AJATEENISTUSEST ENNE TÄHTAEGA RESERVI ARVAMISE SOTSIAALSED TEGURID JA DÜNAAMIKA

Attrition Among Estonian Conscripts – Social Factors and Their Dynamics

Abstract

Conscription has been abolished in many countries; however, changes in the security environment and recruitment needs have brought the question of conscription back to the table. Compulsory military service has its pros and cons from different perspectives – societal and individual. An indicator of the effectiveness of conscription is the attrition rate during the service period. In Estonia, the attrition rate has been relatively high – approximately one fifth of conscripts have been discharged from service. The aim of the study is to explore which social and demographic factors have an impact on the risk of attrition. Based on data from the Estonian Conscript Survey, we estimate the risk of attrition among draftees during their service (8 or 11 months). The survey was carried out in 2016–2017 among conscripts who were drafted during this time period. We apply multivariate survival analysis (Cox proportional hazards regression) to estimate the effect of sociodemographic characteristics and attitudes on attrition. Conscript attrition is also related to the attitudes of significant others and the conscript’s satisfaction with their health and degree of military readiness. Based on the results of the analysis, we found that the attrition risk is higher among older conscripts. This association could be described by other factors: partnership status, poor health and economic difficulties. Namely, the risk of attrition is higher among those who were not satisfied with their health at the beginning of the service and who had difficulties in coping economically. Moreover, the partner’s (spouse, unmarried cohabiting partner) negative attitude towards conscript service was significantly associated with being discharged from service. To conclude, the results indicate that although the official reasons for being discharged from service are related to health problems, our analysis shows that there is a range of other contributing social factors.