Learning Analytics Dashboard

Problem

In a corpus of 34,342 learner samples across 95 L1 backgrounds (5.2M+ language units), replacement errors accounted for 49.5% of all tracked error types. For key structures, error rates reached 20.13% for certain groups, with specific error pairs representing up to 24.88% of occurrences. Patterns persist across experience levels.

Communication breakdown is systemic, not anecdotal.

Organizations lack pattern-level intelligence to prioritize intervention.

Solution

The Learning Analytics Dashboard translates large-scale linguistic data into executive-ready decision support. It surfaces recurring ambiguity clusters, persistence trends, and development priorities aligned to workforce strategy.

This shifts communication training from generic programming to targeted capability investment.

Design Decisions

• Pattern abstraction (no academic terminology): Executives need risk categories, not grammar labels.
  
  
  

• Executive overview layer: Immediate visibility into dataset scale, high-risk distributions, and persistence metrics.
  
  
  

• Ranked breakdown views: Enables prioritization and targeted development design.
  
  
  

• Experience vs. risk comparison: Challenges assumption that tenure alone reduces communication complexity.
  
  
  

• Strategic recommendation layer: Direct mapping from insight to learning design, AI tooling, and measurement strategy.
  
  
  

Reframes communication risk as measurable organizational exposure.