Pilot self-efficacy gains decreased with self-reported expertise: beginners gained most (+1.18), intermediates +0.94, and advanced users +0.46, all significant¶

Claim · All claims
Evidence · 1 study · design r? · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

Reading the evidence codes

q is the design tier, i the impact, n the sample, and the kind with its r the rigour judged within that kind: what each code means.

This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.

q2 i? Beginners showed the largest gains (+1.18), intermediates +0.94, and advanced users smaller but still significant gains (+0.46), with all three subgroups at p <0.001. → Zain Naboulsi 2026

Evidence¶

Zain Naboulsi 2026¶

Zain Naboulsi. (2026). Agentic Education: Using Claude Code to Teach Claude Code. arXiv. https://arxiv.org/abs/2604.17460 doi:10.48550/arxiv.2604.17460

q2 · i? · design · r?

Subgroup analysis of the pilot by self-reported experience level (beginner n= 9, intermediate n= 8, advanced n= 10). The article reports "Beginners showed the largest gains (+1.18)" and that all three subgroups reached p <0.001; no standardized effect size is printed for the subgroup contrast.

"Beginners showed the largest gains (+1.18), intermediates were close behind (+0.94), and advanced users showed smaller but still significant gains (+0.46)."

Discussion¶

Learner Variables¶

Prior Knowledge — predictor: learners who differ on it differ in outcomes

Proximal goals and ability attributional feedback for early learning raise self-efficacy more than distant goals and effort feedback, by a review's report of earlier studies — the general claim this finding bears on, with the studies behind it
The largest self-efficacy effect sizes in the pilot appeared on advanced features such as setting up hooks and creating skills — related
A within-subjects pilot with 27 professional software engineers found statistically significant self-reported self-efficacy gains across all 10 assessed skill areas after completing the cc-self-train curriculum — a broader claim this one bears on
The most effective forms of learning support differed by user expertise: beginners preferred at-a-glance visual cues while advanced users preferred the analysis console. — related
Training improved accuracy in every digital-forensics experience group and most generative-AI familiarity groups, with the largest estimated gains among analysts with no forensics experience and frequent generative-AI users — a broader claim this one bears on

Page metadata

Field	Value
`type`	`claim`
`id`	`cc-self-train-gains-by-expertise-level`
`title`	`Pilot self-efficacy gains decreased with self-reported expertise: beginners gained most (+1.18), intermediates +0.94, and advanced users +0.46, all significant`
`description`	`Pilot self-efficacy gains decreased with self-reported expertise: beginners gained most (+1.18), intermediates +0.94, and advanced users +0.46, all significant`
`status`	`review`
`evidence_strength`	`moderate`
`newest_evidence`	`2026`
`generated`	by `process:wiki-ingest` on `2026-10-10`
`trust`	unverified — no one has confirmed this page's content
`sources`	1 — `zain-naboulsi-2026`

Frontmatter is the page's machine-readable record — see the Schema & Guide.