By Alessandra King, Posted December 12, 2014 –
How many jelly beans are in the
jar? This looks like a typical Fermi problem.
Fermi questions, such as found on http://www.nctm.org/publications/article.aspx?id=20696,
are open-ended questions that require the solver to use common knowledge and
back-of-the-envelope calculations to estimate quantities about which he or she knows
very little. They are one of my favorite types of problem because they foster
problem solving, estimation, and number sense. (I will discuss them more in my
next blog post.) However, this type of prediction problem also gives us the
chance to do a “social” experiment: What happens when we take into
consideration the results of the “team,” the entire group of students who
participate in the challenge? Now the Jelly Bean Jar problem (and similar Fermi
questions) can become a paradigm for studying teamwork
and collaboration.
The NCTM Standards
and the Common Core Standards, as well as
other frameworks for teaching and learning (for example the Partnership for 21st Century Skills),
mention collaboration and communication as well as problem solving as critical
skills that students should master to function well and succeed in the twenty-first
century. The issues they will have to face are complex: The simple problems have
been solved; they have been left with the difficult ones. Matters such as freshwater scarcity,
gender equality, and space exploration are not
problems that can be solved by any one individual. Therefore, the value of teamwork
becomes ever greater. It is well known that a team or a collection of people
often makes more accurate predictions than the individuals in it. This
phenomenon, called the wisdom of crowds,
depends both on talent and on diversity. In fact “the collective accuracy of
the crowd depends in equal measures on the accuracy of its members and on their
diversity,“ as shown by the Diversity Prediction theorem.
A diverse crowd will always be more accurate than its average member and
sometimes more than any member in the crowd, as Scott E. Page explains in this video. Diversity here is
intended as cognitive diversity, referring
to the differences in how we think—the categories and models we use to encode
problems, the tools we employ, and the diverse perspectives we apply to solve
them. Cognitive diversity is connected with the process engendered by different
training, experiences, interests, and outlooks.
So how did my middle school students
fare at predicting the number of jelly beans in the jar? Well, there were about
80 individual guesses, ranging from 168 to 8,000, based on a large variety of
assumptions and approaches. The average of the team was closer to the real
number of jelly beans (1,817) than 85 percent of the individual estimates.
Although often the “crowd” does
very well indeed, we do need to be aware of the limitations
of group decision making for teamwork to be as powerful and effective as it can
be. Fostering diversity and independent contributions and avoiding “group
think” and overconfidence can lead to better decision making. Jelly beans
provide lots of food for thought!
Alessandra
King, Alessandra.king@holton-arms.edu, studies
mathematics with her students at the Holton-Arms School in Bethesda, Maryland.
She has taught mathematics and physics at the middle school and high school
levels and is interested in creative problem solving, critical thinking, and
quantitative reasoning.
Archived Comments
Cool problem! I would love to see a data distribution of your students' guesses. Think of all the data analysis we could do... :) Posted by: KatieH_46014 at 12/19/2014 12:47 PM
|
This is definitely a great idea. It would provide another way to look at this problem and a cool application of data analysis and statistics for the students. Once the data collection is set up properly, one could study the results in relation to students' age, grade, current math class and so on. Posted by: AlessandraK_28293 at 12/23/2014 2:35 PM |