Technology–governance–outcome alignment in digital education: a PRISMA-based configurational meta-synthesis (2015–2025)

Gele Qing, Аliya Akramova

2026

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.3389/feduc.2026.1912384
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Much of the policy discussion around digital education proceeds on the assumption that technically capable systems, if properly implemented, will translate into improved learning outcomes. The empirical record does not consistently support this expectation—and the inconsistency is not random. This systematic meta-synthesis of 147 SSCI-indexed studies from 26 national contexts (2015–2025) examines why, by shifting the analytical lens from what technologies can do to the governance conditions determining whether they will. Following PRISMA 2020 guidelines, studies were coded across five dimensions (Cohen's κ = 0.81) and analyzed through a two-stage pipeline combining hierarchical clustering with crisp-set qualitative comparative analysis (csQCA). Four configurations emerged above the consistency threshold of 0.75. AI/ML tools combined with decentralized governance and explicit outcome alignment yielded positive outcomes in 73.4% of cases (consistency = 0.84); any technology under centralized governance without alignment, only 31.2% (consistency = 0.82). Outcome alignment was present in 94.1% of positive-outcome studies; formal necessity-analysis consistency and coverage values were not calculated, so this is reported as a descriptive association rather than a necessity finding, and the identification problem between alignment-as-cause and alignment-as-institutional-quality-marker remains unresolved. The Technology–Governance–Outcome (TGO) framework repositions institutional design as the constitutive condition of technology effectiveness, not its organizational backdrop. Two structural tensions run through the findings: centralized and decentralized governance both achieve positive outcomes, but through mechanisms that cannot be reconciled into a single policy prescription; and the governance conditions most predictive of success are systematically least available to the institutions most in need of improvement. The synthesis provides an early quantitative reference point for Central Asian digital governance, surfacing intra-national variation that aggregate statistics obscure.