engleski

Inequalities in Achievement in Croatian Primary Schools: Their Relation to Developmental, Social and Educational Level of School Surroundings

Most countries’ educational systems strive to provide equal opportunities for different groups to participate in the educational process, regardless of their social, ethnical, or regional background. Indeed, inequalities in opportunities, access and quality of education have historically been reduced (Baker & Le Tendre, 2005), and some findings suggest that in most countries inequality in educational attainment declined in the last few decades (e.g. Thomas, Wang & Fan, 2011). Several comprehensive studies examined inequality in school attainment and its’ correlates (e.g. Thomas, Wang, & Fan, 2002 ; Ram, 1990). The other group of researchers is interested in inequality of students’ achievement, and they mostly encourage obtaining results from large scale international knowledge assessments like PISA or TIMMS. However, inequality of educational achievement on the national level was considered as inherent to educational system, attributed to characteristics of individual and thus unlikely to be eliminated. In our opinion, equality of national educational achievement should also be achieved to some extent, in particular to retain educational standards, and maintain equivalent school effectiveness level. This is especially important because marked regional differences and differences in urbanization level within countries produce significant educational inequalities (e.g. Qian & Smyth, 2005 ; Tomul, 2011). Individual educational differences are especially important at the primary education level, since it is likely that the disparities formed during earlier educational stages persist or even strengthen in further education (Coleman, 1990). Accordingly, we believe that inequality in school achievement in early educational stages can produce even bigger inequalities in school attainment during later educational path. The conceptualization of this study originated from our earlier studies in which we have examined a number of determinants of primary education in Croatia (Babarovic, Burusic, & Sakic, 2009 ; Burusic, Babarovic, Sakic, 2009 ; Burusic, Babarovic, & Seric, 2012 ; Burusic, Sakic, Babarovic, & Devic, 2013 ; Sakic, Burusic, & Babarovic, 2013). We have found marked educational gaps on the personal level, on the level of certain social groups of students as well as inequalities that occur between administrative and territorial units within Croatia. We became aware that educational achievement inequalities exist, and the goal of this study is to identify the extent of these inequalities within and between Croatian regions. Furthermore, we want to examine some correlates and possible determinants of the observed educational inequalities and relate it to other comparable findings (e.g. Mont, 2011 ; Ferreira & Gignoux, 2011). Participants were students attending fourth and eighth grades in all of 844 Croatian primary schools from 21 Croatian county (overall sample of 46 196 students). In Croatia, fourth grade students are on average aged 10, while eight grade students are on average aged 15, and at that time ending primary education. School achievement was measured by objective paper-and-pencil knowledge tests covering the content of the major subjects from the primary school curriculum. The level of school achievement inequality was measured by coefficient of variation (CV) that represents the ratio of the standard deviation to the mean. The coefficient of variation is suitable for data obtained on individual level, measured on ratio scale, and with normal distribution. Furthermore, CV enables comparison of variability across groups with different means and across different measures. As possible correlates of educational achievement variability within counties, the mean achievement test scores, and urbanization and developmental level of the counties were used. The coefficient of variation of educational achievement in different subjects is calculated on the national and county level and then compared. The indices of educational achievement inequality differ considerably across school subjects and across Croatian counties. The comparison of inequality in two age cohorts revealed no clear or systematic pattern of decrease or increase of education inequality by age. The general relation between mean test scores and measures of inequality suggest that better test scores on the county level are associated with less variance in knowledge within the county. The relation between developmental or urbanization level of the county and within-county educational variance shows the same pattern. More developed and urbanized counties have lower level of inequality in educational achievement. These results are inconsistent with expected Kuznuts Curve – inverted U-shape relation between inequality and the average educational level (e.g. De Gregorio, & Lee, 2002 ; Ram, 1990). It seems that in Croatia negative correlation better depict the relation between the average developmental and educational level and educational inequality, which is more consistent to results of Castelló and Doménech (2002) and Thomas et. al. (2001). These interesting findings and their possible impact on educational practice are discussed.

school achievement; educational inequalities; primary school

nije evidentirano