Public school teachers are required to have specialized training to appropriately address discipline problems and foster positive school culture. Despite good intentions, many teacher-training initiatives fall short of creating an unbiased school atmosphere. This study used data collected by the National Center for Educational Statistics through the Public School Safety and Discipline 2013-14 survey to investigate the relationship between racial/ethnic tension and three categories of teacher training (management, identification, and intervention) across the dimensions of school level, urbanicity, and ethnicity. There is no recent research on public school safety and discipline that attempts to contextualize confounding variables in this way. The innovative research design hinged on articulation of a seemingly unrelated list of safety training courses into three theoretically integrated approaches, which contrast generalizable research outcomes across the United States. Each research outcome was examined across the three independent variables (school level, urbanicity, and ethnicity) through linear regression modeling. Although the complex data analysis associated with this research involved the Jackknife replication (JK1) process within the WesVar program, the use of multiple model-fit indices was built on applications of multiple software packages. Each model of investigation resulted in one or more significant factors at alpha = 0.05. For each of the four regression models in this analysis there were four parameters. Because C(p) = 4 was obtained for each regression model, the statistical analysis reconfirmed appropriate model fit. As a conclusion, the results of this investigation indicated that dimension of school level, urbanicity, and ethnicity jointly played significant roles in shaping racial/ethnic tension and all three teacher-training modalities, which urged local schools to comprehensively align discipline training with racially unbiased practices. Based on this investigation, future research recommendations were proposed on three fronts: (1) A combination of direct data gathering and secondary data analysis should occur to make findings more relevant in local settings. (2) Incorporation of qualitative and quantitative inquiries for teacher training selection is necessary to increase the understanding of results from large-scale data analyses. (3) Upgrading the statistical computing platform is needed to increase the capacity software packages like SAS and WesVar for NCES data analyses. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]