The ability gap within one university is nearly 8 times the gap between universities

7:21 pm 17/Aug/26

The ability gap within one university is nearly 8 times the gap between universities

VCO Group data from 21,000+ applications shows the ability gap between candidates at the same university is nearly 8 times the gap between universities.

VCO Group's analysis of more than 21,000 Young Talent Program applications in Vietnam shows that the ability gap between candidates at the same university is nearly 8 times the gap between one university and another. University name is therefore useful for allocating attraction budget, but not strong enough to screen individual candidates.

Table of Contents

A target university list is a budget decision, not a screening decision

Almost every Young Talent Program in Vietnam starts with a list of universities. That list determines where the booth goes, where the careers talk happens, where the advertising runs and where the recruiting team invests relationship time. That is a sensible way to allocate resources, because budget and time are finite.

The problem appears when the list takes one step further and becomes a criterion in the CV scoring sheet. At that point the university name stops indicating where to find candidates and starts predicting individual ability. Those are two different jobs, and the data shows university name does the first one well and the second one poorly.

VCO Group operates both the application intake and the aptitude testing systems for a number of Young Talent Programmes in Vietnam. That makes it possible to place side by side two facts each company normally only sees half of: a candidate's CV score, and that same candidate's aptitude test result.

The number that matters: where the variation sits

When total variation in aptitude scores is decomposed into a part explained by differences between universities and a part explained by differences between individuals at the same university, the result is unambiguous.

Decomposition of variation in Young Talent candidate aptitude scores

Source of variation	Relative weight	Implication for screening
Between individuals at the same university	Nearly 8 times larger	Knowing which university someone attended tells you nothing about where they sit among their own classmates.
Between one university and another	The smaller share	Enough to rank attraction investment priorities, not enough to reject a specific individual.
Statistical check	Statistically significant	The difference between universities is real, but far smaller than CV scoring sheets currently assume.

The result holds after controlling for field of study, grade point average and age, to ensure the observed gap does not come from those factors.

The most practical reading is this. Pick one candidate at random from a highly ranked university and one from a lower ranked university, and the chance that the second scores higher than the first remains substantial. Meanwhile, pick any two candidates from the same university and the distance between them is usually much larger than the average distance between two universities.

None of this denies the value of a good education. It simply says that university name is an indirect signal, and once you already have a direct measurement in the form of an aptitude test, the indirect signal should not carry heavy weight.

Nearly half of successful candidates come from outside the leading universities

In the dataset, nearly half of the candidates who pass the aptitude round come from universities outside the top five by application volume. This is not a small overlooked group - it is close to half the market's entire supply of qualified candidates.

More notably, these universities also attract fewer competing employers. Candidates there receive fewer offers, respond more positively to a seriously run programme, and typically cost less per qualified candidate than at target universities where every company concentrates its budget at the same moment.

For a programme with a finite attraction budget, this is a reallocation opportunity rather than a call for unlimited expansion. The cleanest approach is to hold investment steady at the target universities and carve out a small share of budget to test 3 to 5 universities outside the list, then judge the result on test scores rather than application volume.

The GPA requirement published in the job posting

Another filter operates before the CV round even begins: the grade point average requirement stated in the job posting. This does not reject applications; it prevents them from being submitted.

The trade-off in publishing a relatively high GPA requirement

Direction of effect	Measured level	Assessment
Candidate supply lost	More than half self-select out before applying	The single largest filter in the entire process.
Quality gained	Under 5 percentage points on the pass rate	A genuine improvement, but small against the supply given up.
Relationship between GPA and aptitude	Moderate positive correlation	Enough to keep GPA in the scoring sheet at moderate weight, not enough to use as an entry condition.

GPA remains worth collecting. The difference is using it as a scoring criterion rather than as a door closed before the candidate can apply.

International context: the gap between stated policy and practice

The shift from credential-based to skills-based criteria has been discussed internationally for years. Joint research by the Burning Glass Institute and Harvard Business School published in 2024 found that the share of US job postings requiring a degree fell markedly, while the share of people actually hired without a degree barely moved. The researchers estimated that fewer than 1 in 700 hires per year benefited from the change.

That gap does not come from bad faith. It comes from the fact that removing a line from a job posting is easy, while replacing it with a direct measurement you can defend to a hiring panel is much harder. Companies that already have a standardised aptitude test and a score database to compare against find the transition considerably easier.

On selection science, Sackett and colleagues' 2022 paper in the Journal of Applied Psychology recalibrated the entire ranking of predictive validity for selection tools, after identifying a flaw in the range restriction correction applied for decades. The general conclusion is that indirect criteria have lower validity than was long believed, while direct measures of behaviour and ability hold their position.

This does not mean abandoning the CV round

At an aggregate level, the CV round in the data works in the right direction. Candidates rejected at this stage score materially lower on the aptitude test than those who progress. In other words, CV reviewers are seeing a real signal and the round is creating value.

What needs adjusting is the weighting of individual criteria inside the scoring sheet, not the existence of the round. Specifically: reduce the weight on university name and on GPA conditions, increase the weight on descriptions of concrete behaviour and results the candidate has produced, and add a step that standardises scores by age group before ranking.

The cheapest way to verify this is a back-test on your own data: take candidates rejected at the CV round last season, compare against the test results of those who progressed, and calculate how many would-be passers each criterion is costing you.

Three things to do

Separate the two functions of the university list: keep it for allocating attraction budget, remove it as a heavily weighted criterion in CV scoring.
Remove the GPA requirement from the job posting and convert it into a moderately weighted scoring criterion.
Carve out part of the attraction budget to test 3 to 5 universities outside the target group, and judge the result on test score distributions rather than application volume.

Frequently asked questions

Should university name be used as a CV screening criterion?

VCO Group data shows the ability gap between candidates at the same university is nearly 8 times the gap between universities. University name is therefore appropriate for deciding where to invest in attraction, but not strong enough to carry heavy weight when scoring an individual CV.

Should a GPA requirement appear in a Management Trainee posting?

A relatively high GPA requirement causes more than half of candidates to self-select out before applying, in exchange for under 5 percentage points on the test pass rate. The more effective approach is to collect GPA and score it within the scoring sheet rather than publishing it as a mandatory condition.

Where do qualified Young Talent Program candidates come from?

In VCO Group data, nearly half of the candidates who pass the aptitude round come from universities outside the top five by application volume. These universities also attract fewer competing employers, so cost per qualified candidate there is typically lower.

About VCO Group

VCO Group operates the assessment infrastructure behind Young Talent Programmes and Management Trainee schemes in Vietnam, covering application intake, aptitude testing, automated interviewing and Assessment Centres. VCO Group serves young talent and candidates with under three years of experience, working with large enterprises and multinationals in Vietnam. Its assessment and candidate management platform is called Testalents. Head office in Ho Chi Minh City.

If you would like to benchmark your own selection data against the market, or discuss the Young Talent Program Insight Report, contact Mr. Dung on 0334 889 068 or visit vcogroup.com.vn.