Statistics for the Dynamic Analysis of Scientometric Data: The evolution of the sciences in terms of trajectories and regimes

TL;DR

This paper introduces entropy statistics and multi-dimensional scaling to analyze the evolution of sciences through trajectories and regimes, emphasizing the importance of configurations among variables and visualizing path-dependent dynamics.

Contribution

It bridges the gap between multivariate and time-series analysis in scientometrics by applying entropy and scaling methods to visualize and test non-linear scientific evolution.

Findings

Configurations among variables are crucial for understanding scientific evolution.

Animations help visualize complex trajectories and regimes.

Entropy statistics can test path-dependent transitions.

Abstract

The gap in statistics between multi-variate and time-series analysis can be bridged by using entropy statistics and recent developments in multi-dimensional scaling. For explaining the evolution of the sciences as non-linear dynamics, the configurations among variables can be important in addition to the statistics of individual variables and trend lines. Animations enable us to combine multiple perspectives (based on configurations of variables) and to visualize path-dependencies in terms of trajectories and regimes. Path-dependent transitions and systems formation can be tested using entropy statistics.