Trends in Access and Retention: A Study of Primary Schools in DPEP Districts
By Yash Aggarwal, NIEPA, New Delhi, Published by Educational Consultants India Limited, 1999, New Delhi, p81
The DPEP aims to support the state governments in their efforts to provide universal access and retention, improve learning achievement and decrease dropout rate with a special focus on reducing gender and social inequities to the minimum. The programme has now been in operation for nearly five years and emerged as a single largest development programme for primary education in India and perhaps in the world. At the national and state level, the progress towards achieving DPEP goals is being monitored regularly through multi-faceted strategies; one among these is an annual statistics of the school system, which provides timely and detailed data on enrolment, teachers, classrooms, infrastructure and other related indicators. The objectives of the present study are to examine the trends in enrolment and related indicators since the inception of the programme; and to raise some issues having bearing on the planning and management of the project. The study is based on the data obtained from DPEP districts where District Information System for Education (DISE) has been implemented in the last few years. The study covers all the 42 districts from Phase I, and 77 out of a total of 97 districts covered under Phase II and expansion districts in Phase I states1. This is a third such review in the last two years. The first review was prepared for the Mid-term In-depth Review Mission in early 1998 whereas the second review focused on the analysis of trends in access and retention for the period 1995-96 to 1997-98.
Implementation of DPEP in Phase I began by the end of 1994, whereas the Phase II was sanctioned in 1996-97 but the actual implementation started in 1997-98 and even later in some districts. Most of the project interventions are of long-term nature and therefore the real impact on output indicators like enrolment, retention and internal efficiency is difficult to measure with a data for two years. Experience has shown that establishment of DISE and its stabilization at the district level had taken about 2-3 years for the Phase I districts. The procurement of hardware, recruitment and deployment of manpower, and establishment of other support systems for data collection, validation and analysis are not fully operational in all Phase II districts. It is recommended that the establishment of information systems should form part of the pre-project activities so that a sound data base is available for comparative purposes. The DPEP Bureau may like to specify a time limit within which the EMIS must be made operational.
When fully implemented, the DISE would cover about 230,000 schools, 31 million primary school learners and about 0.65 million teachers in 139 districts spread over 14 states. Encouraged by the success of DISE, many states are in the process of replicating the model in non-DPEP districts also. Ever since its implementation, DISE has provided school level data for four to five years on key performance indicators like school management, GER, NER, teachers, age Grade structure of enrolment and participation by SC, ST and OBCs in primary education. While these data are significant for planning and management of DPEP activities at the district and lower levels, a need is felt to review the scope and coverage of DISE in the following manner a) to expand the coverage of DISE to include upper primary classes; b) to expand the scope to develop systems for computerization of data for alternative modes of education. This has already been attempted and the software has completed the field testing in Madhya Pradesh; c) to simplify the formats in the light of experiences in the collection and validation of school data and d) to explore the possibility integrating the school information database with that of the household and micro planning data.
The recent estimates of literacy at the national and the state level have shown a significant increase from 52 percent in 1991 to 62 percent in 1997. The increase was significant even among the educationally ward states. Large-scale expansion of formal primary education in the early nineties and the innovative strategies of the primary education development projects like DPEP has contributed substantially to these outcomes. The inter/intra-state level variations are large. The differentials manifest in the form of inequities among gender and social groups.
The recognized primary schools are classified as government and private schools. The government continues to be a major provider of primary education in DPEP states, excepting in Kerala, Maharashtra, Tamil Nadu and Karnataka, where the presence of private recognized schools is large. Kerala tops the list with nearly 50 percent private recognized aided and unaided schools. In addition, significant expansion of alternative modes of education and the proliferation of private unrecognized schools in the recent years has widened access even in smaller habitations that do not qualify for a formal primary school. There is no uniform national framework for the development of private schools especially the unrecognized component. The DPEP has pioneered the concept of non-formal education through Alternative Schools and Education Guarantee Scheme (EGS), whereby the schools are provided on demand in unserved habitations. It is reported that about 38,000 alternative schools function in DPEP districts covering about 1.4 million children under various categories. However, three fourth of these schools are in Madhya Pradesh alone. The programme is yet to make a major dent in other educationally ward states like Bihar, Orissa and Uttar Pradesh, where problems of access are far more acute due to dispersed habitations of smaller sizes. Estimates of enrolment in unrecognized schools are not available anywhere. A recent special study conducted by the author in Haryana indicated that 15-25 percent of children were enrolled in unrecognized schools in DPEP districts.
Maintaining balance between demand and supply of schooling infrastructure through construction of schools and additional classrooms was a major thrust in DPEP. Both overcrowding and excess supply could be problematic. The DPEP supplements many other central and state government schemes in a manner that a judicious choice about the investment in school infrastructure development could be made. The ongoing and completed civil works (as in February, 1999) would provide capacity to seat about 2 million children in DPEP districts. Similarly the improvements in the quality of infrastructure through the construction of additional toilets, provision of water supply and construction of boundary wall would contribute to improvements in school environment. A design manual for the construction of school buildings of various types was developed at the national level and shared with the states. Many DPEP states have shown considerable interest in these designs due to their utility and cost effectiveness.
The effectiveness of the civil works programme under DPEP is captured through Student: Classroom Ratio (SCR). The SCR for Phase I districts in 1998-99 was 38.7 as compared to 40.1 in 1995-96. Thus the civil works has largely kept pace with the expansion of enrolment in Phase I districts. However, the progress of construction has been slow in Phase II districts. This matter should receive the immediate attention of the state project authorities, as the SCR for Phase II was 48.8 for 1998-99, a value higher by about 25 percent than the corresponding average for Phase I districts. West Bengal and Uttar Pradesh had the highest average of about 82 and 66 students per classroom thus indicating overcrowded classrooms. Approximately 42 percent schools in West Bengal had a SCR of more than 90, which is unacceptably high. Assam, Uttar Pradesh and Andhra Pradesh also had significantly higher proportion of overcrowded schools. On the other extreme were a large number of schools with lower than 20 students per classroom. This was due to the norm of two classrooms for each primary school under the scheme of Operation Blackboard and inadequate number of students.
The teacher and textbooks remain the most vital inputs for teaching learning processes at primary stage. States have established different norms for provision of teachers. For the Phase I states, the average Pupil: teacher ratio (PTR) was 38.9 in 1998-99 as compared to 38.1 in 1995-96. The PTR for the Phase II districts was much higher at 47.5 for 1998-99. An average PTR of 66 in West Bengal and Uttar Pradesh was the highest among the DPEP states. Average PTR in Gujarat showed increasing trend for the last two years. School level analysis of the PTR indicated that 41.6 percent schools in West Bengal had a PTR of 90 or more than that and was closely followed by Assam, Uttar Pradesh and Andhra Pradesh. PTR has tended to increase due to increasing enrolment and the inability of the states to appoint new teachers. Some states are also experimenting with the recruitment of para teachers even in regular full time primary schools. This is aimed at ensuring immediate availability of a teacher and reducing the unit cost of primary education. Another important factor is that the teacher is also from the local community. The study has also highlighted the need for rationalization of teacher deployment. Such methods are regularly used in many DPEP districts. It is recommended that rationalization of teachers based on DISE data should be attempted in all districts. Crude estimates indicated that the teacher shortage was maximum in Bihar, Uttar Pradesh and West Bengal, if OB norms and a PTR of 40 is to be implemented in all DPEP schools. Notwithstanding these estimates, the states can estimate the exact requirements by varying the PTR according to state norms/practice. It is recommended that detailed exercises in teacher rationalization should be attempted at the Block and cluster level in all DPEP districts using the DISE data.
Achievement of social and gender equity is a paramount concern of DPEP. These concerns for equity are significant not only for access, retention and internal efficiency but also for recruitment and deployment of teachers and educational administrators. The share of female teachers in all DPEP states, excepting Kerala is far less than 50 percent. Tamil Nadu is very close to 50 percent mark in Phase I districts and has already exceeded this threshold in expansion districts. The female representation in teaching workforce was minimum in Phase II states including West Bengal (17.3 percent); Bihar (21.6 percent); Orissa (22.9 percent), Uttar Pradesh (26 percent) and Maharashtra (26.5 percent). The analysis also showed that 61.3 percent schools, with at least one teacher, were without any female teacher and 18.1 percent schools had only one teacher. Thus the need for increasing the share of female teachers and rationalization of their deployment requires attention of the DPEP states. The states have already initiated action on the former but focus on the latter is practically missing. The states should also undertake studies to assess the impact of female teachers deployment on the internal efficiency and important parameters of the quality of education.
Under DPEP, there was a significant decline in the share of single teacher schools. For the Phase I districts, it declined from 18.4 percent in 1995-96 to 13.7 percent in 1998-99. As noted above, the appointment of para teachers ensures easy availability of a teacher and is also a cost effective method. But the effectiveness and long-term impact of such efforts on quality of education should be examined. Among the states, Gujarat and Bihar were having more than 22 percent of their schools as single teacher schools.
The enrolment trends for the period 1995-96 to 1998-99 were examined for the DPEP districts. At the national level, the primary classes enrolment is witnessing a declining growth rate. With 6.2 percent per annum increase in fifties, it peaked during sixties (7.5 percent per annum) and subsequently declined to about 5 percent in seventies. The eighties witnessed a sharp decline to an average growth rate of 2.65 percent per annum, which further declined to 2.1 percent per annum during early nineties. The latest trends, based on provisional data, indicate a growth rate of (-) 0.37 percent per annum between 1994-95 and 1997-98. The national data for 1998-99 is yet not available. The overall declining trend and the recent stagnation in enrolment in some of the educationally ward states is a matter of concern as the country is still far away from achieving the target of universal primary education.
The enrolment growth, even in formal primary schools, in all DPEP districts taken together has significantly outperformed the national trends. It increased by about 13 percent in Phase I districts during 1995-96 and 1998-99, which translates into a compound annual growth rate of 4.1 percent. By including the enrolment of alternative schools, the overall increase was 19.3 percent between 1995-96 and 1998-99, reflecting a compound annual growth rate of 6.2 percent. However, there were large state-to-state variations with Assam showing 2.6 percent decrease in enrolment.
For the Phase II districts, the data was available only at two points of time and therefore clear trend may not be discemable on output indicators like enrolment, repetition, GER and NER, especially for those districts where EMIS is yet to stabilize. Nevertheless the analysis of 75 districts data shows an increase of 1.63% between 1997-98 and 1998-99 in the formal schools and an increase of 2.55% if the data on alternative schools is also included. Disturbing sign is the decline in enrolment in Assam, Bihar and Maharashtra and marginally in Uttar Pradesh. It may not be out of context to mention that the overall outcome of DPEP II would be governed by the outcome of DPEP in Bihar, U.P. and Orissa, all of which are educationally extremely ward.
The slow progress of enrolment increase in many Phase II districts may be an outcome of the following factors: a) late start of project activities in some districts; b) a reflection of the drag in managing large projects even in decentralized mode of operations; c) inability of the states to fully operationalised the EMIS in a time bound manner; d) many districts covered under Phase II are among the least developed districts in the country and hence more time is required for the programme to take off; e) presence of large private unrecognized schools. Only in-depth micro level studies could identify the real factors affecting the enrolment trends.
Many children are reportedly attending private unrecognized schools in almost all states. Statistics on unrecognized schools are not collected and there are variations in state policy towards these schools. The unfinished task for access and enrolment cannot be understood without a detailed estimate of the number and structure of the out-of-school children. The DPEP instituted a comprehensive model of micro planning and community mobilization for achieving the goals of universal access. Despite significant differences in data collection format and the methodology of data collection, the household surveys provided valuable data on out-of-school children. The present study examined the household data from Haryana and Karnataka. About 10-11 percent children in 6-11 year age group were reported to be out-of-school in Sirsa and Hisar districts of Haryana. The household data from Karnataka indicated that the proportion of out-of-school children varied between 2.7 percent for Bangalore Rural to 12.6 percent in Raichur district. The data from Tamil Nadu indicated that most of the children in 6-11 age group were attending school and only a very small proportion of students were reportedly out-of-school. If a systematic planning for universalisation of primary education is to be attempted, the procedures for household data collection need to be standardized to ensure comparability and reliability of important indicators.
The GER based on formal school enrolment in Phase I districts increased from 83.9% in 1995-96 to 93.6% in 1998-99. However, if the enrolment in alternative schools is included the GER increases to 97.3% for 1997-98 and 99.7% for 1998-99. Based on whatever data on enrolment in unrecognized schools is available, it could be assumed that about 8-10% children attend such schools. If the same is added even at a modest rate of 8-10%, the overall GER for Phase I district will be around 107%, which is very close to the threshold for achieving universal primary education. The district level GER (formal + alternative schools) ranged from 76% for Kaithal in Haryana to 118 percent for Latur in Maharashtra. Notwithstanding the high GER, there are pockets of underdevelopment, the identification of which is the most crucial task for the success of DPEP strategies. It is, therefore, important that increased emphasis on micro planning, teacher mobilization, alternative schools and pedagogical reforms should be undertaken on priority basis in districts/clusters with low GER and maximum out-of-school children. Some of these areas may have a large concentration of deprived population, working children and other marginalized groups. There were seven districts with NER>95 percent. Many district reported NER in the range of 75-95 percent. These NERs do not include enrolment effects of alternative and unrecognized schools.
Monitoring Class I enrolment is significant to understand the progress of universal access to primary education. There were 18 districts belonging to Phase I where more than 5% decline in Class I was observed between 1997-98 and 1998-99. Overall the Class I enrolment declined by 4.3% in DPEP Phase I districts. A similar trend was observed in Phase II, where 39 districts registered a decline of more than 5% in Class I enrolment. The overall decline was to the extent of 7.1% between 1997-98 and 1998-99. The reasons for decline in Class I enrolment have not been examined so far but the likely causes could be: (a) decline in the share of underage and overage children in Class I, (b) Comparatively faster increase in intake of unrecognized schools, (c) A decline in the growth rate of age specific population, (d) reduction in repetition rate, and (e) problems with the enrolment statistics.
The GER and NER although useful for examining participation rates, suffer from many inadequacies and needs to be interpreted with caution. The GER for 66 Phase II districts registered a marginal decline from 84.5% in 1997-98 to 83.7% in 1998-99, which is due to the decline in Class I enrolment. However, if enrolment of alternative schools is included, the GER increases to about 85%, a ratio far behind the 100 percent mark. Thus, in the immediate future, considerable efforts are needed to identify geographical areas where access to primary education is still a problem. Most of the habitations in Uttar Pradesh, Bihar and Orissa are small in size and are sparsely populated. In order to achieve universal access and retention, the GER must increase by about 5-7% per annum for the next few years. This is a major challenge for those districts where the project duration is five years as compared to the normal project cycle of seven years, which was envisaged at the beginning of DPEP. Many such districts also happen to be among the most ward.
The analysis of NER shows that still there were 39 out of 66 districts which had less than 70% NER. Low level of NER show the immensity of unfinished task that lies ahead in terms of improving the intake as well as retention. Districts with low NER should be taken up on priority basis for micro planning, community mobilization and identification of the areas where out-of-school children are concentrated. Formal and alternative schooling facilities may have to be provided in these habitations at the earliest.
Grade repetition is one of the factors, which adversely affects the internal efficiency. There are various exogenous and endogenous factors due to which the repetition rate varies from school to school. Technically the repetition rate should be zero for Grades where ‘no detention policy’ is applicable. But in reality the repetition rates were found to be quite significant. Overall, the repetition rates fell from 7.5% in 1995 to 7% in 1996 and to 5.2% in 1997. Assam, Haryana and Tamil Nadu had significantly higher repetition rates as compared to other states. The repetition rate for Class I was approximately 33% for Assam and 11.6% for Tamil Nadu. A similar analysis for Phase II districts showed an overall repetition rate of 8.8% in 1997. States with high repetition rate were: Assam, Bihar, Gujarat and Himachal Pradesh. The Class I repetition rate was 15.6%. This is unacceptably high as one out of six children in Class I fails to reach Class II. Immediate priority should be given to determine the causes and the remedies for high repetition rates in DPEP II. Schools with high repetition rates can be easily identified for remedial action. Reducing repetition is essentially a school related phenomenon.
Under DPEP, the girls’ enrolment increased faster than that of boys and consequently the Index of Gender Equity (IGE) for 23 out of 42 Phase I districts had crossed the threshold of 95%, showing near absence of male-female disparities in enrolment. Eighteen districts were in the range of 85-95 pointing to the need for identification of areas/clusters with low female participation. Once, the needs of these areas are addressed, the IGE will move in the range of 95+. Similarly, 8 districts out of a total 75 districts (Phase II) had a IGE of less than 75. These districts require more intensive efforts to identify the factors for low enrolment of girls. The repetition rates were generally lower among girls as compared to boys in Phase I districts but opposite is the case for Phase II districts. More intensive efforts would be required to reduce repetition among Phase II districts with gender sensitive strategies.
The Index of Social Equity for the SCs has shown improvement both for the Phase I as well as Phase II districts. Some districts continue to depict high degree of social inequities as far as ST enrolment is concerned. Therefore, micro studies should be undertaken to identify areas where the participation of ST children is low. It is expected that with a special focus on the education of tribal areas, the persistence of inequities between tribal and other population groups would be reduced to the minimum in the coming years.
In conclusion, it may be stated that the outcome of DPEP and other interventions in DPEP districts has been positive and pronounced. The growth of enrolment in DPEP districts has been much higher than the national average for the same period. However, the recent decline in Class I enrolment needs to be investigated through special studies. The repetition rates, both for boys and girls, declined more sharply in 1997 as compared to any of the previous years. These still continue to be high, especially in the context of no-detention policy. Accelerating programme implementation would be needed to improve GER/NER, especially in districts where the total project duration is five years instead of the normal seven years. The GER/NER, despite their usefulness suffer from many inadequacies, which limits their application for educational planning. A unique contribution of DPEP has been a large-scale replication of alternative modes of education, especially meant for the children living in smaller habitations where opening of a formal primary school may not be possible. A minimum 5-7 % annual increase in enrolment through formal and alternative modes of education would be required in the next few years to achieve universal access especially in districts with low GER. This can be achieved by increasing participation of various social groups, on the one hand and improving the internal efficiency, on the other. Continuous efforts will be required for ensuring high quality of teaching learning outcomes and developing strategies for sustaining the cost effective innovations. Strengthening and diversification of EMIS to meet the emerging needs would be a positive step in improving the quality of monitoring.