[Paper Review] Measuring Team Creativity Through Longitudinal Social Signals

Peter A. Gloor, Adam Almozlino+3 more|arXiv (Cornell University)|Jul 2, 2014

Complex Network Analysis Techniques23 references14 citations

TL;DR

This paper proposes a novel method for measuring team creativity by analyzing longitudinal social signals from electronic communications and sociometric badge data. It identifies four key indicators—Rotating Leadership, Rotating Contribution, Prompt Response Time, and network centrality dynamics—that collectively predict team innovation with high reliability across diverse contexts.

ABSTRACT

Research into human dynamical systems has long sought to identify robust signals for human behavior. We have discovered a series of social network-based indicators that are reliable predictors of team creativity and collaborative innovation. We extract these signals from electronic records of interpersonal interactions, including e-mail, and face-to-face interaction measured via sociometric badges. The first of these signals is Rotating Leadership, measuring the degree to which, over time, actors in a team vary in how central they are to team's communication network's structure. The second is Rotating Contribution, which measures the degree to which, over time, actors in a team vary in the ratio of communications they distribute versus receive. The third is Prompt Response Time, which measures, over time, the responsiveness of actors to one another's communications. Finally, we demonstrate the predictive utility of these signals in a variety of contexts, showing them to be robust to various methods of evaluating innovation.

Motivation & Objective

• To identify reliable, data-driven indicators of team creativity that go beyond individual traits or static network structures.
• To address the challenge of measuring collaborative innovation in real-world teams where creativity emerges from dynamic interactions.
• To develop a framework that leverages longitudinal communication patterns to predict team-level innovation outcomes.
• To validate the predictive power of social network signals across multiple contexts and evaluation methods.

Proposed method

• Extracts longitudinal social signals from electronic communication records (e.g., email) and face-to-face interactions via sociometric badges.
• Defines Rotating Leadership as the temporal variation in actors' centrality within the team's communication network.
• Measures Rotating Contribution as the fluctuation in the ratio of outgoing to incoming communications per actor over time.
• Quantifies Prompt Response Time as the average latency in responses between team members, capturing responsiveness dynamics.
• Uses network analysis and time-series metrics to compute the four core indicators from raw interaction data.
• Validates predictive performance using multiple innovation evaluation methods across diverse team settings.

Experimental results

Research questions

• RQ1 Can longitudinal social signals derived from team communication patterns reliably predict team creativity?
• RQ2 Which specific dynamic social network indicators are most predictive of collaborative innovation?
• RQ3 How do Rotating Leadership, Rotating Contribution, and Prompt Response Time individually and collectively contribute to predicting team innovation?
• RQ4 Are these signals robust across different methods of evaluating innovation and varied team environments?

Key findings

• The four social network-based indicators—Rotating Leadership, Rotating Contribution, Prompt Response Time, and network centrality dynamics—collectively predict team creativity with high reliability.
• Rotating Leadership, reflecting shifting influence in communication networks, is a strong predictor of team innovation outcomes.
• Rotating Contribution, indicating dynamic participation patterns, correlates significantly with higher team creativity.
• Prompt Response Time demonstrates predictive power, suggesting that responsiveness enhances collaborative innovation.
• The indicators remain robust across different evaluation methods for innovation, indicating generalizability across contexts.
• The model successfully predicts team-level innovation using only interaction data, without relying on outcome labels or subjective assessments.