Between Subjects Variance and Within Subjects Variance in ICC

Sorry this may be a naïve question. Generally ICC is considered an indicator of reproducibility. Higher ICC is interpreted as good reproducibility , lower ICC not so much.

ICC as such is a ratio of between subjects variances/(between + within subjects variance). So if higher ICC is preferred, that means the numerator, higher between subjects variances is a good signal.

In chemicals , especially those related to phthalates and parabens, how can higher between subjects variance indicate a high reproducibility. Higher variance in phenols and chemical concentration values between subjects should indicate poor reproducibility , is it not ? Please advise.

Hello Sudhi, indeed this is a fundamental concern with methods like ICC – a measurement in a heterogeneous population appears more reliable than the exact same measurement applied to a homogeneous population. I have essentially stopped using ICCs for this reason, and followed prof Harrell’s advice to rely on U-statistics. Related post: