IER

Python library for detecting Insufficient Effort Responding (IER) / careless responding in survey data.

Install

pip install insufficient-effort

Optional extras:

pip install "insufficient-effort[plot]"

The NumPy-only base install includes chi-square flagging, Q-Q quantiles, IRT theta estimation, and response-time mixture scoring. The legacy full extra is retained as an empty compatibility alias.

Quick start

import numpy as np
from ier import IndexOptions, composite, screen

data = np.array([
    [1, 2, 3, 4, 5, 4],
    [3, 3, 3, 3, 3, 3],
    [1, 5, 1, 5, 1, 5],
], dtype=float)

result = screen(data, options=IndexOptions(scale_min=1, scale_max=5))
print(result["flag_counts"])

scores = composite(data, indices=["irv", "longstring", "person_total"])
print(scores)

Learn more

• Getting started
• Index catalog
• Screening workflow
• Composite guidance
• Threshold guidance
• R package notes
• API reference
• Changelog

Citation

@software{ier2026,
  title={IER: Python package for detecting Insufficient Effort Responding},
  author={Lyons, Cameron},
  year={2026},
  url={https://github.com/Cameron-Lyons/ier}
}