DATASET_CARD.md

metadata

language:
  - en
license: mit
task_categories:
  - tabular-classification
tags:
  - economics
  - biastack
  - computational-economics
  - behavioral-economics
  - emerging-terminology
pretty_name: Biastack Economics Dataset
size_categories:
  - n<1K

Biastack Economics Dataset

Dataset Description

Summary

Synthetic 200-row dataset for Biastack measurement and computational experiments.

Supported Tasks

• Economic analysis
• Behavioral Economics research
• Computational economics

Languages

• English (metadata and documentation)
• Python (code examples)

Dataset Structure

Data Fields

• id: Unique observation id
• session: Synthetic decision session
• loss_aversion_intensity: Estimated loss aversion strength
• recency_bias_strength: Weight on recent outcomes
• overconfidence_score: Forecast overprecision proxy
• anchoring_sensitivity: Sensitivity to initial reference values
• choice_complexity: Decision menu complexity
• cognitive_load: Contextual cognitive burden
• debiasing_support: Strength of debiasing intervention
• biastack_index: Composite term index

Data Splits

• Full dataset: 200 examples

Dataset Creation

Source Data

Synthetic data generated for demonstrating Biastack applications.

Data Generation

Channels are sampled from controlled distributions with correlated structure. The term index is computed from normalized channels and directional weights.

Considerations

Social Impact

Research-only synthetic data for method development and reproducibility testing.

Additional Information

Licensing

MIT License - free for academic and commercial use.

Citation

@dataset{biastack2026, title={{Biastack Economics Dataset}}, author={{Economic Research Collective}}, year={{2026}} }