This paper introduces novel approaches for analyzing temporal sequences in ordinal variables. The research explores three complementary methods—frequency tables, transitional graphs, and dependency tables—to investigate causal precedences between political variables.
Frequency Tables: A detailed analysis of value pairs across two variables provides initial insights into their relationships.
Transitional Graphs: These visual representations clearly show which variable typically reaches certain thresholds first in the sequence.
Dependency Tables: Offer a direct test for evaluating if specific values in one variable are prerequisites for changes in another.
The authors demonstrate these methods using data from the V-Dem dataset. Their findings indicate that changes toward electoral democracy consistently follow improvements in freedom of expression and access to alternative information, suggesting these factors as precursors to democratic development.
| Investigating Sequences in Ordinal Data: A New Approach with Adapted Evolutionary Models was authored by Patrik Lindenfors, Fredrik Jansson, Yi-ting Wang and Staffan I. Lindberg. It was published by Cambridge in PSR&M in 2018. |
