{"resourceId":"fischer-rau-rilke-ai-tutor-unrestricted-2026","versions":[{"version":"external-4678e1a3706335f89e514185b2743cd9dcbb5c757fc8cc8ff86cb2f3d7ec0d58","resource":{"id":"fischer-rau-rilke-ai-tutor-unrestricted-2026","title":"RCT: unrestricted AI tutor access raises exam performance 0.23 SD without crowding out reading","organization":"BiB/WZB/IZA; University of Göttingen; (VfS 2026 conference paper via EconStor/ZBW)","sector":"Higher education economics / learning sciences","geography":"Germany (university students; artefactual field experiment)","publishedAt":"2026-02-20","publicationDate":"2026-02-20","eventDate":null,"sourceName":"EconStor / Verein für Socialpolitik 2026","sourceLabel":"Academic RCT conference paper on AI tutoring design","sourceUrl":"https://www.econstor.eu/bitstream/10419/343773/1/vfs-2026-pid-151893.pdf","evidenceClass":"academic-research","outcomeClass":"mixed","topics":["knowledge-work","accessibility-workforce","operating-model"],"finding":"In a randomized experiment with 334 university students preparing for an incentivized exam, AI tutor access raised test performance by 0.23 SD versus control; unrestricted access outperformed a restricted design that forced initial independent reading by 0.21 SD (unrestricted about 0.34 SD vs control). Behavioral logs show unrestricted access produced gradual AI integration while restricted access triggered intensive prompting bursts. Gains were strongest for lower baseline knowledge with stronger self-regulation; authors note long-term retention and inequality effects remain open.","sledRelevance":"Transferable HE tutoring-design evidence for U.S. SLED buyers: tutor UX (unrestricted vs gated) and student self-regulation moderate learning gains—relevant alongside same-day course-chatbot adoption narratives.","evidence":"Artefactual field RCT, three arms (control; unrestricted AI+text; restricted read-then-AI), incentivized assessment, behavioral process data; February 20, 2026 conference paper version on EconStor (hdl:10419/343773).","architectureImplications":"Colocate tutor chat with required readings for the full study window; capture prompt timing and reading engagement; avoid UX that forces post-reading AI bursts without measuring process quality.","governanceImplications":"Treat tutor access rules as pedagogical policy with equity review for students with weak self-regulation; publish evaluation plans before scaling licenses.","securityPrivacyImplications":"Log interactions for research/integrity under institutional control; restrict vendor training on exam-prep prompts; clarify proctoring and academic-integrity rules when tutors are allowed.","caveats":"German university sample; short incentivized exam-prep task; not a semester retention study; self-regulation measures may not map 1:1 to U.S. advising systems; conference paper not yet journal-final.","streamIds":["student-success"],"roles":{"sales":"For teaching-and-learning and student-success buyers evaluating AI tutors, Fischer, Rau, and Rilke’s artefactual field experiment with 334 university students finds AI tutor access raises incentivized exam performance by 0.23 SD, with unrestricted access outperforming a read-first restricted design by 0.21 SD. Benefits concentrate among students with lower baseline knowledge and stronger self-regulation—so do not sell one-size-fits-all seat access as equity or retention insurance. Position tutors that support flexible integration with course texts, plus literacy for weaker self-regulators, and insist on proctored or course-embedded outcome measurement.","engineering":"Engineering should prototype AI tutors that sit beside assigned readings with full-session availability rather than hard gated read-then-chat flows that induced bursty prompting in this study. Instrument baseline knowledge, prompt timing, and reading engagement; segment analytics by self-regulation indicators. Constraints include a short exam-prep lab setting in German higher education, GPT-class tutor dependency, and no long-term retention evidence. Security and privacy need exam-integrity controls and limits on storing sensitive prompts outside institutional tenancy.","delivery":"Delivery is a bounded course-prep pilot: deploy an unrestricted-access tutor next to required readings for a defined study window, collect baseline knowledge and self-regulation measures, and compare incentivized assessment scores to a no-tutor control. Teaching faculty own learning design; academic technology owns tenancy; IR or educational research partners validate effects. Acceptance criteria should include SD-scaled score lifts, reading-engagement metrics, and subgroup reports. Risks include over-generalizing lab gains to semester retention and leaving low self-regulation students unsupported."},"roleDetails":{"sales":{"customerProblem":"Campuses buy AI tutors without knowing whether access rules help or hurt learning for different students.","stakeholders":["teaching & learning","faculty","student success","academic technology","IR/assessment"],"discoveryQuestions":["Will the tutor sit beside required readings for the full study window or be gated?","How will baseline knowledge and self-regulation be measured in the pilot?","What incentivized or course assessment will define learning gain?"],"valueHypothesis":"Flexible, reading-adjacent AI tutoring can raise short-run exam performance when students can integrate support strategically—especially lower-baseline learners with stronger self-regulation.","potentialEngagement":"4–6 week course-embedded tutor pilot with control comparison and process analytics.","unsupportedClaims":["Guaranteed retention or graduation effects","Equal benefits for students with weak self-regulation","That restricted read-first designs are always safer"]},"engineering":{"fit":"Fit for course-prep or homework tutors where readings and chat can share a session UI and analytics exist.","architectureIntegration":"Embed tutor beside PDF/LMS content; stream event logs (prompt bursts, reading time) to learning analytics; optional baseline quiz hooks.","prerequisites":["Institutional AI tenancy","Course reading package in accessible formats","Assessment instrument aligned to tutor scope"],"constraints":["Evidence from short lab-like exam prep","German HE context","LLM version drift over time"],"security":"Institutional logging with retention limits; block exfiltration of exam items; academic-integrity mode for graded settings.","proofOfValue":["Pre-registered score contrast vs control in SD units","Process metrics showing gradual vs bursty prompting","Subgroup report by baseline knowledge and self-regulation"]},"delivery":{"work":"Configure reading-adjacent unrestricted tutor access for a defined prep window; collect baselines; run incentivized or course assessment; debrief faculty on subgroup patterns.","dependencies":["Faculty content scope","Student consent/IRB or local research review if studying","Analytics export path"],"ownership":"Course instructor owns pedagogy; academic technology owns platform; educational research/IR owns evaluation.","skillsAdoption":"Brief students on strategic tutor use; coach low self-regulation students with structured prompts rather than bans alone.","governanceCheckpoints":["Pre-pilot integrity and privacy review","Midpoint check on bursty prompting and equity gaps","Post-assessment go/no-go for scale"],"acceptanceCriteria":["Documented access-rule design (unrestricted vs gated)","Score and process metrics reported with subgroups","Explicit statement of non-retention claims"],"risks":["Lab gains not transferring to semester outcomes","Widening gaps for weak self-regulators","Vendor model changes invalidating prompts"]}},"retrievedAt":"2026-09-25T03:22:55Z","enrichedAt":"2026-09-25T03:28:16Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Lower-baseline students may benefit most, but weak self-regulation predicts null effects—pair tutors with coaching and accessible reading formats rather than access alone.","procurementImplications":"Evaluate tutor products on session UX and analytics for reading/prompt behavior, not only model brand; require institutional evaluation hooks in contracts.","operatingModelImplications":"Operating model must include pedagogical access-rule design and equity monitoring, not only license provisioning.","sourceVerification":{"openedUrl":"https://www.econstor.eu/bitstream/10419/343773/1/vfs-2026-pid-151893.pdf","referenceExcerpt":"AI tutor access raises test performance by 0.23 standard deviations relative to control.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}