The aim of this thesis was to examine to what extent gender, support in the offline world and some aspects of the use of social networks predict loneliness among students. A total of 211 undergraduate and graduate students from the Republic of Croatia participated in the research. Participants filled out questionnaires that measured socio-demographic data, social support in the offline world (tangible support, appraisal support, self-esteem support and belonging support), use of social networks (intensity of use, number of friends and number of likes and comments) and loneliness. The results showed the same level of loneliness among male and female students. No statistically significant correlation was obtained between the variables loneliness and intensity of use. However, students who receive more social support in the offline world experience lower levels of loneliness. Furthermore, the greater the number of friends on social networks and the greater the number of likes and comments for the published content, the less pronounced the loneliness of students. The hierarchical regression model is statistically significant, and the overall model can explain 68.2% of the variance of the criterion variable. Different forms of social support (appraisal support, self-esteem support and belonging support) have the largest independent contribution in explaining the variance of loneliness. Strengthening the sense of self-worth through various programs for students can reduce the feeling of loneliness and encourage the building of quality interpersonal relationships.