Physical and non-physical violence are recognized as a major public health issue in the work environment and one commonly ignored population that may experience work-related violence is teachers/educators. However, to date, there has been no apparent investigation to examine the relations between various exposures of interest and rates of violence, based on hours worked by educators. The objective of the current study was to identify the potential risks for both physical assault (PA) and nonphysical violence (NPV) against licensed Minnesota educators, based on hours exposed.
The target study population was licensed kindergarten through grade 12 educators in Minnesota, who were identified in the Minnesota Department of Education database as of 2003. A random sample of 26,000 educators was selected from this database to determine eligibility; 6,469 eligible participants were included in the study. Data were collected between April 27, 2005 and March 31, 2006, using specially designed mailed questionnaires. PA and NPV event data were collected from respondents for a 12-month time period prior to their respective survey completion dates. Injury rates per 100,000 working hours were calculated, using generalized linear models with Poisson distribution for both PA and NPV events among educators. Multivariate analyses utilized calculated rate ratios to determine the strength of the association between exposures of interest and injury outcomes (PA and NPV); selection of confounders for multiple regression with Poisson distribution was based on Directed Acyclic Graphs, following the methods described by Greenland and colleagues.
Among all eligible respondents, the majority was female (77%). The total NPV rates were higher than the total PA rates (26.37 and 5.31, respectively). For the four subcategories of NPV, the total rates were: threat (34.82); sexual harassment (7.58); verbal abuse (55.48); and bullying (19.62).
Multivariate analyses for respective PA and NPV models revealed increased rate ratios for: educators who were not married (1.28, 1.20) versus married; worked in public alternative (1.73, 1.93), compared to public schools; those who worked in special education (4.39, 1.45) and in multiple activities (4.01, 1.41), compared with standard classroom teaching; and worked with class sizes less than ten (2.71, 1.43), compared to ten to < 25 students. Decreased risks for respective PA and NPV models were identified for: males (0.73, 0.85), compared with females; those who worked as educators for 20-29 (0.66, 0.66), and >30 (0.55, 0.60) years, compared to less than ten years; and worked in their current school for more than 20 (0.39, 0.80) versus less than five years.
Results from this study provided information about factors associated with increased and decreased risks for violence against educators, based on hours exposed. This information provides a basis for further investigation and consideration of possible intervention strategies to reduce violence in the school environment.