Should we change our tuition?
What events are most appealing to alumni?
Should we offer our new program online or on campus?
These are all questions that discrete-choice modeling (DCM) can answer better than direct, standalone survey questions. DCM looks at several factors within a decision RELATIVE to each other, forcing respondents to make tradeoffs among different options. What we learn from this type of data allows us to gain deeper insight into which factors our audiences find most important.
Here is an example of DCM and how it works.
Imagine you are trying to determine the optimal way to offer a new certificate program. You are trying to figure out how much to charge, whether to offer the program online or in-person, and how interested people are in enrolling at your school versus other institutions.
One possibility is to just ask prospective students these questions directly. For example, you could ask them, "How much tuition would you prefer we charge?" and give them options for $5,000, $10,000, and $15,000. The problem? Everyone will choose the lower price.
That's where DCM can be more effective. When setting up a DCM question to determine the optimal certificate program, we would first determine which factors we think will contribute to prospects' decision to enroll in the program. Let's say we decide to focus on the following three factors: tuition, school, and program format.
Then, within each of these factors we determine specific levels to test:
- Tuition: $5,000, $10,000, $15,000
- School: School 1, School 2, School 3, and School 4
- Format: in-person, online, or hybrid
Each survey respondent is then shown multiple combinations of factors and levels and asked to choose which they prefer. For example, we might ask:
- Which of the following programs would you choose? (pick one)
- Online program at Awesome University, tuition $5,000
- In person program at Fantastic College, tuition $10,000
- Which of the following programs would you choose? (pick one)
- In-person program at Super University, tuition $15,000
- Online program at Awesome University, tuition $10,000
And so on.
The analysis of all the responses allows us to answer questions including, “Do prospects prefer the higher cost at a smaller institution or a lower cost at a larger institution?” or “Do they prefer to go to a less expensive school offering an online program or a more expensive school with in-person classes?”
The analysis provides the relative weight of each factor in respondents’ decision-making process. In our certificate program example, DCM might generate the following importance levels:
School: 45% importance
Price: 35% importance
Format: 20% importance
From there, we can also calculate the order of preference of the various levels within each factor. For example, for program format, we might find that in-person classes are students’ most preferred option, followed by hybrid and then online.
So, in summary, we know that prospects will choose the least expensive option if they are asked solely about price. But by using DCM in our example above, we forced respondents to make trade-offs on price vs. other factors and learned that price was NOT the most important factor — the institution was. When forced to make trade-offs, people will sometimes be willing to pay more to get certain offerings that are more important to them. This allows us to get a much more realistic view of what people will do in the real world.
This is just the beginning of the insights DCM can uncover. It can get much deeper and even simulate different scenarios to determine the optimal bundle to offer given competition, cost, market size, etc. Food for thought as you design your next survey!
