Graph-based approach to model the dependency information of graduate attributes for supporting the accreditation process

In the accreditation of an engineering program, the criterion of graduate attributes is particularly challenging due to its outcome-based nature, which involves diverse instructors to collect and analyze data on students’ skills and competencies in course activities. Also, the amount of data can be vast, causing the issues of relevance and consistency of the collected data. In this context, the purpose of this paper is to facilitate the relevant process by modeling the information dependency concerning the measurements of graduate attributes and the responsibilities of stakeholders. The modeling approach is based on the graph representation that focuses on the nodes and their relations. In the graph-based model, information contents are treated as nodes, which are classified into five types: graduate attribute (GA), attribute indicator (AI), program course (PC), learning outcome (LO), and grading component (GC). Then, the contextual interpretation of the GA assessments is specified by the relations that connect these content types. In this work, three types of content relations are defined: refine, measure and associate. Further, three types of stakeholders are identified (i.e., accreditor, administrator, and instructor), along with their relations to specify their responsibilities to the content types. To demonstrate the application of the proposed graph-based model, this paper overviews the use of the Integrated Course Design Tool (ICDT) and the course outline template in the accreditation process. Based on the graph-based model, suggestions are provided toward the development of quality function deployment and software tools.