Abstract:

This study tests the hypothesis that ChatGPT is an appropriate interlocutor to evaluate the structure and outcomes of degree programs in almost all higher education disciplines. This capability arises from the system's vast and ever-increasing reservoir of information sources, which can be applied to structuring the curriculum of programs and courses and then contrasting them with human-designed curriculum and student performance indicators. Three high-enrollment programs in an eastern university in the United States and ten courses per program were selected for analysis. The study consisted of developing and applying an evaluation protocol in five steps:
(1) program outcomes vs. workplace requirements;
(2) faculty-defined outcomes vs. AI-generated outcomes;
(3) course syllabus review;
(4) faculty responses; and
(5) student outcomes.

It facilitated the creation of an artificial intelligence protocol for university or college curriculum evaluation, termed “Multi-Level Curriculum Cohesion Assessment.” Hypotheses were tested using chi-square and correlation analysis.

In summary, the study objectives were to:
(1) evaluate the alignment between program outcomes and workplace requirements;
(2) compare faculty-defined outcomes with AI-generated outcomes;
(3) thoroughly review course syllabi with AI for cohesiveness and relevance;
(4) gather and analyze faculty responses to AI-generated insights; and
(5) assess student outcomes regarding curriculum quality as determined by the AI Protocol.

Based on preliminary results, it can be asserted that faculty members find the curriculum evaluation helpful for improving courses and programs. The study determined a positive correlation between course quality evaluated through the protocol and student performance, engagement, and satisfaction variables.

Keywords:

Artificial intelligence and education, curriculum evaluation, AI-generated curriculum, Ralph Tyler, Curriculum Questions.