Human–AI Complementarity Index
Information SystemsReliability: composite measure1 sample item
About this scale
A composite measure of the value added when humans and AI work together, defined as joint performance minus the maximum of human-only and AI-only performance. Captures whether complementarity actually exists.
Sample item(s)
“Joint accuracy on task = X% | Human-only accuracy = Y% | AI-only accuracy = Z% | Complementarity = X − max(Y, Z)”
Reliability
Reported internal consistency: composite measure. As a rule of thumb, Cronbach’s α ≥ .70 is considered acceptable, ≥ .80 good, and ≥ .90 excellent for established measures.
Notes
Behavioural rather than self-report; useful in lab and field experiments.
Source & citation
Bansal, G., Nushi, B., Kamar, E., Lasecki, W. S., Weld, D. S., & Horvitz, E. (2019): Beyond accuracy: The role of mental models in human-AI team performance. Proceedings of HCOMP 2019, pp. 2–11.
Bansal et al. (2019): Mental models in human–AI teams. HCOMP 2019, 2–11.
Source. Scale details are summarised from the publication cited above. Always consult the original source for the full item set, scoring, and permission to use.
Use it in your research
Need to check the reliability or psychometrics of this scale in your own data? Use the free DrAIStudio research tools:
Cronbach’s alpha calculator
All statistical calculators
Search the full Scale Atlas
Academic writing humanizer