Generative AI literacy and adoptionConnecting learner capability, intention and educational context

Research question

How do learners’ capabilities, intentions and educational circumstances shape the study of generative AI adoption? A useful starting point is to identify the outcome a project wants to explain: willingness to adopt, effectiveness during a learning task or change in educational practice. The papers in this guide examine these questions through a study of adoption intentions, a programming quasi-experiment, a review of educational applications and a synthesis of policy documents. Reading them together creates a route from the learner to the surrounding educational setting. The connections developed here form a research agenda for studying these levels together.

Adoption intentions

Chengliang Wang and colleagues’ mixed-methods study focuses on university learners’ intentions to adopt AIGC technology, using ChatGPT as the example. Practical activities and interviews inform a decision framework, followed by fuzzy-set qualitative comparative analysis of 233 valid questionnaires. The study identifies six configurations associated with high adoption intention and highlights AI literacy and perceived behavioural control. Its contribution is a configurational account of learners’ stated intentions. It is especially relevant when a research question concerns combinations of conditions and when qualitative exploration is used to inform a subsequent questionnaire-based analysis of the intended outcome.

This contribution helps specify what an adoption study needs to measure. A project interested in willingness can define an intention outcome and investigate the circumstances associated with it. A project interested in sustained educational use can add observations of subsequent activities and specify what counts as use in that course. These are different study designs with a possible empirical connection to examine over time. Based on the mixed-methods paper, a proposed extension could investigate whether learners with similar intentions engage in different kinds of activity, and how their AI literacy and perceived control relate to those patterns.

Capability and context

Jing and colleagues’ programming study provides a task-level perspective. Its quasi-experimental investigation of Python graphing connects effective ChatGPT-assisted problem solving with AI literacy, programming knowledge and understanding of ChatGPT. The abstract reports that usage intention did not significantly affect effectiveness within the study, while intention increased after application. This result helps formulate an empirical question about the relationship between wanting to use a tool and being able to work with it effectively. The paper is useful when motivating the measurement of disciplinary knowledge and AI-related capabilities alongside intention in a study of educational adoption.

Chen, Hu and Wang’s systematic review expands the scope to the educational applications of AIGC. It examines 134 publications from EBSCO, EI Compendex, Scopus and Web of Science and organizes work on performance assessment, instructional applications, learning outcomes, advantages and disadvantages, and future prospects. Grounded-theory analysis also informs its account of application scenarios. This review is useful when choosing the educational activity to which an adoption question refers. A proposed study can specify whether learners are evaluating assistance with an instructional task, assessment activity or another application, then select outcomes and participant questions that fit that setting.

Wang and colleagues’ policy study adds the organizational context. It uses meta-ethnography to synthesize 11 policy and guideline documents and develops a transformation action framework across macro, meso and micro levels. The framework connects social and workforce change, educational organization and pedagogical logic, and teaching and learning scenarios. Its contribution is an interpretation of policy documents that can organize questions about educational change. In an adoption project, a researcher could use those levels to describe the institutional setting and examine how course arrangements relate to learner experiences. That application of the framework is a proposed study direction.

Research directions

Taken together, the studies suggest a focused sequence for future research: choose a concrete educational application, describe its course and institutional context, measure learner capabilities and intentions, and follow relevant task activity. Each component has a distinct literature entry point. The adoption study supports configurational questions about intention; the programming study motivates attention to task effectiveness; the AIGC review helps locate an application; and the policy synthesis helps describe the surrounding educational arrangements. A project following this route could investigate how students with different starting capabilities move from an initial intention to particular learning practices. These relationships remain questions for the proposed project to test, with its outcomes and observation period specified in advance.