Introduction and objective Pursuant to the Act on Police Activities and Powers (Official Gazette NN 76/09, 92/14) police activities involve the protection of life, rights, freedom, safety and inviolability of a person; the protection of public peace and order, as well as the protection of property; prevention of felonies and misdemeanours; and other activities. Due to the specifics of their work, police officers are frequently exposed to elevated physical and psychological pressure during their working hours. Due to the nature of their job, the risk of developing various illnesses, such as the circadian rhythm disorders, inadequate sleeping, psychological disorders, metabolic disorders, cardiovascular illnesses, is higher in police officers than in the general population. As there are strong evidence about the influence of physical activity on the aforementioned disorders, the promotion of physical activity in police officers represents one of the fundamental measures implemented for the purpose of improving health and work efficiency. Efficient interventions for the promotion of physical activity should be based on scientific evidence, i. e. on factors that have an effect on undertaking regular physical activities, such as socio-demographic and psychological factors, lifestyle, social support. Therfore, the purpose of this doctoral thesis is to research into the correlates of physical activity in members of various police organizational units of the Ministry of Internal Affairs of the Republic of Croatia. Methods The research was carried out on a representative sample of police officers of the Ministry of Internal Affairs of the Republic of Croatia. The total sample involved 1,820 subjects of male gender, 33.5 % belonging to the uniformed police, 33.1 % belonging to the traffic police and 33.4 % belonging to the criminal police. The sample was stratified according to the categories of police departments (1-4), followed by randomly selecting 12 out of 20 police departments. The data was collected by means of a questionnaire. The questionnaire involved nine separate units: Global Physical Activity Questionnaire – GPAQ, Questionnaire on general data, i.e. socio-demographic factors, The Barriers to Being Active Quiz 5 – BBAQ, Self-Efficacy Surveys for Exercise Behaviours, Physical Activity Enjoyment Scale – PACES, Social support and exercise survey, Simple Lifestyle Indicator Questionnaire - SLIQ), Exercise Motivations Inventory-2 - EMI-2, Physical Activity Stages of Change Questionnaire. Adequate reliability confirmed by the test-retest method was determined for all instruments. The following methods were used for data analysis: regression analysis, Kruskall-Wallis ANOVA, Hi – square test. Results According to the first partial research objective, a statistically significant relation between the set of socio-demographic variables and the level of physical activity of members of uniformed, traffic and criminal police was determined. Socio-demographic characteristics, age, length of service, number of children, number of household members, degree, monthly income, monthly household income, size of the place / town of residence and the number of night shifts per week represent correlates that share 1.9 to 6.3 % of common variance with physical activity in the domains of physical activity. The highest common variance shared with physical activity is the one in the domain of work, 4.4 – 6.3 %. According to the second partial research objective, a statistically significant relation between the physical activity barriers and physical activity in different domains was determined among members of uniformed, traffic and criminal police. The physical activity barriers – lack of time, social impact, lack of energy, lack of will, fear of injuries, lack of skills and lack of resources, share 1,3 % to 16.2 % of common variance with physical activity in different domains. The highest common variance is shared with physical activity in the domain of leisure time and recreation, 11.7 – 15.4 %. Lack of will as a physical activity barrier dominates in members of all police groups. A statistically significant relation of self-efficacy and workout habits with physical activity in all domains was determined among members of various police groups. The psychological factors self-efficacy and workout habits, readiness for consistent adherence to the workout programme and persistence in finding time for working out (Sallis et al., 1988) share 1,1 % to 10.6 % of common variance with physical activity in different domains among members of uniformed, traffic and criminal police. The highest common variance shared with physical activity is the one in the domain of leisure time and recreation, 5.9 – 10.6 %. A statistically significant relation between the enjoyment in physical activity and the overall level of physical activity was determined in all police groups. The strongest relation between the variable related to enjoying physical activity and the level of physical activity is in the domain of leisure time and recreation in all police groups. It is the strongest among the members of uniformed police and they share 10.9 % variance. According to the third partial research objective, a statistically significant relation between the set of variables lifestyle variables and physical activity in different domains was determined among members of uniformed, traffic and criminal police. The lifestyle variables – nutrition, alcohol intake, smoking and stress, share 1.0 % to 21.5 % of common variance with physical activity. The highest common variance is shared with physical activity in the domain of leisure time and recreation, 6.2 – 21.5 %, while the lowest common variance is shared with physical activity in the domain of transport 1.0 – 6.3 %. A significant negative relation of smoking and stress with physical activity was determined in all domains. According to the fourth partial research objective, a statistically significant relation between the set of social support variables and the physical activity in different domains was determined among members of uniformed, traffic and criminal police. The social support variables share 2.7 % to 17.9 % of common variance with physical activity. The highest common variance is shared with physical activity in the domain of leisure time and recreation, 10.0 to 17.9 %. According to the fifth partial research objective, statistically significant differences were determinend in motives for engaging in physical activity were determined among members of uniformed, traffic and criminal police. The first three motives for engaging in physical activity are health, refreshment and strength, followed by mobility, illness, stress, weight control, enjoyment and challenge. The results show the highest level of motivation in members of criminal and uniformed police, while members of traffic police show the lowest motivation for engaging in physical activity. According to the sixth partial research objective, statistically significant differences were determinend in the level of readiness for physical activity among members of uniformed, traffic and criminal police. The differences were determined in three out of five stages or phases of readiness for engaging in physical activity of police officers. In total, the lowest level of readiness for physical activity is among members of traffic police, i.e. the greatest share of aforementioned subjects belong to the phases: lack of intention (precontemplation) and forming intention (contemplation). Conclusion This research has identified a significant relation of socio-demographic and psychological factors, lifestyle and social support with physical activity of members of various police groups of the Ministry of Internal Affairs of the Republic of Croatia. Furthermore, significant differences were determined in a large number of analysed variables among members of uniformed, traffic and criminal police. The results point to the need for various approaches to the promotion of physical activity among members of different police groups. The scientific contribution of this research is primarily evident through the contribution to understanding physical activity correlates among police officers, as this is, to the best of author’s knowledge, the first research conducted on a representative sample of police officers that determines comprehensive physical activity correlates in various domains among members of uniformed, traffic and criminal police.