Selvin joined the faculty of the School of Public Health at UC Berkeley in 1972, and in 1977, he became the head of its biostatistics division. As the head of the Undergraduate Management Committee, he was instrumental in developing the school's undergraduate program. In addition to his work at UC Berkeley, he also served from 1990 to 1998 as an adjunct professor of epidemiology at the University of Michigan and since 2005 as a professor of biostatistics at the Johns Hopkins University in Baltimore.[1][2]
UC Berkeley bestowed several awards on Selvin for his achievements in teaching. He received the Berkeley Distinguished Teaching Award in 1983 and the School of Public Health Distinguished Teaching Award in 1998. In 2011, at 70, he was awarded a Berkeley Citation. Selvin published over 200 papers and authored several textbooks in the fields of biostatistics and epidemiology.[1]
In February 1975, Selvin published a letter entitled A Problem in Probability in the American Statistician. In it he posed and solved a problem later known as the Monty Hall problem. After receiving criticism for his suggested solution, Selvin wrote a follow-up letter entitled On the Monty Hall Problem, published in August of the same year. This was the first time the phrase "Monty Hall Problem" appeared in print. Selvin proposed a solution based on Bayes' theorem in this second letter and explicitly outlined some assumptions concerning the moderator's behavior. The problem remained relatively unknown until it was published again by Marilyn vos Savant in her column for Parade magazine in 1990. This publication generated a lot of controversy and made the problem widely known worldwide. As a result, quite a few papers were published on the Monty Hall Problem over the years, and it is featured in many introductory probability & statistics classes and textbooks.[3]