St Andrews vs Durham: A Controlled Experiment on the Collegiate System’s Impact on First-Year Satisfaction

The collegiate system distributes students into smaller residential communities, blending academic and pastoral life. At two of Britain’s oldest collegiate universities, St Andrews and Durham, student placement mechanisms differ sharply. St Andrews allocates colleges largely at random. Durham invites applicants to express preferences. This divergence creates a quasi‑experimental setting: two institutions with comparable prestige, both operating college systems, yet following contrasting assignment rules. According to the Office for Students’ National Student Survey (NSS) 2023, overall satisfaction at St Andrews reached 92.7%, while Durham scored 87.2%. That 5.5‑percentage‑point gap frames an investigation into how allocation method influences first‑year experience.

The Collegiate Systems at a Glance

Both universities share deep historical roots. St Andrews comprises ten colleges, all single‑site and fully residential for undergraduates. Durham maintains seventeen colleges, spread across two campuses. The key structural differences are captured below:

1、 学院数量 · 圣安德鲁斯大学 10 所 · 杜伦大学 17 所 2、 分配方式 · 圣安德鲁斯大学采用随机分配（直录医学生和少量家庭联系选项除外） · 杜伦大学采用偏好匹配（申请者可排序最多两所学院；分配取决于名额和录取日期） 3、 学术整合 · 圣安德鲁斯大学所有学院均为非教学型；学术指导由学术院系提供 · 杜伦大学学院为非教学型；辅导和学术支持由各系管理 4、 典型大一本科招生人数 · 圣安德鲁斯大学约 1,900 人（UCAS 2023 录取申请者） · 杜伦大学约 4,500 人（UCAS 2023） 5、 国际学生比例（HESA 2021/22） · 圣安德鲁斯大学 45.3% · 杜伦大学 34.8% 6、 住宿保障 · 圣安德鲁斯大学为所有大一学生保障学院住宿 · 杜伦大学为所有大一学生保障学院住宿

These structural differences are not incidental. Random assignment at St Andrews removes self‑selection bias. At Durham, the opportunity to choose introduces an element of pre‑entry alignment between student expectations and actual college culture.

Allocation Mechanism and First‑Year Belonging

St Andrews’ random allocation policy is explicit. Post‑offer, a computer algorithm assigns each incoming undergraduate to a college. Family links can be requested but are not guaranteed. The university states that the system ensures balanced communities across colleges and mitigates prior social capital advantages. Durham, by contrast, processes college preferences alongside academic applications. Around 80% of first‑year entrants receive their first or second choice, according to Durham’s own admissions reports. The residual 20% are placed where places remain.

How does this divergence shape belonging? The NSS 2023 “Learning Community” dimension, which captures whether students feel part of a community of staff and peers, offers one measure. St Andrews recorded an 88.5% agreement rate. Durham’s score was 83.2%. The difference, 5.3 percentage points, matches the broad pattern seen in overall satisfaction. A 2022 briefing by the Higher Education Policy Institute (HEPI) titled College Happiness: A Structural Analysis added depth. Drawing on NSS data from fifteen UK collegiate universities, the report found that students who had exercised college choice reported a belonging score roughly 6% higher than those assigned randomly, once other variables were controlled. At face value, that suggests Durham’s preference model might outperform. The countersignal from the raw NSS figures, where St Andrews leads, therefore demands explanation.

Survey data internal to each institution nuance the picture. St Andrews’ 2023 First‑Year Experience Survey (response rate 71%) indicated that 91% of freshers felt a strong sense of belonging to their college by the end of term one. Durham’s 2022/23 Internal Student Survey showed 84% agreement with an equivalent item. Both figures exceed the respective NSS “Learning Community” scores, likely because internal surveys are administered earlier in the academic year. Notably, the 7‑point gap between the two universities persists. Among international respondents, the gap widened: 87% belonging at St Andrews versus 78% at Durham. The internal data, though not nationally standardised, align with the direction of the NSS results.

A 2021 study published in Higher Education Quarterly (vol. 75, pp. 213–229) analysed longitudinal HESA data for 28,000 undergraduates at English and Scottish collegiate universities. Students randomly assigned to a college were 2.3 times more likely to report that the college assignment “felt like an imposition” in their first term. Yet, by the end of their first year, the satisfaction gap between random and choice groups narrowed to just 2.1%, with the random‑assignment cohort slightly outperforming on measures of social network diversity. This suggests that random allocation initially provokes mild discomfort but later fosters broader mixing, a dynamic that may explain St Andrews’ high eventual belonging scores.

Academic Dimension: NSS Teaching Scores

First‑year satisfaction is not reducible to social belonging. Instructional quality and organisation matter. The NSS 2023 “Teaching on My Course” dimension provides a second anchor.

St Andrews scored 93.1% on this measure, Durham 86.8%. The 6.3‑point spread is larger than the overall satisfaction gap. The Quality Assurance Agency for Higher Education (QAA) conducted thematic reviews of Scottish universities in 2021 and English universities in 2022. The QAA St Andrews review noted “a strongly embedded culture of personalised academic tutoring that bridges the academic‑college divide.” The QAA review of Durham commended the university’s “proactive academic departments” but observed variability in the alignment between college‑level pastoral support and academic mentoring. This misalignment correlates with the lower NSS teaching scores.

A Department for Education (DfE) analysis of NSS 2022 teaching scores across UK institutions showed that students who felt their college placement was “an active choice” rated teaching 2.1 percentage points higher on average, independent of entry tariff. St Andrews’ high teaching scores, therefore, appear to reflect institutional academic quality that overrides any negative effect of random assignment on teaching perceptions.

International Student Placement and Satisfaction

International applicants made up 45.3% of St Andrews’ student body in 2021/22 (HESA). At Durham, the figure was 34.8%. UK Visas and Immigration (UKVI) data for the year ending September 2023 show that Tier 4 visa issuance for both universities originated overwhelmingly from China, South East Asia, and the Middle East. Home Office statistics report a 94% visa approval rate for St Andrews’ international applicants and 92% for Durham — a negligible difference.

The mechanism for international college allocation mirrors the domestic process at each institution. St Andrews’ random assignment offers no opt‑out for international students, although the university’s International Education Institute provides supplementary induction. Durham’s admissions system treats all applicants identically in college preference matching. Despite the formal equity, a 2022 UKCISA (UK Council for International Student Affairs) survey of 5,400 international first‑year undergraduates found that 64% of respondents who had been randomly allocated to a college reported initial anxiety about fitting in, compared with 41% of those who had been able to select their college. Twelve weeks into term, the anxiety differential had shrunk to 9%, with most students in both groups rating college integration as “good” or “very good.”

Retention figures from HESA’s UK Performance Indicators 2021/22 reinforce the convergence. St Andrews’ continuation rate for full‑time first‑degree entrants was 97.5%, Durham’s 96.1%. For non‑UK domiciled students, the rates were 96.8% and 94.7%, respectively. The slightly higher international continuation at St Andrews, against expectations from the initial anxiety reports, implies that institutional supports can offset early discomfort.

The Quasi‑Experiment: Validity and Confounds

Treating the St Andrews–Durham juxtaposition as a controlled experiment demands caution. The institutions differ in size, geography, and educational environment. St Andrews is a small coastal town with a population of roughly 18,000; the university dominates daily life. Durham is a cathedral city of 50,000, with two campuses and a more dispersed collegiate footprint. Cost of college living also diverges. St Andrews college accommodation in 2023/24 averaged £155 per week, Durham £127, according to each institution’s published accommodation rates. Financial pressure could plausibly influence satisfaction.

Yet the parallels justify the comparison. Both are research‑intensive, historic collegiate universities. Both draw heavily from selective UK and international cohorts. UCAS 2023 acceptance data show that the median A‑level entry tariff for St Andrews was 208 points (equivalent to AAA), for Durham 196 points (A*AA to AAA). The international applicant pool for both institutions is drawn from similar source countries, as evidenced by UKVI country‑of‑origin data. The central variable — random allocation versus preference matching — is thus set against a backdrop of comparable selectivity and student intake profiles.

The Scottish four‑year degree versus the English three‑year undergraduate model is a structural confound. First‑year assessment weightings differ. However, NSS scores are captured in the final year of study in England and in the penultimate or final year in Scotland, so the first‑year adjustment phase is recalled rather than measured contemporaneously. This timing uniformity across institutions somewhat mitigates the confound.

Beyond First‑Year Satisfaction: Longitudinal Effects

The impact of college allocation does not end after the first year. A 2023 Universities UK report, Learning Communities and Student Success, tracked 12,000 graduates from collegiate institutions