engleski

Regression Analysis of the Effects of Socioeconomic and Demographic Characteristics on Students’ Academic Achievement

During the history of education the researchers have tried to determine which variables have an impact on the academic achievement of students, and how to make the education system more efficient. It is known that one of the most influential variables on academic achievement is IQ. Many researches focused on the impact of IQ on academic achievement, discovered that about 20% of the variance of success in the exam it is explained with the IQ, while the remaining 80% of the variance of success in the exam is explained by other variables. This paper examines the influence of selected socioeconomic and demographic variables on the success of students at the Department of Professional Studies, University of Split, Zagreb Teaching Centre, on the colloquium of business statistics in the 2nd semester, the academic year 2009/10. Independent variables in the model are: the percentage of students’ presence at lectures and exercises, the amount of time spent in studying for the exam, gender, age, employment status, personal conditions for learning and parenting status. The dependent variable is the number of points achieved at the colloquium. The research included the sample of 230 students, and data were collected from the survey questionnaire and official records from a database of students at the Faculty. Using statistical programs SPSS20 and EViews3, the multiple linear regression model was developed with the number of points at the colloquium as the dependent variable, and with other variables being independent. The validity of the model was tested by regression diagnostics procedures, and no problems were found. The research results confirmed some assumptions. The regular attendance of students at both lectures and exercises has a positive influence on the number of points at the colloquium. Positive impact on the success of the colloquium was confirmed for the variable the time spent studying, too. Also, students who have rated their own learning conditions by grades good (3) and very good (4), on a scale from 1 to 5, had significantly better results than other students. The other variables such as: age, gender, employment and parenting did not shown to be statistically significant.

education, higher education, liner regression model, regression diagnostics model validation

nije evidentirano