Urban development-induced cadastral boundary digitization and informal plot dispute adjudication pathways in Ghanaian municipal courts: an instrumental-variable analysis of survey-coverage exposure

The central problem is that such analyses treat survey-coverage exposure as an exogenous instrument, yet the finite-sample behaviour of the IV estimator under realistic data-generating conditions remains uncharacterised. Six hundred data sets of N = 500 were drawn from a declared data-generating process (intercept 12.0, exposure 0.42, group -3.1, severity 1.35, error SD 4.2). The analysis found that estimated bias for the intercept was 0.02931 and for exposure -2e-05; root-mean-square error was 0.89992 for the intercept and 0.01617 for exposure; empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. The results indicate that under the stated process the estimator recovers the exposure coefficient with negligible bias and near-nominal coverage, while intercept precision is more sensitive to sampling variation. IV designs in cadastral-dispute research must report finite-sample diagnostics of this kind before causal claims about digitization and adjudication pathways are advanced.