Search

In Bangladesh gross and net enrolment rates are used to measure overage and underage enrolment in the education system. However, due to the limits of these methods in exploring the issue of age in grade, the paper uses the CREATE Community and School Survey (COMSS) data from Bangladesh. COMSS was a longitudinal survey of 6,696 households with 9,045 children of 4-15 years old in 2007 and 2009. The paper shows that 69.4 percent of 6-15 year old children, enrolled in primary and secondary schools in 2007 are age in grade incongruent and that in the early grades of primary school this proportion is even higher. Age in grade incongruent children come from relatively low income families and have relatively poor health. Age in grade incongruent children attend school irregularly and perform worse than the congruent children of the same grade. Less than 50 percent of primary Grade 1, 2 and 3 children progress to the next grade on time. Slow progressing children also come from relatively low income families, have relatively poor health and inadequate learning materials compared to regularly progressing children of their grade group. The slow progressing children attend school irregularly compared to the regularly progressing group and get less primary school stipend money. Both these age in grade incongruent and slow progressing children are increasing the number of "silently excluded" children in classrooms. (Contains 5 tables and 17 figures.)

The main objective of the study is to gain an understanding on educational expenditure at primary and secondary levels in Bangladesh. In estimating educational expenditure by source, it has been sought to determine: (1) sources of financing of primary and secondary education; (2) rural-urban variation; (3) variation between boys and girls; (4) variation between types of educational institutions; and (5) variation among students with respect to socio-economic conditions of their families. The focus of the survey is on the students drawn from government and non-government schools, madrashas, and community (at primary level only) schools located in both rural and urban areas. The reference period is the calendar year 2005. Data presented in various Tables of this study, unless otherwise specified, are from this survey and for the year 2005. The present study has revealed that public spending on primary and secondary education remains low in comparison with other developing countries and insufficient to ensure acceptable quality. Per student public expenditure in a year in government primary schools for 2005 was Tk 1,728 followed by Tk 1,656 in the ebtedayee madrasas, and only Tk 664 in registered nongovernment schools. At the secondary level, the numbers were Tk. 2,461 per student in the government-assisted schools (98 percent of all general secondary schools) and Tk 3,810 for the secondary level (dakhil) madrasas. The data collected from the school records showed the amounts received by, schools from the government for salaries, construction and maintenance and did not include the amounts spent on stipends at primary and secondary levels and distribution of textbooks in primary school. It was evident that very little was available for quality inputs, such as, provision of teaching aids, training and supervision of teachers, provisions for libraries and laboratories. Nor were enough resources to provide for an adequate number of teachers, classrooms and a manageable class size (rather than a primary class of over 50 students and over a 100 in some secondary classes). The meagerness of resources was partially mitigated by household spending on education. Of total spending per child in government primary schools and government assisted secondary schools--the most numerous type of institution in each category--59 percent and 71 percent respectively came from household sources. Cost of private tutors was the single largest item of household expenditure. Forty-three percent of government primary school students and 85 percent of government-assisted secondary school students had private tutors. Household costs for education and dependence on private tutoring clearly aggravated educational inequity. Household spending on education per child of the lowest socio-economic quintile at both primary and secondary levels was less than half of the spending of the topmost quintile. Public spending in different types of institutions varied widely, but no rationale was apparent for this variation. Madrasas at the secondary level received the highest government subvention on a per capita basis and urban institutions were favoured in contrast to rural ones. There was, however, no capitation formula on the basis of the student population in institutions or in geographical units, often applied when the state takes seriously its obligation to provide for compulsory education. A policy reorientation and consequent increase in public resources for education and changes in the budgetary allocation pattern in favour of the poor and the disadvantaged are called for. (Contains 13 tables and 2 figures.) [This paper was published by Campaign for Popular Education (CAMPE).]