The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2728
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T17:15:34.299312Z
status: published
lens: g_space
sector: education
headline: A Meta-Analysis of the Effect of AI Use on English Learner Engagement
dek: This study investigates the effectiveness of artificial intelligence (AI) use on English learner engagement and conducts moderator analysis to identify the variables that exert a moderating effect on AI’s impact. A multi-database search was conducted to locate 846 records, and 17 primary studies yielding 36 effect sizes were included in this meta-analysis after deduplication, screening and full-text eligibility assessment. Aggregated results show that AI use has a large average impact on English learner engageme…
gain_title: A meta-analysis of 17 studies (36 effect sizes) found AI use raised English learner engagement by a large average margin (g = 0.955, 95% CI 0.720 to 1.191).
problem_title: (none)
trace_subject: (none)
gain_reading: A meta-analysis of 17 studies (36 effect sizes) found AI use raised English learner engagement by a large average margin (g = 0.955, 95% CI 0.720 to 1.191).
gain_evidence: AI use has a large average impact on English learner engagement ( g = 0.955, 95% CI [0.720, 1.191])
problem_reading: (none)
problem_evidence: (none)
quick_read: A meta-analysis in Language Teaching Research pooled 17 primary studies, yielding 36 effect sizes, to test how AI use affects engagement among learners of English. The pooled effect was large: g = 0.955, with a 95% confidence interval of 0.720 to 1.191.

The authors began with 846 records from a multi-database search and narrowed them through deduplication, screening and full-text eligibility checks. Because the included studies differed substantially from one another, the pooled figure is an average across them, not a fixed effect for every classroom.

Moderator analyses found that the effect varied significantly by the role AI played and by learners' major. The authors say the findings are relevant to researchers, instructors and administrators working in English as a second or foreign language settings.
limitation: Substantial between-study heterogeneity means the pooled effect is an average across the included studies, not a figure that applies uniformly to every setting.
tag: Evidence-backed gain
key_points: A multi-database search located 846 records; 17 primary studies met the inclusion criteria after deduplication, screening and full-text review. | The 17 studies contributed 36 effect sizes to the analysis. | The pooled effect of AI use on English learner engagement was g = 0.955 (95% CI 0.720 to 1.191). | Between-study heterogeneity was substantial, so the pooled estimate is an average across studies. | The effect varied significantly depending on the role AI played in learning. | The effect also varied significantly by learners' major. | The study covers English as a second language and as a foreign language contexts.
rundown: The study searched multiple databases and found 846 records. After removing duplicates, screening and checking full texts for eligibility, 17 primary studies remained, which together supplied 36 effect sizes for pooling.

The aggregate result showed a large average impact of AI use on engagement among English learners, with g = 0.955 and a confidence interval of 0.720 to 1.191. The authors caution that substantial heterogeneity between studies means this is an average, not a uniform effect.

Moderator analyses tested which variables change the size of the effect. Two emerged as significant: the roles AI plays in learning and the learners' major. The authors frame the results as evidence on AI's effectiveness for engagement in English as a second and foreign language contexts.
sources:
- peer_reviewed | Language Teaching Research | https://doi.org/10.1177/13621688261498089 | 2026-11-03
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

f623601cc7cc7bb41c22565c2d8d8373d168c7b3ba56740578d0f060ec8ba2da

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000