Determinants Of Schooling Among Primary School Age Children In Ethiopia

Investing on schooling is the backbone of economic growth and economic development, and that primary school lays the foundation for a more productive labor force through promoting literacy and numeracy and judging economic decision. Primary education serves as the basis of any education and training. Primary schooling has direct and positive impacts on earnings, farmer productivity as well as future health and poverty alleviation benefits (World Bank Policy Paper on Primary Education, 1999).

Primary education also has great role to play in a nation’s economic wellbeing, providing on average the highest public returns to investment for the state and a critical driver of economic growth. However, Ethiopia still has one of the lowest primary enrolment and highest illiteracy rates in the world. The theory of human capital recognizes participation in education as an investment decision made now in exchange for return later in life (Mince, 1958). In case of children in developing countries, such as investment decision are generally made by parents through implicit cost-benefit assessments. Parents are assumed to compare the future benefit of sending their children to school against the immediate cost. These benefits are for the children as well as for the parents themselves, especially in the context of developing countries where children are considered as old-age security for parents (Husiman and Smit2009).

Factors that are related with children schooling among primary school age children (7-17) are complex ranging from the individual level to the regional level and the country level factors. A number of studies have been conducted on determinants of schooling in both developed and developing countries. A study in Norway on determinants of school enrollment, for instance, used OLS regression to show that individual level and household level factors are important determinants of school enrollment (Aakvik.et al, 2005). However, this study is limited in terms of its methodology, where it employed traditional OLS regression to analyze a data that has violates the independence assumptions of regression and gives an incomplete picture to understand the true association of educational outcomes and its determinants.

A study in Pakistan using the Lahore as the case study on determinants of enrollment in primary education using OLS regression and log it model show that wealth index, age of child, educational level of household, works status of household are major determinants of enrollment in primary education (Mazharul and Saima, 2013). However, in this study not distinguish between and within explanatory variables with outcomes variables.

A study in rural Ethiopia using the Tigiray region as the case study, on determinants of children’s school using both prompt model and Hickman’s sample selection model survey data gathered from four villages. The result show that age of child, birth order of child, household head, literacy of adult male family and household income are most important determinants of children’s school ( Abafita and Kim,2015). However, they are used data limited in scope and coverage as well as single analysis which may be not adequate indicator of national level educational outcomes.

A study rural Ethiopia in on determinants of current enrollment status and relative grade attainment among primary school children using instrumental variables estimation method data from the Ethiopian Rural Household Survey (ERHS) between 1989 and 2004.Thus, show that household level factors are major determinants of primary school (Mani et al.’s ,2013). However, this study was focused to measure educational outcomes single indicator. Educational outcomes influenced various factors, so single indicators not representative of children’s educational outcomes.

Very few studies make distinction between and within different levels of analysis in the framework of a multi-level analysis in the previous studies. For instance, A study in India on determinants of primary level educational attainment using multilevel analysis data from cross sectional survey on social consumption show that demand side factors are main determinants of primary level education ( Chandan, Kumar, and Nauriyal,2013). However this study was used limited scope and coverage of data.

Most of the above revived studies in used single level analyses as their major data analyses methodology. Using single level analyses, did not distinguish between and within variation explanatory variables that are associated with educational outcomes. Thus, to avoid these problems, this study was used multilevel logistic analysis, using factors at the individual level, household level and regional level. Besides, the most of the above studies did not consider various measures of children’s educational outcomes. Using single measure of educational outcomes not clearly indicated educational outcomes. Due to the fact that educational outcomes are influenced different set of factors. Thus, to avoid these problems this study was used various measures of educational outcomes (current enrollment, every attained school and age for grade equivalent ).

In general, most above revived studies reported based on either OLS analyses or less sophisticated models that didn’t address variation in children’s educational outcomes and they have put in considerable effort in a bid to understand pertinent issues concerning primary education. Beside, these most of studies focused single measure of educational outcomes in Ethiopia except Abafita and Kim (2015). None specifically analyzed between and within regions and its determinants among primary school age children’s in Ethiopian the framework of a multi-level analysis. This study contributes to the literature by addressing this methodological gap in knowledge using the recent2016 EDHS dataset by framework of multilevel analysis and take into account various measuring child educational outcomes gaps.