UA Ledger · Measurement Lab · draft for review
Synthetic teaching inputs. No real campaign results or industry defaults. Change the values to inspect the model. Calculations run locally; this page sends no data and does not retain inputs after closing.
Use the randomised assignment population rather than only people who completed a downstream event. Derive each expected count from the total assignments and planned split, then sum squared observed-minus-expected differences divided by expected counts. The worksheet reports the statistic without a p-value or automatic pass threshold. Predefine the statistical review and investigate the actual assignment mechanism.
Only a varies; all other inputs are held constant. This is not a probability range.
(a+b)*share(a-(a+b)*share)**2/((a+b)*share)+(b-(a+b)*(1-share))**2/((a+b)*(1-share))Additional constraints: (share<1). The field minimums also apply.
Very small expected cells, clustered assignment and repeated checks need specialist treatment. A balanced allocation also cannot establish correct event logging, absence of spillover or a causal effect on revenue.
Chi-Square Goodness-of-Fit Test · primary documentation checked 19 September 2026.