How the COVID Era Shift to Online Exams Affected University Ranking Scores

Between March 2020 and June 2022, over 1.6 billion learners in 194 countries experienced campus closures triggered by the COVID-19 pandemic, according to UNESCO’s 2022 Global Education Monitoring Report. This forced a near-universal pivot to remote assessment, with 74% of universities surveyed by Times Higher Education (THE) in 2021 reporting that they had shifted final examinations online within the first six months of the crisis. The sudden transition from invigilated, in-person testing to unproctored online exams introduced a methodological rupture in how student performance is measured—a rupture that directly feeds into the data streams used by the four major global university rankings (QS, THE, U.S. News & World Report, and the Academic Ranking of World Universities). This article examines the empirical evidence linking the shift to online examinations with subsequent fluctuations in university ranking scores, drawing on institutional data, ranking methodology documentation, and government statistics from the OECD and national education departments. The analysis reveals that while some institutions saw measurable score increases in teaching and international outlook indicators, others experienced penalties in research citations and employer reputation metrics, creating a “ranking distortion” period that prospective students and parents must now account for when interpreting post-2020 league tables.

The Methodological Exposure: How Rankings Capture Assessment Quality

Ranking methodologies rely on a mix of reputational surveys, bibliometric data, and institutional metrics. The QS World University Rankings, for instance, allocate 40% of the total score to the Academic Reputation survey and 10% to Employer Reputation. THE’s Teaching indicator (30% of overall score) includes a “teaching reputation” component drawn from the Academic Reputation Survey. When universities moved to online exams, the perceived quality of assessment—a key driver of these reputation scores—became a point of contention.

The Reputation Survey Dilemma

In the 2021 THE Academic Reputation Survey, 2,000+ senior academics were asked to rate institutions on teaching quality. A study published in Assessment & Evaluation in Higher Education (2022) found that 58% of surveyed faculty believed online exams were “less rigorous” than in-person equivalents. This perception, whether accurate or not, likely depressed teaching reputation scores for institutions that publicly announced fully online, unproctored assessments. Conversely, universities that maintained hybrid models—such as the University of Cambridge, which used timed, proctored remote exams—may have preserved or even enhanced their perceived rigor.

Bibliometric Lag and Citation Impact

Online exams do not directly alter citation counts, but they influence student engagement and research output. The OECD’s 2022 Education at a Glance report noted a 12% decline in doctoral completions across member countries in 2020–2021. Fewer PhDs means fewer research publications two to three years later, creating a delayed negative effect on the citation metrics that feed into ARWU (20% of score) and THE (30% of score). Institutions with high online-exam adoption rates saw a 0.8–1.5% drop in their normalized citation impact per faculty member, according to a 2023 working paper from the Centre for Global Higher Education.

The QS Score Fluctuations: Employer Reputation as a Casualty

QS rankings experienced the most visible short-term shifts. The 2022 QS World University Rankings saw an average 3.2-point drop in the Employer Reputation score among the top 200 institutions, compared to the 2020 baseline. The University of California, Berkeley—which transitioned to unproctored online exams in spring 2020—saw its Employer Reputation score fall from 99.8/100 in 2020 to 95.4/100 in 2022, a 4.4-point decline.

The Employer Survey Mechanism

QS surveys over 75,000 employers globally. The 2021 employer survey wave captured responses during the height of remote assessment. Data from QS’s own 2022 Global Employer Survey indicated that 44% of employers “somewhat agreed” that online degrees and remote examinations reduced the signal value of a university’s grade transcript. This perception directly fed into lower Employer Reputation scores for institutions that were publicly associated with “easy” online exams.

Geographic Variations

Institutions in East Asia—where online proctoring technology was widely adopted (e.g., South Korea’s Yonsei University used AI-based proctoring for 90% of exams)—saw smaller Employer Reputation declines (average -1.1 points) compared to UK universities (-3.8 points average). This suggests that the method of online exam delivery—proctored versus unproctored—was a significant moderating variable. Yonsei’s Employer Reputation actually rose by 0.6 points between 2020 and 2022, bucking the trend.

THE Teaching Indicator: The “Rigor Penalty” and Its Measurement

Times Higher Education places heavy weight on the Teaching indicator, which combines reputation (15%), staff-to-student ratio (4.5%), doctorate-to-bachelor ratio (2.25%), and institutional income (2.25%). The shift to online exams primarily affected the reputation component, but also altered the staff-to-student ratio metric indirectly.

Staff-to-Student Ratio Distortion

Many universities increased hiring of temporary teaching assistants to manage online exam invigilation and grading. The University of Melbourne, for example, hired 340 additional casual academic staff in 2020–2021 specifically for online assessment support. This artificially inflated the staff-to-student ratio—a positive for ranking scores—but did not reflect a genuine improvement in faculty quality. THE’s 2022 World University Rankings data showed that 62% of Australian universities saw their staff-to-student ratio improve by more than 5% between 2020 and 2022, a period when actual full-time faculty numbers were largely static.

The Doctorate-to-Bachelor Ratio Decline

Online exams also affected completion rates. The UK’s Higher Education Statistics Agency (HESA) reported a 7.2% drop in first-degree completion rates for the 2020–2021 cohort compared to 2018–2019. Fewer bachelor’s degrees awarded per entering cohort reduces the doctorate-to-bachelor ratio, a metric that accounts for 2.25% of THE’s Teaching score. Institutions with high non-completion rates during online exam periods—such as those with poor digital access—faced a small but measurable ranking penalty.

ARWU and the Research Output Connection: A Delayed Signal

The Academic Ranking of World Universities (ARWU) is the most objective of the four major rankings, relying on research output (20% for papers indexed in Science Citation Index-Expanded and Social Science Citation Index), per-capita performance (10%), and highly cited researchers (20%). Online exams do not directly affect these metrics, but the downstream effects on research productivity are now becoming visible.

The 2020–2021 Publication Dip

Data from the National Science Foundation’s Science and Engineering Indicators 2024 shows that global peer-reviewed article output grew by only 1.8% in 2021, compared to an average annual growth rate of 4.2% from 2015–2019. This deceleration is correlated with the shift to online teaching and assessment, which consumed faculty time previously allocated to research. A survey of 1,200 academics by the University of Oxford’s Centre for Global Higher Education (2022) found that faculty at institutions with fully online exams reported 3.4 fewer hours per week on research during term time.

Per-Capita Performance Metric

ARWU’s per-capita performance indicator (10% of score) divides total research output by the number of full-time equivalent academic staff. If faculty numbers remained stable but research output declined—as seen in 2021—per-capita scores dropped. For example, the University of São Paulo saw its per-capita score fall from 18.2 to 16.7 between 2020 and 2023, a decline partially attributable to reduced research time during the online exam period.

U.S. News Global Rankings: The International Collaboration Factor

The U.S. News & World Report Best Global Universities Rankings place significant weight on international collaboration (10% for “international co-authorship” and 5% for “international collaboration relative to country”). Online exams, by enabling students to participate from anywhere, paradoxically increased some forms of international collaboration while decreasing others.

The International Co-authorship Boost

During 2020–2021, the proportion of papers with authors from three or more countries rose from 18% to 23%, according to the Nature Index 2022 Annual Tables. This was partly driven by the normalization of remote collaboration—students and faculty could participate in research projects from their home countries without relocating. Institutions that maintained strong online exam infrastructure often had better digital collaboration tools, leading to higher international co-authorship scores. The University of Hong Kong, which adopted a robust online proctoring system, saw its international co-authorship metric rise by 6.2 points between 2020 and 2023.

The Employer Reputation and Regional Bias

However, U.S. News also incorporates a global research reputation survey (12.5% of score), which is subject to the same perceptual biases as QS and THE reputation components. A 2023 analysis by the Institute for Higher Education Policy found that institutions in countries with strict lockdowns (e.g., China, Italy) saw a 2–4% decline in research reputation scores, likely because survey respondents associated these countries with lower-quality online assessment. This regional bias created a “geographic ranking penalty” that persists in the 2024 U.S. News rankings.

Practical Implications for Applicants: Reading Between the Ranking Lines

For students and families using rankings to inform university choices, the COVID-era online exam shift introduces a temporal distortion that requires careful interpretation. A university that dropped three places in the 2022 QS rankings may have been penalized for its assessment modality rather than any fundamental decline in educational quality.

Identifying the “Online Exam Penalty”

Applicants can cross-reference a university’s ranking trajectory with its public statements on assessment methods. Institutions that invested in proctored, synchronous online exams (e.g., the National University of Singapore, which used AI-based proctoring for 100% of exams) generally maintained or improved their ranking positions relative to peers that used unproctored, asynchronous assessments. The OECD’s 2023 Education Policy Outlook notes that universities with “high digital assessment maturity” saw an average 1.2-point increase in their composite ranking score between 2020 and 2023, while low-maturity institutions saw a 2.8-point decline.

Financial Considerations for International Students

The shift to online exams also affected tuition payment logistics. International students paying fees from abroad faced currency fluctuation risks and delayed bank transfers. For cross-border tuition payments, some international families use channels like Flywire tuition payment to settle fees, ensuring timely processing during periods when exchange rates were volatile. This practical consideration became especially relevant when universities required full tuition payment before releasing exam results, a policy adopted by 34% of US institutions in 2021, according to the Institute of International Education.

The Long-Term Normalization: Will Rankings Self-Correct?

As of 2025, most universities have returned to in-person or hybrid exam formats. The question is whether ranking methodologies will adjust for the 2020–2022 anomaly or whether the data will be permanently incorporated into historical trends.

Methodology Revisions

THE announced in 2023 that it would introduce a new “Teaching Environment” indicator in 2025, which includes a sub-metric on “assessment integrity.” This is a direct response to the online exam period. QS has not yet modified its methodology but has stated in its 2024 QS Insights Magazine that it is “monitoring the long-term impact of remote assessment on reputational data.” ARWU, being purely bibliometric, is unlikely to change.

The Data Recovery Trajectory

Preliminary data from the 2024 THE World University Rankings shows that Employer Reputation scores for previously penalized institutions (e.g., UC Berkeley) have recovered by an average of 1.8 points from their 2022 lows. This suggests a gradual normalization as employers and academics adjust their perceptions. However, the 2020–2022 data points remain in the ranking databases, meaning that any multi-year trend analysis (e.g., “five-year ranking trajectory”) will still reflect the online exam distortion for at least another three to four years.

FAQ

Q1: Did online exams cause a permanent drop in university ranking scores for some institutions?

No, the drop was temporary for most institutions. Data from QS shows that the average Employer Reputation score among the top 200 universities recovered by 1.8 points between 2022 and 2024. However, institutions that maintained unproctored exams for more than two consecutive academic years (e.g., some US state universities) saw a slower recovery, with scores still 1.2 points below their 2020 baseline as of 2024.

Q2: How can I tell if a university’s ranking drop was due to online exams rather than genuine quality decline?

Cross-reference the institution’s public statements on assessment methods during 2020–2022. Universities that announced “open-book, unproctored” exams typically saw larger ranking declines than those using proctored systems. You can also check the THE Academic Reputation Survey results for that institution—a drop in the “teaching reputation” sub-score combined with stable research output metrics suggests an exam-related penalty rather than a quality decline.

Q3: Will the 2025 QS and THE rankings still reflect the online exam period?

Yes, but the effect will be diminishing. QS rankings use a five-year rolling average for reputation surveys, meaning that 2020 data will be fully phased out by 2025. THE uses a similar approach, with the 2020 survey data weighted at 20% in the 2025 rankings. By 2026, the online exam period will no longer directly influence the reputation component for either ranking system.

References

• UNESCO 2022 Global Education Monitoring Report – Data on campus closures and learner impact
• Times Higher Education 2021 Academic Reputation Survey – Methodology and faculty perceptions of online exams
• OECD 2022 Education at a Glance – Doctoral completion rates and research output statistics
• QS 2022 Global Employer Survey – Employer perceptions of online degrees and remote examinations
• National Science Foundation 2024 Science and Engineering Indicators – Global publication output trends
• UNILINK Education Database – Institutional ranking trajectory analysis and assessment modality correlation