A Big Four audit firm is offering universities the chance to profile students by age, social background and nationality to judge how likely they are to drop out of study.
Deloitte has said that UK universities could use this information to provide more help to students at risk of dropping out, or to recruit those who are most likely to complete a course - raising the prospect that universities could try to enrol particular demographic groups.
The firm said it gained the insights from an analysis of students who left the business school of a large Australian university. Its research looked for a correlation with factors including satisfaction, academic performance, nationality, average rents and low socio-economic status.
In an interview with Times Higher Education, Julie Mercer, head of education at Deloitte, said that the information would give an institution more "data points" in its admissions process than the GCSE results, A-level grades and personal statement now used.
Deloitte's data could allow an institution to conclude that "this student, with this background, from this school is more likely to do well at this university", Ms Mercer said.
Asked whether there was a risk that universities might accept fewer poor or ethnic minority students if they were found to be more likely to drop out, she said there was always a chance that data would be used "inappropriately".
Deloitte has stressed that the service is intended primarily to help universities support students at risk of dropping out so that they complete their courses. For example, it could be used to flag up a student who had failed to hand in two pieces of coursework on time. If the student were deemed at risk because of his or her background, the university would be better prepared to intervene, Ms Mercer explained.
"The data help them [universities] to know what kind of wrap-around services [are needed]" to support different types of students.
They could also help to "improve the commercial viability of an institution", she noted.
Almost 10 per cent of young first-degree students from areas with the lowest levels of higher education participation drop out in the first year, according to 2009-10 data from the Higher Education Statistics Agency, compared with 7.2 per cent across all students.
Roger Brown, professor of higher education policy at Liverpool Hope University, said there was a danger that universities might use the data to focus their marketing on a demographic that was less likely to drop out, although he thought it unlikely that many institutions would do so.
But he noted that there were only a "limited number" of things universities could do to retain people who face "outside pressures" such as employment or children.
David-Hillel Ruben, a visiting professor of philosophy at Birkbeck, University of London, cautioned against "fetishising" retention rates. Every university had students "who should not be there" because they did not take the course seriously but who had stayed on "because [of] the university's commitment to retention rates", he said.
Paul Temple, a reader in higher education management at the Institute of Education, University of London, professed surprise that Deloitte thought that it could "make money on what is a routine piece of...research for most universities - ie, what are the characteristics of our successful and unsuccessful students?"
A spokesman for the University and College Union said that institutions should provide more pastoral care to students with little family experience of university.
Assess your online reputation
Deloitte is also offering to assess the online profile of a university by "scraping" the internet for comments about the institution on Facebook, Twitter, YouTube and other sites - "where the real conversation is happening", said Danny McConnell, a partner at Deloitte.
The firm has conducted a "student sentiment analysis" for a Russell Group university by searching about 1,500 online repositories for 30,000 pieces of data.
The information was used to create a chart showing the ratio of "highly positive" to "highly negative" comments about teaching quality, research, employability and location, Mr McConnell explained.
He said that although 75 to 80 per cent of comments about the university were positive, the survey revealed that students associated it with elitism. "The view wasn't really what the university would have expected."
A traditional perception questionnaire completed by students had failed to register the perception of elitism and other criticisms found in comments online, he explained.
Mr McConnell said that such views expressed in comments online were likely to have only a "minimal" influence on whether prospective students would apply to the top 10 universities in the UK, "but if you look at the middle ... they are the ones that are going to have to stand out".
He added that the service would cost a university "tens of thousands" of pounds rather than "hundreds of thousands".