TRV-2026-2198Version 1 · Certified

Written 2026-10-22 22:04:25 UTC · current record

Reason for this version

Certified into the record

Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-2198
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-22T22:04:25.674161Z
status: published
lens: g_space
sector: education
headline: How Perceived University Support Shapes Marketing Students’ AI Self-Efficacy: A Multigroup Dual-Mediation TAM Analysis Across Three Chinese Universities
dek: The proliferation of artificial intelligence (AI) in marketing has prompted higher education institutions to invest substantially. However, a critical pedagogical question remains: whether and how these institutional investments translate into students’ psychological readiness to deploy AI tools effectively. Students’ AI self-efficacy constitutes a foundational prerequisite for technology adoption and career adaptability. A lack of such confidence can precipitate technology avoidance behaviors and diminish learn…
gain_title: Perceived university support increased marketing students' perceived usefulness and ease of use of AI, which in turn raised their AI self-efficacy, with ease of use the stronger pathway.
problem_title: (none)
trace_subject: (none)
gain_reading: Perceived university support increased marketing students' perceived usefulness and ease of use of AI, which in turn raised their AI self-efficacy, with ease of use the stronger pathway.
gain_evidence: the indirect effect via PEOU was consistently stronger than that via PU
problem_reading: (none)
problem_evidence: (none)
quick_read: A structural equation modeling study surveyed 538 marketing students at three Chinese universities. It tested whether the support students feel from their university translates into confidence in using AI tools, through two beliefs: that AI is useful and that it is easy to use.

The findings suggest that spending on AI infrastructure alone may not build student confidence, and that teaching which makes AI tools seem less complex and shows their professional relevance may matter more. The data are cross-sectional and self-reported, so the causal pathway and its wider applicability remain uncertain.
limitation: The study relies on a cross-sectional, self-report design and a sample from a single country, which limits causal inference and generalization.
tag: Evidence-backed gain
key_points: The sample was 538 marketing students from three Chinese universities, analysed with structural equation modeling. | Perceived university support significantly increased both perceived usefulness and perceived ease of use in every sample. | The indirect effect through ease of use was consistently stronger than the one through usefulness. | The direct support-to-self-efficacy path stayed significant at two of the three institutions.
rundown: The study uses the Technology Acceptance Model as a framework for how institutional resources shape students' psychological readiness to use AI. It asks whether the mechanism holds across institutionally different settings.

Both usefulness and ease-of-use beliefs carried the effect of support onto AI self-efficacy, which makes this a dual-mediation pathway. The direct effect of support persisted at two of the three institutions, so it did not hold uniformly.

The authors recommend that universities channel investment into pedagogical strategies that demonstrably reduce the perceived complexity of AI tools and illustrate their professional relevance, rather than into infrastructure alone.
sources:
- peer_reviewed | Journal of Education and Training Studies | https://doi.org/10.11114/jets.v15i1.9189 | 2026-10-22
prev: 0000000000000000000000000000000000000000000000000000000000000000
sha256
05846c0c8e8b69f4698af64dda6ae1e72ac3fb9d3c2078131f9bc5c65954835d
previous
0000000000000000000000000000000000000000000000000000000000000000
Verify this record
How to verify without trusting this page

Fetch the canonical text of any version from /api/record/TRV-2026-2198 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.