Back to projects
Project

LLM-Generated Self-Efficacy Scale Validation

Validation work examining whether LLM-assisted item generation can support defensible self-efficacy measurement.

2026 In Analysis
Measurement Validation

Overview

This project examines whether large language models can help generate or refine self-efficacy scale items without sacrificing measurement quality.

The aim is not novelty for its own sake. The goal is to test whether AI-assisted scale development can be rigorous enough to support real research and applied measurement work.

What I Am Looking At

  • Item quality, coherence, and construct coverage.
  • How generated items compare with more traditional development workflows.
  • What kinds of human review and validation remain necessary before use.