Where \(g\) is the observed proportion, \(e\) is the expected proportion, \(n\) is the effective sample size, and \(d_{eff}\) is Extra deff (see Weights, Effective Sample Size and Design Effects):
\[\begin{align} z = \frac{g - e}{\sqrt{d_{eff}e(1-e)/n}} \end{align}\]
where \(p \approx 2(1-\Phi(|z|))\) if \(|g - e| \ge 0.0001\) and NaN otherwise.