研究數據管理發展基金（2023）

The project will benefit from a clear research data management plan. We aim to enhance the existing transcripts and improve the accuracy of the tagging of parts of speech of the child and adult utterances. Improvement of accuracy in tagging will greatly enhance the overall quality of the corpus.

The data will allow us to determine the developmental milestones for the children under investigation at different developmental stages and address important theoretical and empirical research questions and assess a number of factors that contribute to bilingual and trilingual development including the role of parental input, language dominance, degrees of balance in the target languages and crosslinguistic influence.

With a clear research data management plan, the project provides a successful case to show how good RDM practices can facilitate and enhance research productivity. We made attempts to enhance the existing transcripts and improve the accuracy of the tagging of parts of speech of the child and adult utterances. Improvement of accuracy in tagging in turn led to enhancement of the overall quality of the corpus and accuracy of quantitative and qualitative analysis.

The datasets allow us to study grammatical features as well as developmental milestones for the children under investigation at different stages and address important theoretical and empirical research questions and assess a number of factors that contribute to bilingual and trilingual development including the role of parental input, language dominance, degrees of balance in the target languages and crosslinguistic influence.

Mendoza, C., Law, C., Zhou, J., Li S., & Yip. V. (2026). Using Augmented Reality (AR)-compatible storybooks as input to support Hong Kong trilingual children’s early language learning. Early Multilingual Education Research Pathways from Policy to Practice, pp. 151-164. Tirant Humanidades.

This dataset features two video demonstrations of Augmented Reality (AR) materials designed to enhance vocabulary acquisition in Hong Kong preschool trilingual children during teleassessment and intervention. The videos showcase AR-compatible storybooks that promote interactive parent-child shared reading sessions, highlighting key functions, effectiveness, and engagement in language learning.

2. https://doi.org/10.48668/UFYV8N

Lok, C., J. H. N.Lee, S.Matthews and V.Yip. 2026. Responses to Cantonese A-not-A questions by Cantonese-English bilingual children. International Journal of Bilingualism
https://doi.org/10.1177/13670069251405684

Yip, V., & Matthews, S. (2007). The bilingual child: Early development and language contact. Cambridge University Press. doi: doi:10.21415/T5MS3Q

The CHILDES Cantonese-English Yip/Matthews Corpus, also known as the Hong Kong Bilingual Child Language Corpus, documents longitudinal development in eight children (Alicia, Charlotte, Darren, Janet, Kasen, Kathryn, Llywelyn, Sophie, Timmy) raised bilingually in Cantonese and English from birth in Hong Kong. The data feature naturalistic interactions, often following one-parent-one-language input patterns, with some code-mixing in parental speech, and cover ages roughly 1;03 to 4;06 depending on the child. Files are in .CHA format for analysis with CLAN (TalkBank) software, including audio/video-linked transcripts for select sessions, and support studies of simultaneous bilingualism, syntactic transfer, code-mixing, and relative clause development.

Mai, Z., C. Mendoza, Y. Liang, S. Mallhews, and V. Yip. 2025. “Issues in Research Data Management for child language corpora.” A masterclass conducted at the Australian Linguistic Society (ALS), Griffith University. Abstract posted on https://als.asn.au/Conference/Past-Conferences/2025/masterclass2025

Yip, V.. J. Lai and S. Matthews. 2024. “Investigating bilingual development using CHILDES corpora.” Masterclass conducted at the Australian Linguistic Society (ALS 2024), Australian National University. Abstract posted on https://als.asn.au/Conference/2024/masterclass2024

Reference was made to the project in Yip’s keynote speech (ALS 2024):

Yip, V. 2024. Through the Chinese looking glass: from bilingual to trilingual development Plenary given at the Australian Linguistic Society (ALS 2024), Australian National University

Specifically, I will use China and US export and import data to investigate how does firms’ financial reporting quality affect their export and import value and the number of trading partners. First, I will use survey data from executives to measure accounting quality and conduct country-sector-level analyses. Second, I will use firm-level international trade data to conduct more detailed analyses.

International trade is crucial for the global economy. The evidence provided in this project extends the understanding of the real economic effects of high quality financial disclosure. Moreover, there are important policy implications for developing countries that heavily rely on international trade for economic growth and at the same time have low financial reporting quality.

Through this innovative methodology, we aim to establish the first crime map in China that provides an accurate footprint for each individual crime instance. This proves invaluable for scholars, as it provides deeper insights into the spatial distribution of crimes, instead of just looking at the number of incidents. By analyzing the geographical patterns and clusters of criminal activities, we can uncover concealed trends and correlations that illuminate the dynamics of criminal behavior.

Through the creation of such crime map, we aim to equip policymakers, researchers, and communities with the knowledge necessary to develop effective crime prevention strategies and enhance urban safety in China.

Our team is working on the construction of a dataset based on court judgment documents. The dataset focuses on crimes with clearly identifiable incident locations—such as theft and robbery—and provides geocoded information for these cases. By moving beyond aggregate crime counts, it enables researchers to analyse geographic patterns, examine local contexts, and compare variations across regions and over time. The resulting dataset is intended to be an open and reliable resource that supports future research in crime, law, and spatial analysis.

To promote broader use and standardization, we also provide example code for geocoding additional crime types. This code demonstrates how researchers can extract location information from court documents, process it into a standardized geographic format, and integrate it into the dataset. We aim to make the dataset extensible and encourage other scholars to replicate, adapt, or expand it for their own research purposes.

This project is backed by CUHK research data management tools, ensuring our data is well-managed and accessible for future use. These tools provide secure storage, systematic version control, and long-term preservation of datasets, which guarantees the sustainability of our work.

Yuheng Lin, Zheqi Dai, Qiuqiang Kong, MusicScore: A Dataset for Music Score Modeling and Generation, ArXiv, 2025

The project has promoted proper RDM practices by providing training and supervision to research staff and sharing experiences with colleagues and postgraduate students. Additionally, the research assistant hired has served as the Data Champion representative for the Faculty of Law over the past year.

Also, the link below is a research seminar that was held to promote good RDM practices in the Faculty of Law: https://www.law.cuhk.edu.hk/app/events/faculty-research-seminar-research-data-management-and-a-hong-kong-sentencing-database-by-prof-kevin-cheng/

Cheng, K. K. Y., & Chan, Z. B. H. (2026). Sentencing with or without guidelines: a cross-jurisdictional analysis of England & Wales and Hong Kong. Legal Studies, 1-30.

A systematic and self-explanatory database will be established and the data will be deposited at the CUHK Research Data Repository (https://researchdata.cuhk.edu.hk). The database will be accessible to approved researchers and students undertaking research into a wide range of age-related topics and will contribute to improve our understanding of healthy ageing. We are open to various collaborations. Further information on this cohort and the database can be found in our website (http://www.jococ.org/zh-hk/mros-msos.php).

The well-structured database facilitates the use of information for various analyses related to healthy aging and other associated topics. Data can be retrieved from the CUHK Research Data Repository (https://researchdata.cuhk.edu.hk). On the project’s index page, there are two main folders: a data folder and a publication folder. Data is organized in the data folder by visit intervals (baseline, 2-year, 4-year, 7-year, and 10-year follow-ups). Health and medical data from the first 10 years, including fracture and mortality information, are openly accessible. Data from years 11 to 20 can be retrieved with approval from the principal investigator. Over 200 publications are stored in the publication folder. Various collaborators will be invited, and both undergraduate and postgraduate students are welcome to use the data for analyses on different topics. The dataset is also valuable for conducting systematic reviews.

We are open to various collaborations. We have shared information on our website http://www.jococ.org/zh-hk/mros-msos.php previously. After the data deposited in the CUHK, our website will redirect the link to CUHK Research Data Repository to enhance the use of CUHK database, at the same time promote RDM. The fund promotes good RDM by directly supporting the open access publication of project data. This action ensures data is findable and reusable, adhering to core RDM principles and regulations.

Research Objectives: This study aims to investigate what factors may affect specialist outpatient participation in medical decision-making in Hong Kong.

Methods: This study includes two phases. In Phase I, in-depth reviews (n = 15) will be conducted to explore patients’ experiences and their perceived barriers and facilitators of SDM in specialist outpatient settings. In Phase II, a cross-sectional survey (n = 600) will be conducted among specialist outpatient attendees. Informed by the insights from Phase I and the literature, the survey will gather quantitative data on patient demographics, preferred and actual involvement in medical decision-making, and factors influencing patient involvement in SDM. Additionally, the questionnaire will incorporate a subset of open-ended questions to investigate factors affecting patients’ experiences related to SDM.

Conclusion: This will be the first study combining qualitative and quantitative methods to explore factors influencing patient involvement in decision-making in Hong Kong. The outcomes of this research can inform healthcare practices, enhance patient-centered care, and contribute to the implementation of SDM in specialist outpatient settings.

The research was conducted in two phases. In Phase I, interviews with 30 patients helped identify the main barriers and facilitators. These included personal factors (such as health literacy and health status), behavioural factors (such as trust in doctors and the quality of interaction), and environmental factors (such as consultation time and the overall quality of care).

In Phase II, a survey of 600 patients provided a broader picture. The findings showed that some groups of patients were more likely to participate in shared decision-making than others. Younger patients, those in poorer health, those living with their families, and those who felt their doctors provided high-quality care were significantly more engaged in SDM. These results highlight that both patient characteristics and the healthcare environment play important roles in shaping patient involvement.
The project delivers valuable insights that can guide improvements in healthcare services. By identifying the factors that encourage patients to take part in decision-making, hospitals and policymakers can design strategies to support more inclusive and patient-centred care. This could mean offering more health education to boost patient confidence, improving communication training for doctors, or ensuring enough consultation time to foster meaningful discussion.

The project was made possible through funding support, which also helped ensure good Research Data Management (RDM) practices. Data were collected, stored, and analysed securely and responsibly, protecting patient confidentiality while enabling high-quality research. This ensures the findings are reliable and can inform better healthcare policies and practices in the future.

During the funding period, the project completed two comprehensive systematic reviews of existing healthcare utilization measures and developed a preliminary bilingual questionnaire tailored to the Hong Kong healthcare system. The team also conducted evidence synthesis, cross-comparison, and preparatory work for questionnaire development. While the larger-scale validation study remains ongoing and will continue beyond the project period, the groundwork established through this project provides an important foundation for the future completion and validation of the questionnaire.

A key focus of the project was the establishment of strong research data management (RDM) practices. With support from the Research Data Management Development Fund, the research team developed a comprehensive Data Management Plan using the CUHK DMPTool and adopted FAIR principles to make research data more findable, accessible, interoperable, and reusable. Standardized procedures were established for data collection, documentation, file naming, version control, quality assurance, storage, backup, security, and future data sharing.

To promote reuse and transparency, project data were organized using standardized formats, metadata, and data dictionaries. Plans were established for depositing research outputs in the CUHK Research Data Repository, making them searchable and accessible to other researchers and supporting future research and teaching activities.

The fund has significantly strengthened the team’s capacity to manage research data responsibly throughout the research lifecycle. As the validation study continues, the established RDM framework will continuously support efficient data collection, secure storage, transparent documentation, and future sharing of research outputs, thereby maximizing the long-term impact and value of the project for researchers, students, and the wider community.

Paper in submission: Huo Z, Xiong X, Lee ATC, Chan EYH, Quan J, Yip BHY, Lam LCW. Validation of Healthcare Utilisation Questionnaires in Children and Adolescents: A Systematic Review. Value in Health

Conference poster presentation:
1. Huo Z, Xiong X, Lee ATC, Yip BHK, Lam LCW. Validation Evidence of Self-Reported Measures for Healthcare Resource Utilization in Adult Populations: A Systematic Review, ISPOR 2025 at Montreal Canada, https://doi.org/10.1016/j.jval.2025.04.1254
2. Huo Z, Xiong X, Lee ATC, Yip BHK, Lam LCW. Validation Evidence of Healthcare Utilisation Measures for Paediatric Populations: a Systematic Review, ISPOR 2026 at Philadephia USA

Here, we share three processed and validated magnetic resonance imaging (MRI) datasets acquired from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN), the Harvard Aging Brain Study (HABS) and the Hong Kong SuperAgers study (ClinicalTrials.gov ID: NCT05728801), including high-resolution T1-weighted structural MRI and resting-state functional MRI (rs-fMRI) at 3 Tesla. These three datasets contain structural and resting-state functional MRI (rs-fMRI) neuroimaging data that has been acquired from 643 English healthy participants, 220 American participants and 488 Chinese old adults. Alongside, we share large-scale gradients estimated from each modality and constructed radiomic models. Our processed datasets will facilitate future research examining the coupling between brain macrostructure, mesostructure, connectivity, and cognitive function.

Beyond the deliverables, this project used the “re-generated” data and had novel finding in normal aging adults. All the brain scans are stored in the CUHK central clustering following the FAIR principles (Findable, Accessible, Interoperable, Reusable). The Research Data Management Development Fund pays well for procesthis long-term archiving, version control, and reproducible workflows, so other teams can verify or build on the results even years after the project ends. A live data management plan ensures transparency throughout.

In this proposed 6-month project, we aim to set up the common practice guidelines for storing these sequencing data into the CUHK Research Data Repository. We will first invite a group of students and researchers to input different kinds of data to the system, and via discussing and interviewing the participants, we will formulate a protocol based on the feedback. One workshop will then be carried out in the School of Life Sciences to train the students and researchers, and necessary modifications to the protocol will be made if deemed appropriate. Another workshop will be made to everyone in the University, and the established protocol will be made publicly available at the CUHK Research Data Repository website.

Throughout the 24-month project, we successfully developed robust RDM strategies that emphasize the meticulous organization, documentation, and storage of high-resolution imaging data. Key deliverables included customized training sessions, workshops, and comprehensive online resources that educated Research Postgraduate Students (RPGs) and faculty on RDM best practices. We also actively involved students through hands-on data management mentorship, fostering an institutional culture that values proper data archiving.

The Research Data Management Development Fund was instrumental in achieving these goals. It provided crucial support for hiring dedicated research and student personnel to manage the complex microscopy platforms, funded necessary equipment usage (including Confocal and TEM/ET microscopy), and enabled our team to share our RDM frameworks at the Symposium on Biological Cryo-EM/ET. Ultimately, this project has not only safeguarded valuable cellular imaging data for long-term reuse but also set a standard for RDM practices in the broader plant cell biology community.

Presentations and training materials shared during the CUHK workshops and external collaborative webinars.

Various synthetic protocols for common colloids have been systematically categorized according to deformability, composition, and structural characteristics. Source data were collected from previous and current projects. Then, they are validated and converted into a standardized metadata and protocol format to ensure consistency and ease of use. Categories currently included in the database encompass inorganic colloids (e.g. gold nanoparticles, silica, barium sulfate) and polymeric colloids (e.g. poly(N isopropylacrylamide), polymethyl methacrylate, polystyrene), among others. All curated records have been deposited in the CUHK research data repository to enable broad access by researchers and to encourage reuse and reproducibility of established protocols.

As part of this initiative, the CUHK research data repository, the CUHK Data Management Plan (DMP) tool and the above-mentioned AI platform have been integrated into our colloid research workflow now. The integrated system accelerates project initiation by providing ready access to validated protocols and standard metadata templates, and it ensures that data produced in future projects are captured, standardized, and appended to the repository. Going forward, this RDM framework will support continuous updating of the knowledge base, promote better practices in data management and enhance research productivity and reproducibility across the colloid science community.

The integrated RDM workflow combining the CUHK repository with Tallgeese AI enhances reuse of validated synthetic procedures while maintaining local, on‑device data control. As a result of this initiative, two research projects have been completed and their results were presented in international conferences.

The project aims to contribute to the understanding of how different learning strategies impact the acquisition of sign language, which could have implications for educational practices, particularly in the context of teaching the hearing-impaired. By making the database accessible to researchers and educators, the project supports the principles of open science and collaborative research. The findings from this study have the potential to inform educational strategies and contribute to the broader field of cognitive neuroscience and language acquisition.

To test our hypothesized relationships, we plan to conduct a cross-lagged panel study that observes 600 older employees (i.e., age 50+) living in Hong Kong over the time frame of one year and two experimental studies. The findings will highlight the relevance of older employees’ digital self-efficacy for their work continuance motivation and reveal factors that help to build and maintain digital self-efficacy in older age. Moreover, the results will provide new insights into the role of the organizational context (i.e., digitization of an organization and ageism in the workplace) and how it shapes whether digital self-efficacy translates into motivation to continue working.

The first dataset followed employees aged 45 and above over the course of a full year, tracking how their work experience evolves through time. The second dataset captured the work experience of employees across the entire adult age range over a one-month period. Together, these datasets offer new insights that have been integrated into two manuscripts currently under review at peer-reviewed international journals.

The project also created valuable training opportunities. A practicum student and postgraduate researchers were involved throughout the data collection process, gaining hands-on experience. A postgraduate student submitted a project-related poster that was accepted and presented at a leading international conference in the field. A postgraduate student from the PI’s lab also served as a data specialist for research data management through a scheme organized by the CUHK Library.

Finally, the project actively promoted sound research data management practices by organizing sessions for student helpers and lab members. These sessions covered preparing metadata, managing longitudinal datasets, and working with a formal research data management plan. To ensure the data can continue to support student teaching and training, the completed datasets and all relevant meta-information will be uploaded and shared openly through the CUHK Research Data Repository.

Study 2: https://doi.org/10.48321/D1D1714F26