Abstract:

Introduction:
Undertaking studies in the field of public health is not always an educational success. There are numerous potential factors that may have influence on the risk of failure during studies. To minimize the ratio of resignations / cross-outs from the list of students, appropriately selected criteria concerning accuracy should be applied during admissions. From the point of view of selecting the best candidates, criteria considered more beneficial are such that are characterised by a greater specificity allowing to avoid a situation in which there would be people with an inadequate level of preliminary competences among the candidates.

Aim of study:
Assessment of selected socio-demographic factors and admission criteria as potential predictors of educational success of MA students at the Public Health Department at the Medical University of Warsaw (MUW).

Materials and Methods:
151 students of the Public Health Department who undertook studies of the 2nd degree at the Faculty of Health Sciences at MUW in 2007 qualified for the study. The age average was 22.8 ± 2.03; about 77% of the studied group was women; 87% of the students completed their 1st degree at MUW. The failure coefficient was 32.4% and mostly concerned students of the 1st year.
Predictive analysis was performed using the method of logistic regression. In the suggested logistic model, five predictors were used that belonged to the group of socio-demographic factors: gender, age, place of completing the 1st degree, professional title, and mode of study. Moreover, the model also included the score obtained by the candidate at the entry exam test in the following five areas: Epidemiology, Organization in Healthcare, Methodology and Statistics, Health and Human nutrition, and Social science. Whereas the dichotomous variable, dependent on the analysed model, was the fact of completing / not completing studies (a variable coded as 0 and 1, respectively). Estimation method of quasi-Newton was applied here with the determination of asymptotic standard errors.
.
Results:
The suggested model of logistic regression which would assess the influence of potential predictors on the risk of failing to complete studies showed no statistical relevance (chi2 = 9.515, P = 0.484). No variable in the group of socio-demographic factors, nor admission criteria proved relevant predictors of failure during studying.

Conclusions:
Admission criteria applied during the recruitment process are characterised by insufficient validity in the evaluation of candidates’ competences on entry as they do not contribute to the efficient selection of those candidates for whom the risk of failure in studying is high. It is necessary to increase the validity of the selection process by introducing additional admission criteria. Further monitoring of the quality of the accepted admission policy at the Public Health Department at the Medical University of Warsaw is necessary.

Keywords:

Educational status, school admission criteria, student dropouts.