Scholars can implement bootstrapping techniques to increase the accuracy of estimates and overcome the limitations associated with small sample sizes. This chapter provides a theoretical and mathematical overview of bootstrapping, as well as a brief discussion of the applicability of bootstrapping to criminal justice and criminology. Bootstrapping can be used to estimate regression coefficients, standard errors, confidence intervals, and p values, to name a few. Bootstrapping during the estimation of a regression model can be easily achieved or may generate problems depending upon the configuration of the data. Consistent with other Bayesian techniques, Bayesian bootstrapping permits the integration of preexisting information into the selection of the bootstrapped subsamples. Cluster bootstrapping represents a bootstrapping technique used to account for the clustering of data in the creation of the bootstrapping distribution. In general, SPSS, Stata, and R offer case resampling and regression-based bootstrapping options when estimating various coefficients.
|Original language||English (US)|
|Title of host publication||The Encyclopedia of Research Methods in Criminology and Criminal Justice|
|Subtitle of host publication||Volume II: Parts 5-8|
|Number of pages||5|
|State||Published - Jan 1 2021|
All Science Journal Classification (ASJC) codes
- Social Sciences(all)