A Conversation between Theory, Methods, and Data.

Area of Science:

• Psychometrics and Quantitative Psychology
• Human Systems Dynamics
• Information Theory

Background:

• The 10-year anniversary of the COGITO Study prompts a re-examination of the Cattell data box (persons × variables × time).
• Different research questions necessitate distinct statistical approaches across the data box dimensions.
• Integrating theories and methods across the three dimensions of a large-scale data cube (e.g., COGITO) is complex.

Purpose of the Study:

• To discuss the three dimensions of the Cattell data box in relation to distinct researcher questions.
• To illustrate statistical challenges using the speed-accuracy tradeoff.
• To present a conceptual model for understanding effects across time, including dynamics and learning.

Main Methods:

• Utilizing the COGITO data cube (200 persons × 100 variables × 100 occasions).
• Developing a conceptual model for the speed-accuracy tradeoff to account for cross-sectional, short-term, and long-term effects.
• Analyzing nonstationarity in human systems, contrasting dancing and conversation.
• Applying information theory to understand the theory-methods-data interaction.

Main Results:

• Identified fundamental differences in ordering and time scaling between the time axis and other dimensions.
• Highlighted nonstationarity as a pervasive issue in the time dimension of human systems.
• Proposed an information-theoretic perspective viewing the theory-methods-data interaction as a 'conversation' rather than a 'dance'.

Conclusions:

• A conceptual model can integrate cross-sectional, dynamic, and learning effects within the Cattell data box framework.
• The 'conversation' metaphor, measured by entropy changes, offers a metric for scientific progress in understanding theory-method-data interactions.
• Addressing time-related complexities like ordering, scaling, and nonstationarity is crucial for advancing research in human systems.