摘要：The analysis of the school dropout phenomenon at the level of disadvantaged population raises a series of issues in officialstatistics in Romania because there are no statistics on the school dropout on population grouped by nationality. More, for Romapopulation there is the practice of early studies abandon by some youngsters. For this reason, this study will be highlighted on thedrop-out characteristics of the Roma population using a series of statistical methods. The main objective of this study is todetermine to what extent school dropout rate is different for the Roma population compared with total population from fourdevelopment regions from northern Romania. For the analysis were used the following methods: regression method andstatistical data grouping method. To estimate the model parameters were used data series on school dropout rate by type ofeducation at counties level for the period 2000-2009 and recorded data from the 2002 Population Census. The obtained resultsallow us to assert that the school dropout rate is higher in most of the counties that have a higher proportion of Roma populationin total population. In general, the Roma population is more vulnerable to school dropout than other populations.
