Abstract This paper presents the results of a pilot study investigating the language proficiency of Romani children before they start school and during their first three years of primary education. The children were tested in their first language (Romani, L1) and the language of instruction (Slovene, L2). The testing included 29 children from two Slovene regions: fourteen were from Prekmurje (north-east Slovenia) and fifteen were from Bela krajina (southeast Slovenia). An oral language instrument adapted for Romani was used to assess comprehension, word recall, syntax and text comprehension. Composite indices were created for L1 and L2 knowledge. Simple regression analysis revealed a weak to moderate positive correlation between L1 and L2 proficiency, though this was not statistically significant. In the moderated regression model, this positive correlation was statistically significant after adjusting for region, with no evidence of a difference in magnitude between the two regions. At an average level of L1 proficiency, average L2 achievement is higher in Prekmurje. These results suggest that systematic development of L1 can support L2 acquisition. Further research is needed on a larger sample, as well as on the effect of testing measures, and additional linguistic training is required for employees who teach Romani children in Slovene schools.
Zorčič et al. (Tue,) studied this question.