Topics Coding Scheme
This project is developing an automated Topics Coding Scheme for use with Profiler Plus to enhance the analyses that can be conducted on leader individual differences and behavior.
The topic categories are based on those used in the U.S. Policy Agendas Project to take advantage of existing hand-coded data for testing and evaluation.
Researchers: Shao-Yun Tsai (New York State Division of Homeland Security and Emergency Services) and Michael Young
Rhetoric and Terrorist Attacks
This project uses survival analysis to assess whether the rhetoric of terrorist groups can be used to predict their next attack. We are currently examining the rhetoric of three terrorist groups in the Middle East but intend to extend the study to additional groups.
Researchers: Natasha Mather, Shilpa Hanchinal and Michael Young
Critical Language Leader Assessment (CLLA)
The Critical Language Leader Assessment project will expand the scope of leadership assessment to languages that are critical to the national security and prosperity of the United States of America and address a long-standing bias toward English language texts in international
relations. The benefits of critical language coding schemes are fourfold:
- Critical language coding schemes significantly increase the volume of source text on which leadership profiles can be constructed.
- They are likely to bring about more accurate profiles since they are based on leaders’ utterances in their native tongue.
- They help answer novel empirical questions as well as revisiting and maybe challenging established insights using a more rigorous methodology.
- They broaden the scope of leadership profiling beyond the five percent of the global population in the Anglo-American core.
Initially the Critical Language Leader Assessment project will focus on Russian, Arabic and Mandarin Chinese coding schemes for Leadership Trait Analysis and motive analysis, with other languages and coding schemes to follow.
Researchers: Michael Young (PI), Sercan Canbolat (University of Connecticut,Co-PI), Ella Honcharenko, Lu Gao, and Katherine Hallin (Edinburgh University)
Digital Presidents
The Digital Presidents project is a collaboration led by researchers at the School of Public Affairs, Xiamen University to evaluate the capability of Large Language Models to emulate the speech of national presidents using Retrieval Augumented Generation. The fidelity of the emulated speech will be assessed using a variety of measures for leader assessment including measures from Leadership Trait Analysis, motive analysis, and Operational Code Analysis.
Researchers: Wang Weiguang (Xiamen University), Wu Chang (Xiamen University), Michael Young and Lu Gao
Cognitive Mapping Coding Scheme
This project is developing an automated cognitive mapping coding scheme based on the hand coding protocols developed by Femke van Esch.
A previous attempt using machine learning failed and this project will enable direct comparison of the two approaches and perhaps demonstrate the superiority of a rule-based approach for such endeavors.
Researchers: Femke van Esch (Utrecht University), Jeroen Snellens (GreenMont Systems) and Michael Young
Russian Language Motive Coding Scheme
Using documents translated into English for the assessment of individual differences greatly limits the number and variety of documents that can be used for analysis and at the same time machine translation systems do not yet have the fidelity to translate often implicit meaning.
In this project we are developing an automated motive coding scheme for the Russian language and hope to use it to analyze individual differences among Russian leaders.
Researchers: Michael Young and Ella Honcharenko (Ukraine)
Completed Projects
The Textual Assessment of Leaders Individual Differences (TALID) Dataset
The Textual Assessment of Leaders Individual Differences (TALID) dataset is a growing dataset of individual difference scores for more than 700 global leaders.
Scores are included in the dataset for:
- Leadership Trait Analysis
- Operational Code Analysis
- Motive Analysis
- Conceptual/Integrative Complexity
- Verbal Behavior Analysis
Researcher: Michael Young
See Michael D. Young, Textual Assessment of Leaders Individual Differences: Exploring TALID, 6 April 2024
Validation of the Conceptual Complexity Coding Scheme for Profiler Plus
This project is validating the Conceptual Complexity Coding Scheme against gold standard human coding, exploring improvements to the coding schemes and assessing its performance.
Journal editors are requiring more information on validity and replication, and this is one of what may be several projects assessing the validity of automated coding schemes that run on Profiler Plus.
Researchers: Cassidy LoSasso and Michael Young
See Cassidy LoSasso, Michael D Young, Faith Renewed: Validation of the LTA_Classic Conceptual Complexity Coding Scheme, /Foreign Policy Analysis/, Volume 20, Issue 3, July 2024, orae019, https://doi.org/10.1093/fpa/orae019
