Education Case Studies: Introduction | MinSIM | College Decision Making | High School Quality

System Dynamics and Collegiate Decision Making

Problem Statement: Enrollment and financial shortfalls led the Vice President of Academic Affairs at a small women’s college to pose the question:

Might system dynamics provide to the college new and better perspectives for understanding and then for better managing the problem?

Approach: We implicitly drew upon the logic of our Ladder of Engagement:

• To engage a disparate audience of college administrators in sharing their KNOWLEDGE of the problem. What had been the recent, underlying dynamics of student enrollment and finances and how were those dynamics forecast into the future.
• To organize and integrate that disparate knowledge into an UNDERSTANDING of the feedbacks through which those enrollments and finances were connected and exerted mutual control in a set of feedback loops. and
• Finally, through computer simulation and projection of future trends, to identify higher leverage policy options that could INFLUENCE the situation to reverse the recent, worrisome trends.

KNOWLEDGE

What do we know about the “behavior(s)” of the system?

The challenge of defining the problem became obvious in the earliest gathering of college administrators from the Finance, Admissions, Academic Affairs, and Student Life Offices. Each office had distinct mental models of what it needed to do to contribute to the overall health of the college, but none understood how its operation impacted others. In addition, each office could provide time-series data on parameters under their direct control (e.g. basic enrollment and retention data, fund-raising and financial balance sheet data, financial aid distribution), but few data that were related to the concerns of other offices.

What was clear, however, was that the overall dynamics driving the college’s problems transcended any single office’s knowledge base, and these offices had little information on those interactions. This led us to identify initial problems of universal interest and curiosity: How are all these elements integrated into student perceptions of overall “college value,” measured in terms both of costs and benefits (quality). How do those perceptions influence their decisions to matriculate and/or to persist? This focus permitted each office to share information about student numbers, college finance, and curricular and extracurricular quality, in an open and non-threatening manner and in a common effort to better understand the dynamic interplay between them.

UNDERSTANDING

What “drives” the behavior of the system?

Initial collection of data on student inquiries, matriculation, persistence, and graduation rates allowed us to develop a student “aging chain” model (where, over time, students enter, advance or depart, and then graduate). We linked this to a basic budget model (income from tuition and other sources; expenditure to support programs). These two models allowed the group to extrapolate enrollment and finances into the future, albeit without much in the way of dynamic feedback to control those changes.

This model led the group to recognize the need to incorporate a third powerful element into the model. This was student perception of the college’s overall programmatic “quality.” Tangible student enrollments and balance strands interacted with less tangible perceptions of quality and value to describe the complex dynamic world of college administration. Student perceptions of “quality” were dynamic, and could be enhanced by decisions to invest in that quality. Quality and cost to the student then combined to define the “value” of the college. Students then compared the college’s value with that of other colleges to inform matriculation and continuation decisions. That relatively simple concept- or feedback-map (see immediately below) allowed the administrative groups to recognize the challenges in using a limited budget to realize programmatic improvements or tuition reductions in a very competitive recruitment environment. Those realizations also informed our choice of which elements of the system needed especially careful attention, as we converted this conceptual map into a computer simulation that could objectively help evaluate the leverage that could be attained by a variety of potential policies.

Concept Map - College Model:

This was the fundamental conceptual foundation for our work helping an ailing college look for policy leverage in addressing a dangerous financial shortfall. Could we collaboratively build a structure that captured our mental models of how students, dollars, and the less tangible “quality” or “value” aspects of college interact? To walk quickly through the map:

Enrolled students (A) pay tuition (B) that, along with other sources generates income (C) into the college’s general fund (D). Fixed obligations contribute to the overall expenses (E) that drain those funds. Should funds remain after those expenditures, they can be used to fund projects that enhance the quality of the college (F). That quality, which can degrade in the absence of conscious enhancement, contributes, along with the cost of the education (B) to the students’ perception of college value (H). That determination is not made in a vacuum, but is compared to the value of other competing colleges (other H) to define the student’s probability of enrolling or remaining at the college (I), thus closing the loop.

This map helped focus conversations on possible policies to alleviate the college’s cash flow problems: What are the issues (costs and benefits) around raising or lowering tuition? Can quality be enhanced without counteracting tuition increases? How valid is it to think of the student body as a uniform whole when considering which of a palette of enhancements to invest in? How critical is it to ‘outsmart’ the competition?

INFLUENCE

Can this understanding be used to generate better policies to manage the system?

Integration and activation of this final element of “comparative value” with the student and finance strands offered an opportunity to simulate a variety of scenarios. These explored managerial strategies the college might implement in the face of strategies or actions undertaken by peer institutions. Sadly, what became clear in using the model to simulate various strategies was the realization that the time for undertaking corrective measures had effectively passed; we were unable to discover approaches to overcome the college’s significant inertia that resists rapid redirection. Within eighteen months of completing the work the college was forced to close. Lessons learned from the experience, however, continue to guide participants, many of whom continue their academic careers in other institutions.