In addition, the number of characteristics included in the accident statistics is limited, so that in particular the influence of latent variables, such as personality traits or psychological dispositions, cannot be investigated. Finally, the cross-sectional character of previous studies limits the possibility of deciphering causal structures in relation to the occurrence of accidents and injuries. The GUS-study aimed at uncovering these causal patterns of school injuries by conducting a panel study.
By conducting a nationwide (with the exception of Bavaria and Hamburg) representative panel survey of about 10,000 schoolchildren over six survey waves, predictors of injuries in the school environment could be identified on the basis of individual injury biographies. In order to confine the subject of the study, the focus was on injuries that occured in the school context. A central distinction was made between injuries in school sports, on the schoolyard, in the school building and on the way to and from school. The analysis strategy employed the use of bi- and multivariate analysis methods that are adequate for panel data as well as for the specific data structure with different levels of analysis.
Level of Education: School
Lead Organisation: DGUV
