Susmita Pratihast, Erin Mote

Proceedings of the 13th International Conference on Society and Information Technologies: ICSIT 2022, pp. 19-24 (2022); https://doi.org/10.54808/ICSIT2022.01.19

Susmita Pratihast
InnovateEDU, New York, United States

Erin Mote
InnovateEDU, New York, United States

Cite this paper as:
Pratihast, S., Mote, E. (2022). Blueprint for Inclusive Research & Development in Education: A Data-Driven Approach to Organize and Articulate Research Data Needs for Equitable Evaluation. In N. Callaos, S. Lunsford, B. Sánchez, A. Tremante (Eds.), Proceedings of the 13th International Conference on Society and Information Technologies: ICSIT 2022, pp. 19-24. International Institute of Informatics and Cybernetics. https://doi.org/10.54808/ICSIT2022.01.19

In the field of education, there is a gap between research and practice. Lack of data standardization and collection inhibits comparability and generalizability of findings in the context of population heterogeneity. Mapping of education research inquiries to local heterogeneous data representations from different repositories is expensive and creates barriers for content comprehensiveness in research. The paper focuses on design and usability of the Blueprint, an organizing data framework that summarizes key elements to represent education data needs to improve articulation of data needs and allow practitioners to be part of an inclusive research and development process in collaboration with the research community. The design process included an exhaustive landscape analysis of established data standards, conceptual frameworks in education domains as well as publicly available databases and indicators. This was supported by the analysis of current evidence and research base with the use of natural language processing techniques and topic modeling approaches. This methodology was used to validate and inform the inclusion of different elements in the Blueprint. A large stakeholder review process was adopted to further enhance and refine the Blueprint. With a generalizability score of 77% across multiple pilot studies, the paper focuses on the results of the pilot studies and demonstrates the real-world use cases for widespread applicability and scalability in the sector.