ECONOMIC ACTIVITIES OF THE ROMA IN SERBIA
1. Differences between two censuses
Table 1 shows the data on the economic activities of the Roma in the Republic of Serbia, gathered at the censuses held in 2002 and 2011.Based on the differences between participation in certain categories it can be seen that within the 10-year period, the percentage of employed Roma decreased significantly, by 5.95%, whereas there is an increase in the category of persons doing housework (housewives), by 2.19%, and an increase in the number of persons from the category ‘the others’, by 3.57%.These are statistically very significant differences, since the Chi Square test value (which is 5.233,6) is much higher than the theoretical ones (12.6 for the level of freedom 6 and the level of significance of 5%, and 16.8 for the same level of freedom and 1% of the level of significance).
Compared to the economic activity of the Roma in the whole territory of the Republic of Serbia, in the region of Vojvodina there is a somewhat smaller decrease in the number of the employed Roma (but it is still a significant one), an increased number of housewives and persons from the category ‘the others’ (Table 2). Differences observed between the two censuses are in this case also highly statistically significant, although the empirical Chi Square test shows a somewhat smaller value than the values for the whole territory of Serbia.It amounts to 1.337,2, whereas the highest differences between empirical and theoretical frequencies can be seen in the categories of the employed Roma, housewives and the others. However, its value leads to a conclusion that the null hypothesis that the order of frequencies according to categories of population from 2011 is in line with the order of structures from 2002 should be rejected.
Unlike in the region of Vojvodina, in the central Serbia area (the region of Belgrade, Sumadija and the west of Serbia, just like the east and south of Serbia) there is the largest decrease in the number of employed Roma as compared to the census from 2002, but also a smaller increase in the number of women from the category of housewives (Table 3). This increase is only by 1.43%, whereas in the region of Vojvodina the number of women in this category increased by 4.09%.It is also noticeable that in Vojvodina, the percentage of unemployed persons decreased by 1.54% and in 2011 they were probably in the category of ‘the others’ that included the persons with other personal incomes, and those who received social welfare assistance.In Central Serbia, the number of the Roma in this category increased by 1.91% as compared to the census from 2002.The differences between the two censuses, shown in Table 3, are also statistically very significant and the value of empirical Chi Square test is much higher as compared to Vojvodina region, which indicates larger differences between the two censuses in Central Serbia.It amounts to 4.423 and it also leads to a conclusion that the null hypothesis on the equality of structures in the two censuses should be rejected.