Learn about diversity at ASEE
ASEE would like to acknowledge the generous support of our premier corporate partners.



Guessing Games

Core engineering knowledge can’t be outsourced to a crowd.

Photo: HENRY PETROSKI - There are limitations to, if not the downright undesirability of, design by committee. On the opening day of class, I often ask students to state the height of the Eiffel Tower or a characteristic dimension of some other famous structure.

Taking their silence as an under-appreciation of the scale of structural engineering achievements, I ask each of the students sitting around the seminar table for their guesstimate. The most recent set of answers I recorded for the tower’s height are typical: They ranged from 450 feet to 1,500 feet, resulting in an average value of 846 feet, which is 15 percent lower than the tower’s 990 feet (300 meters – we Americans tend to think metric only parenthetically).

My exercise is a variation on the jelly-bean experiment described in a recent issue of Modern Steel Construction by Jason Cohen, author of Best Kept Secrets of Peer Code Review. According to Cohen, who is also a marketing and small-business blogger, the experiment gained fame in 1987 through the work of financial economist Jack Treynor.

Treynor showed students a large jar of jelly beans and asked for guesses of their quantity. The average of 56 responses was only 2.5 percent off from the correct answer of 850. Twenty years later, at the Columbia Business School, Michael Mauboussin repeated the experiment with a larger class and a larger jar. The guesses ranged from 250 to 4,100, and the average was 3 percent off from the actual number of 1,116.

Cohen’s objective in reporting on these experiments and their results, which ostensibly show that “the crowd is smarter than the individual,” is to debunk this “wisdom in the crowd” in applications that go beyond factual guesstimating. In particular, he questions the value of focus groups and believes that “design by committee is the antithesis of innovation.” He convincingly argues that identifying “average products designed for the average consumer is the path to small-business failure.”

According to Cohen, crowd wisdom works only when erroneous guesses cancel out, leaving a surprisingly accurate estimate of jelly beans in a glass jar or feet in a tower’s height. But in the case of seeking creative advice, “votes eliminate the interesting edges, leaving only the boring residue.” He concludes, “That’s not how great products are made, or how great structures are designed and built.”

My purpose in asking students the size of a famous structure had not been so profound. I mainly wanted to emphasize to them what they did not know about even some of the world’s most famous engineering accomplishments. This lesson, I have hoped, would motivate them to pay attention to the quantitative as well as the qualitative aspects of the many structures we were going to discuss throughout the semester.

I have in addition wanted to emphasize that the distance from the ground to the top of the Eiffel Tower or between the towers of the Golden Gate Bridge was not just a number. The 300-meter height of the Eiffel Tower is not just a measure of its size; it was also a vertical goal that engineers had been striving for from the early 19th century. And the bridge’s 4,200-foot span was a measure of the technological achievement that allowed the Golden Gate to be the greatest in the world for almost three decades. We should know these things not as disembodied numbers and factoids but as fundamental ingredients of historical and technological literacy.

I will continue to introduce my students to the quantitative side of structures by testing their ignorance, but in the future I will also use the results of guessing games as an opportunity to say something about the limitations of, if not the downright undesirability of, design by committee.

Henry Petroski is the Aleksandar S. Vesic Professor of Civil Engineering and a professor of history at Duke University. His latest book is The Essential Engineer: Why Science Alone Will Not Solve Our Global Problems.




© Copyright 2010
American Society for Engineering Education
1818 N Street, N.W., Suite 600
Washington, DC 20036-2479
Telephone: (202) 331-3500