The Power of Personal Endorsements

Many college applicants agonize over whom to ask to fill out formal recommendations, typically from high school teachers or, for graduate programs, from college professors.

The annual report from the National Association for College Admission Counseling consistently says that the top factors in admissions are grades, the rigor of the curriculum and standardized test scores. But 11 percent of admissions leaders surveyed said teacher recommendations had considerable importance on making decisions, and another 46 percent said that these recommendations were of moderate importance.

But what if another kind of recommendation may have more clout?

A new study published in the journal Administrative Science Quarterly suggests that informal endorsements -- when someone reaches out and offers praise for a candidate -- may have more clout than personal letters of recommendation. The study was based on admission to an M.B.A. program and so results could be different in other graduate and professional fields and in undergraduate admissions. But the findings suggest that not all recommendations are alike.

The study examined 21,324 applicants to a full-time M.B.A. program. Of these applicants, 4.8 percent were "endorsed," the term used by the study's authors to describe this kind of recommendation. (The authors are Ben A. Rissing of Cornell University and Emilio J. Castilla of the Massachusetts Institute of Technology.)

The research found that 82 percent of endorsed applicants were interviewed, compared to 34 percent of applicants without an endorsement. Of those interviewed, 64 percent of those with endorsements were admitted, compared to 52 percent of those without them.

Of course one obvious question would be whether the endorsements reflected superior qualifications or just connections? And this leads to the question of whether such admissions endorsements perpetuate the advantages of those (generally wealthy) who have connections.

Significantly, the researchers found that the endorsed candidates were not otherwise superior to those without endorsements. Those who were endorsed did no better later in academic pursuits (as measured by grades and awards) than those who were not endorsed. Those who were endorsed sometimes looked a little better "on paper" than did non-endorsed applicants. But those without endorsements ended up doing better in interviews, on average, than those with endorsements.

There is one area where a correlation existed between being endorsed and later behavior. Those who enrolled with endorsements were more likely than others to be donors, and to be large donors to the university.

The findings left Rissing, one of the authors, concerned about equity issues.

"There are smart, qualified and community-minded individuals in all of these groups. Is their potential lack of awareness of these endorsement channels going to limit their opportunities?" he said in a statement about the research "Decision-makers must be attentive to the reality that access to these types of social connections, such as endorsements, are not ubiquitous."