Like National Sample Survey Organisation (NSSO), International Institute for Population Sciences (IIPS, Bombay) conducted a household based `National Family Health Survey' (NFHS) during April 1992 to June 1993. Twenty-five states covering about 99 per cent population of the country were covered and about 90 thousand ever-married women in the age group 13-49 were selected and interviewed in the survey.

Though, largely it was family health survey, information on a number of educational variables, such as, literacy (6+) and school attendance rate have also been generated. In addition, information on behavior of literate/illiterate family and health practiced has also been generated which give some fresh insights and evidences of being literate on these variables. Though, information on these items are also available either through Census or regular publications of MHRD, it is worthwhile to compare these estimates with those generated through NFHS. Data in NFHS is generated through three structured questionnaires, namely, household, woman's and the village questionnaires. Educational information is collected through household questionnaires. `Can you read and write', `has ever been to school', `what is the highest grade completed' and `is still in school?' are some of the questions canvassed in the survey. Like Census, NFHS too simply based on questions rather than based on some test to know ability of the respondent to read and write with understanding. However, in Census, each and every member of house is supposed to ask questions personally, in NFHS, reading and writing head of the household respond ability of household. In woman's questionnaire, questions like `can you read and write' and `what is the highest degree you obtained were also raised. In the village schedule, questions relating to educational facilities and distance from the nearest facility available were also raised but results are not reported in the all-India document.

The sample design adopted in each state was a systematic, stratified sample of household, with two stages in rural and three stages in urban areas. In order to generate estimates, adequate attention is paid and weights are assigned to different states. At the national level, the overall weight for each household assumed to be the product of the design weight for each state and the state weight. One of the major limitations of NFHS is a illiterate respondent. Since, educational variables are responded by them, the quality of data might have adversely affected. Of the total respondents, 63.1 per cent were illiterate and only 11.7 per cent had completed Primary schooling. Not only the respondents were illiterate but even their husband were also illiterate (92.5%) and only few had completed Primary schooling.

The NFHS results with particular reference to educational clientele, literate population and quality of literate produced in terms of their educational level, attendance rate and completed years of schooling have been critically analysed. Keeping in view the available statistics, only two components of EFA, namely, universal enrolment and universal literacy have been critically analysed in the present article.

Invariably, it has been noticed that Census estimates compares well with those produced on the basis of sample survey. Not only literacy rate is comparable, but its male-female and rural-urban distribution also matches well with those produced by NFHS. The discrepancy between the two estimates is small and insignificant which is perhaps due to difference in date of reference and definitions used. The comparison at the state level also reveals that both the estimates of literacy rate are quite comparable and no significant deviation is noticed. The high male-female and rural-urban disparity noticed in Census is reconfirm by the NFHS estimates. Irrespective of area, the highest literacy rate is noticed in age-group 10-14 years but about two third population of 6-14 years is still found to be illiterate. Further, it has been noticed that of total littermates of age-group 6-14 years, only a little more than twenty five per cent had completed Primary schooling.

A large gap has been noticed between GER at Elementary level and corresponding attendance rate. If 25 per cent is considered an estimate of overage and underage children, the NER at the elementary level comes out to be 68.37 per cent. Lower attendance rate especially in age-group 6-10 years may be due to two reasons. First, a large number of overage and underage children are included in the Official enrolment and second, all children who have taken admission may not be attending schools or they might have dropped-out from the system.

The statewide attendance rate shows that in a majority of states, a large number of children of age-group 6-14 years were out-of-school. The highest percentages of out-of-school children in year 1992-93 were in Bihar followed by Rajasthan, Uttar Pradesh, Madhya Pradesh. The low median number of completed years of schooling indicate that a large number of children dropped-out from the system before they complete a particular education cycle. The median number of years of schooling for girls is 0.0 which indicate that the majority of females (all ages) have never been to school.

Thus, information on attendance rate, completion rate and non-enrolment ratio gives enough indication that goal of education for all in the country is not right in the sight. Sporadic attempts have been made in the recent past to bring all children under the umbrella of education. But, coverage both interms of area and out-of-school children do not suggest that a significant dent can be made in the near future. However, if the existing programme spread to new areas and entire out-of-school populations is covered and retain in the system, the same can dramatically change the scenario and goals can be realised earlier than projected. Unless, age-grade matrix is available, the true picture of UEE cannot be presented. It is high time that age-grade matrix should form part of regular collection of statistics which can be used for proper monitoring and implementation of different programmes currently under implementation in the country.