Researchers Expose the Advantage of Prestige in Spreading Academic Ideas
View the research study here: https://www.researchgate.net/publication/328391438_Prestige_drives_epistemic_inequality_in_the_diffusion_of_scientific_ideas
What factors impact the spread of an idea across academic communities? How important is the quality of an idea compared to the prestige of the institution it is coming from? Researchers at the University of Colorado, led by Allison C. Morgan, investigate these questions by conducting a research study on the spread of ideas in the computer science community between academic institutions across the United States and Canada.
First, the researchers created a network of academic institutions based on a dataset of tenure-track faculty hirings. The nodes are the universities. There is a directed edge from university A to university B when there is student who received their Ph.D. from university A and was granted a tenure track-position at university B. When discussing cascades in class, we allowed each node to choose between “A” and “B”. Similarly, in this study, the nodes are either “S” (susceptible) or “I” (infected). All nodes begin as “S” and can transition to adopt “I”. The transition from S to I occurs when a node adopts the idea that is being tested on the network. Once a node is in state I it cannot revert to state S. The team abstractly models the quality of the idea by applying a probability that an idea will spread between two institutions in their simulation. This probability plays a similar role as the threshold that we discussed in class. So, a better quality idea will have a higher probability of spreading and a lower threshold required to spread between two institutions. The probability does not depend on the prestige of the institution itself. The prestige of each university is a data driven estimate.
To begin the simulation for an idea, one of the nodes is chosen to transition from S to I. The simulation is finished once no more nodes will transition to I–or once all nodes have transitioned to I . In the latter case, a cascade has been generated. The team repeated this simulation 10,000 times varying the prestige of the academic institution of their “starting point” and varying the quality of the idea being spread. They measure the length of each simulation to be the number of time-steps the simulation takes which quantifies the speed at which the idea spreads. They also measure the number of institutions that undergo the transition from S to I to quantify the scale of the diffusion of the idea across the network.
The results revealed that ideas that begin at a prestigious university spread farther than ideas of equivalent quality that begin at a less prestigious university. The researchers found that low quality ideas that originate at prestigious universities can still generate a large-scale cascade. An idea that comes from a prestigious institution may spread simply because it came from a highly respected institution and not necessarily because the idea itself is high quality. Additionally, the results show that ideas that originate at a less prestigious university, in the periphery of the network, must be very high quality in order to spread as far as a low quality idea that starts at a prestigious school. The team concludes, “These findings suggest that idea dissemination within academia is not meritocratic, even when the assessment of the idea’s quality (transmission probability p) is entirely objective”. Perhaps these research findings encourage people to reassess how they evaluate the quality of academic ideas and whether or not an idea’s source sways their opinion. Above all, their research provides experimental evidence to explain an unfair advantage that prestigious schools have that we intuitively might expect.
Citation:
Morgan, A. C., Economou, D. J., Way, S. F., & Clauset, A. (2018). Prestige drives epistemic inequality in the diffusion of scientific ideas. EPJ Data Science,7(1). doi:10.1140/epjds/s13688-018-0166-4