Bootstrap confidence intervals are interval estimates of a population parameter constructed by resampling the observed data with replacement, rather than by relying on an analytic sampling distribution or an assumed parametric form ^[1]. In the bootstrap procedure, a large number of bootstrap replicates are drawn from the original sample, each replicate having the same size as the sample and containing observations drawn uniformly at random with replacement ^[2]. The statistic of interest is recomputed on each replicate, yielding the bootstrap distribution of the statistic; this distribution serves as an empirical estimate of the sampling distribution from which standard errors, bias estimates, and confidence intervals are derived ^[1]. Several interval constructions exist: the percentile interval uses the empirical quantiles of the bootstrap distribution; the basic bootstrap interval reflects the difference between the estimate and its bootstrap quantiles; and bias-corrected and accelerated intervals adjust for bias and skewness in the bootstrap distribution ^[3]. Bootstrap confidence intervals are distinguished from analytic intervals by their reliance on computationally intensive resampling and by their weaker assumptions about the underlying distribution ^[2].