This paper empirically and systematically assessed the performance of bootstrap resampling procedure as it was applied to a regression model. Parameter estimates from Monte Carlo experiments (repeated sampling from population) and bootstrap experiments (repeated resampling from one original bootstrap sample) were generated and compared. Sample sizes of 20, 30, 50, and 100 were considered in the simulation. Ten independent Monte Carlo experiments and 10 independent bootstrap experiments were conducted respectively for each sample size condition, with 1,000 samples (resamples for bootstrap) for each experiment. Estimates for standardized regression coefficients were obtained from each sample, and the mean estimates across samples were evaluated in relation to the population parameters. The results indicate that, as the number of resamples increases, the mean bootstrapped estimates did not show a clear tendency to converge on the population parameters. But, with the increase of the original bootstrap sample size, the quality of the bootstrapped estimates improved. For the case of regression analysis, the results raise some concern about the validity of the assumption underlying the bootstrap procedure. (Contains 27 references and 9 figures.) (Author)