Why Your MBA Cohort Matters Most
- Raffles Jakarta
- 3 days ago
- 9 min read
Graduates of a serious postgraduate program often say that what surprised them was not a module. It was the people with whom this phenomenon is associated. This phenomenon is such a ubiquitous finding in alumni testimony that it has become a cliché, and clichés tend to be dismissed rather than examined. The assertion is a mistake here, because that claim turns out to be supported by a substantial body of research in sociology, economics, and organizational behavior, and understanding why it is true changes how a candidate should choose a program.
The point is not the sentimental bit about lifelong friendships. It’s narrower and more consequential: the group you learn with shapes both the quality of your thinking and the structure of the professional network you leave with. Neither is a by-product of the curriculum. The aspects of education that cannot be replicated include the network argument, the syllabus alone, and the reason your closest colleagues are the least useful people you know.
The basic finding here is one of the most cited papers in the social sciences. Mark Granovetter was interested in how people actually found jobs and discovered that the useful information rarely came through close ties.
The connection was through acquaintances, people the subject knew a little, saw infrequently, and considered something less than friends. His explanation is structural, not psychological. Close contacts tend to live in your social world and thus have mostly the same information. Weak ties bridge social worlds, and it is over those bridges that truly new information travels (Granovetter, 1973).
Ronald Burt took the logic from information to ideas. In his work with structural holes, he identified managers located between otherwise disconnected groups within organizations and found that they are more likely to generate ideas rated as valuable by senior colleagues and that they are more likely to be promoted and better paid. He reads in a deliberately unromantic fashion: people with networks that span otherwise disconnected groups are not inherently more creative. They are exposed to a broader range of ways of posing a problem, and much of what appears to be creativity is actually the transfer of an idea that is commonplace in one group to a group where it is not (Burt, 2004). James Coleman’s parallel work on social capital made the complementary point that the value of a network is not only in the information that it moves but also the obligations, expectations, and trustworthiness it brings (Coleman, 1988).
When you consider those three findings within the context of a mid-career manager's normal professional network, the implication becomes uncomfortable. Many of us know many people who do more or less the same as we do, in more or less the same sector and in more or less the same city. The network is dense, comfortable, and informationally poor.
Homophily, and why networks do not diversify by themselves
Researchers have well documented the reasons these networks form. One of the most robust regularities in social network studies is homophily, the tendency for association to occur among similar people. It cuts across race, education, occupation, age, and social class so that the information received by any individual is heavily filtered through the similarity of those around them (McPherson, Smith Lovin, and Cook, 2001). Being homophilous is not a moral failing. It’s what people do by default when they meet each other: at work, in the neighborhood, or through existing acquaintances.
If left to its devices, it creates exactly the dense, redundant network that Granovetter and Burt identify as the least useful. That is the particular structural problem to which a proper cohort provides the answer. A program intentionally builds weak ties and structural holes among professionals from different industries, functions, nationalities, and career development stages in a compressed time frame, where continued collaboration requires acquaintances to become substantive. It is difficult to build that in any other way and almost impossible to do that accidentally.
The decision quality argument
The second reason the cohort matters has nothing to do with networks and everything to do with the quality of thinking in the room. Lu Hong and Scott Page have a formal result that states that, under some conditions, a diverse set of problem solvers will outperform a set of the most able problem solvers. The mechanism is that there is a correlation between ability and similarity of approach: the best individual performers tend to converge on similar heuristics, so a group composed entirely of them explores a narrower space of possible solutions than a group whose members approach the problem differently (Hong and Page, 2004). The result is mathematical, not motivational, and the conditions attached are real, but the underlying intuition survives translation: a room where everyone reasons the same way has fewer moves available to it.
The experimental literature on social diversity points the same way, explaining something that surprises people. Katherine Phillips summarized a body of research showing that diverse groups often perform better than homogeneous groups on complex tasks, and they do so while finding the experience less comfortable and being less confident about it. People tend to think they did well in homogeneous groups and poorly in diverse ones. Often the confidence is the inverse of what actually takes place (Phillips, 2014; Rock and Grant, 2016). Inversion is the most practically important finding in this entire literature and it explains a great deal about why organizations underinvest in the conditions that produce effective decisions. The friction is not a bug in the process. Friction's the thing. If you explain your reasoning to someone who does not share your assumptions, then those assumptions become explicit.
What happens without it
The counterfactual is in the literature. Irving Janis’s analysis of groupthink looked at policy failures where cohesive, capable, well-informed groups made disastrous decisions and identified a recognizable syndrome: pressures toward uniformity, self-censorship by doubters, an illusion of unanimity, and the emergence of members who shield the group from disconfirming information (Janis, 1972).
The other half is filled in by Amy Edmondson’s work on psychological safety. Cognitive diversity yields better results only if the minority are prepared to voice their opinions. In teams with high interpersonal risk, diversity exists on paper but disappears during meetings (Edmondson, 1999). So a cohort needs two properties: a real difference in the room and conditions for expressing that difference.
That's why small class sizes are not a luxury. In a group of two hundred, the cost of speaking up is high, and the marginal contribution of any one person is low. Most people remain silent and do not act. In a small enough group that everyone knows each other, silence is noticed, and participation is expected.
An honest note on the evidence
Intellectual honesty requires admitting that peer effects in education are more difficult to measure than it appears because students choose their environments and outcomes are confounded by that selection. The strongest evidence comes from contexts in which peers are randomly assigned. Bruce Sacerdote’s study of randomly assigned university roommates found true peer effects on academic outcomes but also on social behavior, where the effects were specific rather than global and stronger for some outcomes than others (Sacerdote, 2001). The honest summary of the wider literature is that peer composition matters, that its effects are real but not unlimited, and that whoever claims a cohort will transform a career independent of the individual's own effort is overstating a finding that does not support that weight. The sensible position is a modest one. The people you study with will influence your exposure, what’s normal for you, and your future network. It’s not doing your work for you."
What should you look for in an MBA cohort?
The research above converts into four questions a candidate can actually ask an admissions team, and the answers are more informative than most published statistics.
How many industries are represented? A cohort drawn overwhelmingly from one sector reproduces the network the candidate already has. The value lies in the range.
How many nationalities are represented, and how genuinely mixed is the teaching staff? International composition matters only if students are required to work across those lines rather than clustering by language and nationality in group work.
What is the range of professional experience? A cohort of uniform seniority has less to teach itself than one spanning early managers and people already running functions.
How large are the classes, and how much of the assessment is collaborative? Diversity that is never brought into contact is a statistic rather than an education.
Where can I find an international MBA cohort in Jakarta?
That’s the principle on which the Master of Business Administration at Raffles Jakarta is built. It’s a program that brings together students of different industries and cultures and makes those differences the means, not the ends. It’s collaborative learning that turns differences into something usable, not just present.” Three structural features support this approach.
The program is entirely conducted in English, which is a precondition for a truly mixed cohort rather than a nominally international cohort. Lecturers in small classes know the work of each student, not just their name. This is the condition Edmondson's research identifies as necessary for voicing difference rather than suppressing it.
In the language of this article, that network is a particularly direct way of getting weak ties between markets. The program is offered for one year with four intakes per year in January, April, July, and October. The program is offered either on campus or in a hybrid format, which combines online study with scheduled campus sessions.
One caveat for anyone considering the hybrid route is in order and follows from the research above. Contact is the basis for the formation of networks, and intermittent contact produces weaker relationships than continuous contact.
If a hybrid student views the scheduled campus sessions as merely an administrative requirement rather than a substantive part of the program, they will graduate with the qualification but without any of the benefits of being part of the cohort. The fix is unglamorous: attend the sessions, take the group work seriously and treat the people in the room as the second half of the program paying for it. The curriculum at a top MBA program is no secret. The module list is published, the textbooks are available, and a disciplined person could work through much of the technical material on his own.
A room of forty people from different industries and countries, forced to reach a joint conclusion about a problem that none of them would have defined in the same way, in a space where disagreement is safe enough to be uttered aloud. That cannot be replicated alone. That's where the difficulty lies and that's the point of the exercise. The research here is remarkably consistent. Diverse groups are better at making decisions but less happy Strong ties carry information that weak ties do not. Networks don’t diversify themselves, because homophily is the natural state. So choosing a program for the strength of its cohort is not a soft criterion that you apply after the serious ones. It is one of the tougher questions a candidate could ask about the evidence.
Arman Poureisa
Marketing Manager
References
Burt, R. S. (2004). Structural holes and valuable ideas. American Journal of Sociology, 110(2), 349 to 399. https://www.journals.uchicago.edu/doi/10.1086/421787
Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95 to S120. https://www.jstor.org/stable/2780243
Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350 to 383. https://journals.sagepub.com/doi/10.2307/2666999
Granovetter, M. S. (1973). This article discusses the strength of weak ties. American Journal of Sociology, 78(6), 1360 to 1380. https://www.jstor.org/stable/2776392
Hong, L., and Page, S. E. (2004). Groups of diverse problem solvers can outperform groups of high ability problem solvers. Proceedings of the National Academy of Sciences, 101(46), 16385 to 16389. https://www.pnas.org/doi/10.1073/pnas.0403723101
Janis, I. L. (1972). Victims of groupthink: A psychological study of foreign policy decisions and fiascoes. Houghton Mifflin. https://psycnet.apa.org/record/1975-29417-000
McPherson, M., Smith Lovin, L., and Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual Review of Sociology, 27, 415 to 444. https://www.annualreviews.org/doi/10.1146/annurev.soc.27.1.415
Phillips, K. W. (2014). How diversity makes us smarter. Scientific American. https://www.scientificamerican.com/article/how-diversity-makes-us-smarter/
Raffles Jakarta. (2026a). Master of Business Administration. https://www.raffles-indonesia.com/mba
Raffles Jakarta. (2026b). Think bigger: The Raffles Jakarta MBA 2026. https://www.raffles-indonesia.com/think-bigger-mba-jakarta
Raffles Jakarta. (2026c). Our worldwide campuses. https://www.raffles-indonesia.com/our-worldwide-campuses
Rock, D., and Grant, H. (2016). This article explains why diverse teams are smarter. Harvard Business Review. https://hbr.org/2016/11/why-diverse-teams-are-smarter
Sacerdote, B. (2001). Peer effects with random assignment: Results for Dartmouth roommates. Quarterly Journal of Economics, 116(2), 681 to 704. https://academic.oup.com/qje/article/116/2/681/1904199
World Economic Forum. (2025). The future of jobs report 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/high-ability
