Statistics Dep’t Founder Dead at 89

C. Frederick Mosteller, the founder of the Harvard’s Statistics Department, died of sepsis on July 23 at Powhatan Nursing Home in Falls Church, Virginia. He was 89.

Mosteller first came to Harvard in 1946 as a professor in the Department of Social Relations, and he remained on the faculty of the University for the rest of his academic career. He helped found the Stats department in 1957, and in doing so, helped distinguish statistics as a distinct field of study across academia. As its first chair, he led the department for over a decade.

“When he was a student, major universities did not have statistics departments, only mathematics departments,” said his daughter, Gale Mosteller. “Now major universities typically have both.”

In addition to establishing statistics as its own respectable field at universities, he worked to introduce statistics in high schools. He helped write teachers’ manuals in the field and taught a televised course on statistics in NBC’s “Continental Classrooms” in 1961.

He later chaired the Department of Biostatistics at the School of Public Health (HSPH), and both developed and chaired the school’s Department of Health Policy and Management. He also offered classes on the applications and methodology of statistics at Harvard Law School and the Kennedy School of Government.

“As the person who organized and led the Department of Biostatistics at HSPH, he developed one of the strongest departments in the nation,” said Professor of Medicine Howard H. Hiatt ’46, who was HSPH Dean while Mosteller was at the school.

Throughout his life, Mosteller aimed to find practical applications of probability and statistics. Even at a young age, he showed a knack for statistics as he worked on his father’s road construction crew to earn money for college. On rainy days, the crew would play poker—and Mosteller would consistently win.

“As I recall, my father made more money from poker than from working on the road crew,” said Gale Mosteller. “A clear incentive to learn the odds and play well.”

In college Mosteller became even more interested in probability when he met a math professor, Edwin G. Olds, at Carnegie Mellon University (CMU), where he earned his bachelor’s and master’s in mathematics. Mosteller later went on to receive his doctorate in mathematics from Princeton University.

During his sophomore year at CMU, Mosteller encountered a question in one of his classes that would propel him towards a career in statistics—when three dice are rolled, what is the chance that the sum of the faces will be 10?

Though all the students in the class got the correct answer by counting on fingers, Mosteller wondered how the problem could be solved for more dice than one could keep track of on their hands. Professor Olds later approached Mosteller and invited to teach him how.

“So we went to his office, and he showed me a generating function. It was the most marvelous thing I had ever seen in mathematics,” Mosteller told Donald J. Albers and Gerald L. Alexanderson, the authors of the 1990 book “More Mathematical People.”

It wasn’t long until Mosteller would captivate the minds of his own students.

“My most memorable meeting with Professor Mosteller was my very first meeting with him,” said James W. Vaupel ’67, who was the first student to graduate with an undergraduate Statistics degree from Harvard.

During one of Vaupel’s first meetings with Mosteller, the professor kept spinning a coin and recording whether it came up heads or tails, according to Vaupel.

When Vaupel finally asked Mosteller why, “he gently explained that it was more random to spin a coin than to flip a coin, because a coin spun on a desk would spin many more times than a flipped coin would turn over,” Vaupel said. Mosteller further explained that it was a penny made in 1942, and thus contained alloys other than copper because of the need for copper during the war.

“He showed me the results—the coin came up heads almost two-thirds of the time,” said Vaupel. “[Mosteller] smiled and said, ‘This, Mr. Vaupel, is a very useful coin.’”

In addition to his teaching, Mosteller was an accomplished researcher and writer. He published over 60 books and 300 papers during his lifetime.

Some of his most prominent research includes the Tennessee class size experiment, which studied the effects on students randomized into small and large classes. This experiment showed that small classes enhanced learning, and that the beneficial skills students acquired in small classes continued to be seen even after they returned to large classes. The results of this experiment “reportedly helped spur President Bill Clinton’s request for 100,000 new teachers to reduce classroom sizes,” according to The Washington Post.

Mosteller also worked with Professor David L. Wallace of the University of Chicago to determine who had written 12 of the 57 Federalist Papers by calculating sentence length and the frequency of certain words. They eventually concluded that James Madison had written them.

While Mosteller was a remarkable statistician, his daughter said he would possibly be remembered most for his role as a teacher and the influence he had on his students, because “he made classroom teaching [and] mentoring students a priority.”

“It seemed as though everywhere he traveled, he’d run into a student who had enjoyed one of his classes, whether or not he was a statistics major,” Gale Mosteller said.

Gale Mosteller also said that much of her father’s success was due to the support he received from his family. Both Gale and her brother William helped provide samples for dice experiments when they were young, while Mosteller’s wife, Virginia, “made it her business to support his work.” Gale went on to help more extensively on research and manuscript editing as she got older.