Campus Speech Code
A decision analysis exercise where a college president chooses to abolish or maintain a speech code, evaluating outcomes in campus disruptions, state funding, and enrollment using weighted utility analysis.
THE IDEA
EXERCISE 48 Campus Speech Code
You are the president of a state-funded college and are strongly opposed to the college’s current policy of disciplining students and faculty for insensitive speech. In the wake of several controversial, widely publicized incidents in which both students and faculty members at the college were disciplined for insensitive speech, you are considering whether to abolish or continue the college’s speech code. You are mainly concerned about three things:
- Whether your decision will prompt public demonstrations and other disruptions of campus activities (2 in the supplied prompt; the worked analysis below uses a different notation).
Three outcomes: no disruptions, peaceful disruptions, and violent disruptions. The probability for each outcome, if the code is abolished, is .1, .7, and .2, respectively. The probability, if the code is continued, is .8, .2, and 0, respectively. 2. How the state legislature will react to the decision; that is, how state funding (which covers two-thirds of the college’s annual budget) might be affected (.5).
Three outcomes: fewer funds, same funds, more funds. The probability for each outcome, if the code is abolished is .1, .6, and .3; if continued, .3, .5, and .2. 3. How the decision will affect student enrollment (.3).
Three outcomes: smaller enrollment, same enrollment, greater enrollment. If the code is abolished, the probabilities are .1, .3, and .6; if continued, .3, .5, and .2.
Using utility analysis with multiple classes of outcomes, analyze the problem and determine which option is preferable. Assign utility values as you see fit.
I analyzed the problem as follows. In Step 1, I identified three classes of outcomes (campus disruption, state funding, and student enrollment) and weighted them .2, .5, and .3, respectively. In Step 2, I constructed an identical utility matrix for each outcome class (Figure 16-16), identifying the perspective in each matrix as the college president and listing the two options. I then performed Steps 3 through 6 with each matrix (Figure 16-17), assigning utilities and probabilities, computing expected values, and entering the totals in the “Total EV” columns. In Steps 7 through 11, I constructed a merged matrix (Figure 16-18), entered the total expected values from the three classes-of-outcomes matrices, multiplied these values by their respective class weights, entered the products in the “Total Weighted EV” column, and determined from the ranking that abolishing the campus speech code was the strongly favored option. In Step 14, I asked myself whether the results made sense intuitively, which I deemed they did.
The referenced matrices and assigned utilities are not included in this excerpt. The author’s ranking therefore cannot be reproduced from the probabilities alone; your ranking will depend on the utilities you assign.
When to use it
A business leader deciding on a new product launch, considering market reception, cost, and timeline.
A government official choosing between two policies, evaluating impact on different stakeholder groups.
A student choosing a college, weighing academic programs, location, and cost.
FROM THEORY TO EVERYDAY LIFE
In practice
Campus Speech Code Problem
A college president compares abolishing and continuing a speech code across campus disruptions (weight 0.2), state funding (weight 0.5), and student enrollment (weight 0.3). The author reports that abolishing the code ranks higher using his chosen utilities. The excerpt supplies probabilities but omits the matrices containing those utilities.
A MOMENT TO REFLECT
Put it to work.
How would you assign utility values to each outcome? What rationale would you use?
How would the decision change if the weights for the outcome classes were different?
What other factors might a college president consider that are not captured in this model?
CONTINUE THE CONVERSATION
Source texts
- The Thinkers Toolkit by Morgan D Jones