Student Success by DesignA practitioner’s guide to theory, data, and AI in student success

4.4 · The Student Success Ecosystem

Components and what the evidence shows

Each component contributes, but effect sizes vary widely and depend on quality, dosage, and access.

Short shows the summary, takeaway, and questions.
Interactive chart · Strength of evidence by component

Select a row for the details.

Integrated models (CUNY ASAP-type)Strong

Randomized trials show about 18 percentage points on three-year graduation, with replications in Ohio and Westchester.

Supplemental instruction and tutoringStrong

Grade effects of d ≈ 0.29–0.60 in a systematic review; stronger with higher dosage.

Basic needs (link to attrition)Strong

Dose-response links between unmet needs and attrition markers. Causal evidence for specific services is more limited.

Proactive advising and case managementModerate

Proactive advising with follow-up beats information-only outreach. Stay the Course found gains in persistence and completion for women.

High-impact practicesModerate

Consistent associations with engagement and retention; access is uneven. First-year seminars alone show small effects (δ ≈ 0.11).

Belonging and validationModerate

Small-to-moderate correlations with motivation and engagement; brief belonging interventions can narrow gaps.

Paid internshipsModerate

Linked to about 15 percent higher early earnings in matched analyses. Selection effects are likely.

Transitions and pathwaysEmerging

Promising descriptive and quasi-experimental results; few rigorous causal tests.

Standalone early alert and technologyMixed

Prediction is reasonable. Impact depends on high-quality human follow-up.

Advising and case management
Comprehensive, proactive advising with follow-up outperforms transactional or information-only approaches. Mentoring and coaching show modest effects on retention (d ≈ 0.15) and graduation (d ≈ 0.10). Evidence: moderate to strong.
Supplemental instruction and tutoring
Grade effects of d ≈ 0.29–0.60 in a systematic review, stronger with higher dosage and often for underrepresented students. Evidence: strong in better-controlled studies.
High-impact practices
First-year seminars, learning communities, undergraduate research, internships, and service learning are consistently associated with engagement, learning, and retention, sometimes with larger gains for Black and Latino students. First-year seminars alone show small retention effects (δ ≈ 0.11). Evidence: moderate.
Transitions and entry
Orientation, transfer supports, and clear pathways show promise in descriptive and quasi-experimental work. Critical for first-generation, transfer, and adult learners. Evidence: emerging.
Basic needs and mental health
National surveys often find about a third of students food insecure and nearly half housing insecure, with dose-response links to lower GPA and attrition. Counseling reduces distress and is associated with better academic functioning. Evidence: strong correlational; moderate for counseling.
Career and experiential learning
Paid internships are linked to higher early-career earnings, about 15 percent or roughly $3,000–$5,500 one year after graduation in matched analyses, and better job quality. Other experiential forms improve perceived preparedness more variably. Access is unequal. Evidence: moderate.
Belonging and validation
Small-to-moderate correlations with motivation, self-efficacy, and engagement; weaker with performance and retention. Brief social-belonging interventions can narrow gaps. Evidence: moderate correlational; emerging causal.
Data systems and early alert
Predictive models identify students reasonably well. Outcomes are mixed: alerts help only when paired with high-quality human action. Georgia State’s analytics-informed microgrants are a positive example. Evidence: moderate for prediction; mixed for impact.
Leadership, policy, and culture
Leadership commitment, staff development, policies such as registration holds, and shared responsibility enable everything else. Evidence: strong practical consensus; limited causal research.
Cross-cutting supports
Case management and coordinated care networks assign ownership and route students across silos. Assessment, external partnerships with K-12, employers, and community groups, and physical spaces for belonging and accessibility hold the system together. Evidence: strong for comprehensive case management; emerging for full coordinated care networks.