LINKS
ABOUT
SOCIAL
  • White LinkedIn Icon
  • White Facebook Icon
  • White Twitter Icon
  • White YouTube Icon

SG: eugene@talentdatalabs.com

NL: stefan@talentdatalabs.com

SING: (+65) 9047 9451
NL: (+31) 6404 365 40

67 AYER RAJAH CRESCENT

#02-10/17

SINGAPORE

© 2018 Talent Data Labs

Subscribe to our blog!

How much should you care about the university your applicants attended?

 

Attending top universities still matters

Students struggling with challenging selection processes and paying hefty tuition fees know how important attending the right university is. In fact, a research conducted by the Institute for Fiscal Studies, Cambridge University, Harvard University, and the UCL Institute of Education in the UK found that attending a prestigious university is among the key determinants of financial success. Furthermore, robot readers and several application processes require candidates to have attended top universities. However, plenty of bright people can be found in “normal” universities as well. Including university requirements in screening makes the selection process faster and simpler, but exposes organizations to the risk of losing several competitive candidates. One of our clients had a case that  is a good example of how data can be used to find a solution to this problem.

 

Entering an exclusive network for professionals

Our client offers an exclusive online network for students and professionals, and we were asked to investigate whether or not having attended highly ranked universities should be one of the key criteria to meet in order to be accepted in the network. At the moment, admission is based on the screening of CVs, a Video Score and several Psychometric Assessments Scores, with the additional restriction that to enter the selection process applicants need an invite from a member of the network. Our client’s network, as most networks, had the bulk of its members in specific geographic areas and, consequently, universities. The invite rule made this even more evident, as people are more likely to vouch for individuals from their same university.  

 

As a result, the applicant database included universities with a lot of candidates  and others with only a few of them. Therefore, we split the sample of universities into those with a high number (at least 30) of applicants and those with less than 30 applicants. We then compared estimated prestige of universities and acceptance rates of applicants. Note that here prestige is determined by the percentage of research papers of a university that end up published in top journals.

 

A counterintuitive result

Table 1. Summary statistics for the whole sample of universities, and two restricted samples (universities with at least 30 applicants, and less than 30).

 

By looking at the table, we see how universities with at least 30 applicants tend to be more prestigious, with 6.25% of their papers published in journals which are top 5 in their field, compared to 5.62% in the case of their counterparts. Furthermore, and perhaps surprisingly, it appears that people applying from universities outside the network enjoy higher acceptance rates. In other words, this preliminary analysis suggests that applicants from top universities are less likely to be admitted than those coming from less renowned institutions.This seems counterintuitive, and once we took the analysis one step further we realised that there is more to the story than that.

 

University prestige is not the only factor

Table 2. Correlations between acceptance rates and university prestige for the whole sample and the two subsamples.

 

Correlations between university prestige and rate of admission is significantly higher for universities with at least 30 applicants than for the other ones (0.487 against 0.184 according to our preferred measure). Thus, it seems that the prestige of your university is predictive of admission only for candidates applying from universities where the network already has a high number of applicants, and less so for the others.

 

But how can we make sense of such difference? A reasonable explanation seems to be that people from outside the network are way more unlikely to be invited in the first place.The sole fact that one of them has found out about the network and managed to get an invite implies some merit on his or her side. He is probably somebody that looks out for professional opportunities and is career-focused. In other words, people that reach the assessment stage from outside the network are probably among the best and most determined people of their university, and thus more likely to be admitted.

 

Leveraging data to find the perfect selection criteria

What would then be the best way for our client to attract talented people in its network? Probably, it would be to assign a dynamic weight to the criterion of coming from a top university. The importance of such criterion, in fact, should be increasing in the number of applications coming from the assessed person's university. Furthermore, we would suggest taking in higher consideration applications coming from people who are students or alumni of universities with very few applicants, for the reasons reported above. Our team of data scientists is currently working on the implementation of these features, which requires a good amount of technical knowledge.

 

In conclusion, this case exemplifies why stating in your posting that only applications from top universities will be considered may not be a good idea. Top universities have tough selection processes, and consequently they are excellent pools of talent. Still, there are plenty of brilliant people that are not enrolled in them because of their high costs, or because of a socioeconomic background that simply did not encourage them to apply. Giving these candidates a chance might turn out to be the key to hire great professionals. Also, you may end up doing the entire society a favour by increasing social mobility, as top schools are often attended by the richest students.

 

 

If you are interested in finding out how to create more efficient selection criteria for your company, or you want to know more about how to implement data analytics to talent management, you can contact us at insights@talentdatalabs.com.

 

More articles are coming soon! You can subscribe to our newsletter to not miss them. Or follow us on social media: 

 

YouTube: https://www.youtube.com/channel/UCDAh5_jbw0XltpxlerVP9_w 

Facebook: https://www.facebook.com/TalentDataLabs/

Twitter: https://twitter.com/TalentDataLabs1

LinkedIn: https://www.linkedin.com/company/18306801/

Medium: https://medium.com/@talentdatalabs

 

 

Share on Facebook
Share on Twitter
Please reload

GENDER INEQUALITY AT WORK DEPICTED IN 4 CLEAR GRAPHS

August 15, 2019

1/2
Please reload

RECENT POSTS
Please reload

ARCHIVE
Please reload

FEATURED POSTS