MODELS
← Browse the encyclopedia

Encyclopedia · Free preview

Proxy

A measure used to represent a distinct attribute or outcome through an assumed or established connection. Its usefulness depends on that connection and the intended inference.

A proxy is a measure used to stand in for a different attribute or outcome. The connection must be specified: course completion might indicate engagement, but it does not directly establish understanding. A useful proxy can be imperfect and still inform a decision when its limits and validation are understood.

Ask what would make the proxy change without a corresponding change in the target, and what target changes it might miss. Incentives, new populations, or altered conditions can weaken the connection. Use direct checks where feasible and multiple complementary indicators where needed, without assuming that adding metrics automatically produces validity.

When to use it

When choosing metrics, interpreting scores, or investigating a divergence between reported improvement and the outcome that matters.

How it can help

Encourages checking whether changes in an indicator correspond to changes in the target, including under incentives or changing conditions.

Keep exploring

Read the full page.

Create your free access to continue reading and explore the complete library.

Register free with ChatGPT →

Already registered? Use the same button to sign in.

Sign-in shares your email with Michael Simmons to create your site access. No payment required. Newsletter signup is separate. How your data is used