reproduce the analysis should apply the same documentary source, eleven-day collection window, comment inclusion criteria, and coding framework. The original coding definitions should be consulted to ensure consistency in interpreting the abbreviated category codes.
Institutions
- Syarif Hidayatullah State Islamic University JakartaWest Java, Ciputat