Depression has lately become one of the most prevalent mental disorders; the number of people affected by it increases daily. A lot of studies analyze the connection between depression and stress commonly end up at the same conclusions: depression is predominantly found among younger people, especially among students which tend to have high levels of stres. Research focuses on detecting two types of variables: those that cause depression and those that protect from depression. Some typical examples of variables used to predict depression are gender, emotional intelligence, and perfectionism. These variables are often described as good predictors, even though they have not been consistently affirmed as relevant on a global level. In this study the focus was placed upon affirming the relevance of the aforementioned variables. To prove their relevance a sample of 410 students was collected, consisting of 306 female and 104 male students. The research participants completed the Scale of Depression, Anxiety, and Stress, Emotional Competence Questionnaire (UEK-45), and the Multidimensional Perfectionism Scale. They also gave out information about their gender and age. We used the data to substantiate a few different claims. The first one was that there is a noticeable gender inequality when analyzing depression, emotional intelligence and perfectionism. The second claim was that there is a relevant connection between emotional intelligence and depression. The third one was that there is a relevant connection between perfectionism and depression. After proving these claims, we focused on examining how perfectionism and emotional intelligence can be used to predict depression. We managed to prove that there is a connection between gender and emotional intelligence, but no connection between gender and the other two variables. We also found a negative correlation between depression and emotional intelligence (perception and understanding, expression of emotions and emotion management) and a positive correlation between depression and perfectionism and maladaptive perfectionism. And last, we used perfectionism and emotional intelligence as predictors, and managed to develop a model that explains 37% of the total variance. Emotion management and adaptive perfectionism proved to be positively correlated with our predicted value, while maladaptive perfectionism proved to be positively correlated with our predicted value.