Skip to main content
Figure 5 | EPJ Data Science

Figure 5

From: Inequality and cumulative advantage in science careers: a case study of high-impact journals

Figure 5

Empirical distribution of citation impact values conditioned on publication number n . We aggregate the normalized citation z values of researcher profiles with first publication year y i , 0 j ≥1970 for Nat./PNAS/Sci. and y i , 0 j ≥1960 for the economic sciences, and with L i ≥5 and 5≤ N p ≤20. Each panel shows the probability distribution P(z(n)) conditioned on publication number n=1,…,10. The z-scores represented by each P(z(n)) represent a subset of the aggregate set of z values, independent of L i and N p . Because the unconditional distribution of z values is approximately normal with mean 0 and with units of the standard deviation ( σ z =1), we also plot a normal distribution Normal(0,1) in each panel for reference (red curve).

Back to article page