ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Contextualising Inter-, Intra-religious and Gendered Literacy and

Indian Muslims as a whole lag behind other religious communities in terms of educational attainment. This paper seeks to place Muslim literacy and education as relational and its locatedness in a larger spatial context in order to propose that there can be no one unilinear process in conceptualising religious differences in matters of literacy and education, which might be produced variously through individuals and the larger structures of which individuals are a part.

Special articles

Contextualising Inter-, Intra-religious and Gendered Literacy and Educational Disparities in Rural Bihar

Indian Muslims as a whole lag behind other religious communities in terms of educational attainment. This paper seeks to place Muslim literacy and education as relational and its locatedness in a larger spatial context in order to propose that there can be no one unilinear process in conceptualising religious differences in matters of literacy and education, which might be produced variously through individuals and the larger structures of which individuals are a part.


ince independence there has been a remarkable expansion of educational opportunities at almost all stages and avenues of specialisation and for various segments of population [Kamat 1989; Tilak 2004] and yet access remains far from equal even in terms of universalisation of elementary education [Singh 2005; Tilak 2006] in the post-independence period.1 The most pervasive axes of inequalities are caste, gender, region and religion out of which disparities in religious groups remain one of the least explored themes.

Until very recently there were no reliable statistics to take stock of the educational situation of religious minorities in India and much of the discussions were based either on guesstimates or sporadic micro-level surveys, primarily pointing out how Muslims occupy the lowest rungs in educational attainment [Kareem 1989; Mondal 1997; Salamatullah 1994; Khalidi 1995; Jeffery and Jeffrey 2000; Shariff and Razzack 2006]. However, it was for the first time in 1987-88 that the National Sample Survey (NSS) in its 43rd round took up educational and other parameters related to religious groups, which continued in the subsequent 50th and 55th rounds. These documents substantiated the claims that there exist considerable educational gaps between and among religious groups and that Muslims lag behind other communities. The Census 2001 supported the findings of the NSS reports.2 So do the data from other sources [Shariff 1999]. Although these statistics do point out the disparate levels in educational attainment of Muslims vis-à-vis other religious groups, the explanatory dimensions remain inadequately attended too despite rising awareness and aspirations for education among Muslims in general and poor Muslims in particular [Hussain 2005].

Quite often, lower literacy amongst Muslims is seen as entrenched in their religious location and therefore intrinsic to the community [Hunter 1869; Baig 1974; Sharma 1978] whereas some hold the socio-economic status of Muslims responsible for the absence of demand for literacy in the community [Jain 1969; Ahmad 1981; Jain 1986; Ahmad 1987; Ansari 2001]. We argue that these explanations are rather simplistic as well as flawed and that literacy attainment amongst Muslims needs to be scrutinised through multilayered, complex and spatially contextualised locations rather than as undifferentiated averages [Shariff and Razzack 2006]. This is essentially because in addition to socio-economic, historical and cultural specificities that may have created differences in Muslim literacy, their literacy tends to follow trajectories similar to other religious groups (predominantly Hindus), albeit at a slower pace, as Muslims do not operate in contextual isolation [Hasan and Menon 2004].3

To caution, we neither intend to undermine the differences that may arise because of communal biases within overarching structure, nor do we intend to play them up in reifying the process of “othering”, and yet the paper seeks to place Muslim literacy as relational and its locatedness in a larger spatial context in order to propose that there can be no one unilinear process in conceptualising religious differences in literacy, which might be produced variously through individuals, and through structures of which individuals are part [Archer 2003; Balagopalan and Subramanian 2003].

This study derives from a doctoral work and is based on the survey of 300 households in the summer of 2002 in two blocks in the districts of Patna and Purnia in Bihar.4 Although these two districts contain a sizeable population of Hindus and Muslims (Muslim households accounted for 60 per cent in the sample and the rest belonged to the Hindus), they occupy almost polar ends in terms of developmental parameters. Given this, the varying spatial contexts in combination with varying educational and other socio-economic characteristics are expected to have a differential impact on Muslim literacy.

Section I of this paper provides an overview of the literature on educational backwardness of Muslims. Section II follows with a detailed account of the educational profiles of the religious groups, mainly Hindus and Muslims, drawn from the field survey in rural Bihar. Section III is a comparison between two differing localities in order to spatially contextualise literacy and educational disparities between the two religious groups. In Section IV an attempt is made towards developing an explanatory framework so that some of the processes contributing to emerging disparities can be understood. Section V concludes the paper.

I Lower Educational Attainment among Muslims: Theoretical Formulations

There are two popular explanations for low literacy amongst Muslims. One explanation has its root in what may be called “particularised theology hypotheses”.5 This hypothesis, in brief, assumes that religion provides a pattern of belief system, prescribes values, norms, attitudes and sets behaviour and orientation towards life, which individuals share in groups. One religion differs from others in such matters and plays an essential role in controlling, influencing, determining, limiting and guiding the socio-economic achievement of its followers [Mayer and Sharp 1962; Versoff et al 1962; Muller 1980; Hirshman and Falcon 1988; Sanders 1992].

In the Indian context, it was Hunter (1869) who for the first time argued that for Muslims, education has intrinsically been linked with religion as Islam emphasises religious education. It is because of this that Muslims give preference to Islamic education imparted through madrasas and ‘maktabas’ over modern education in schools or other institutions. He further observed that Muslims saw the new system of education during the British regime as opposed to their tradition, unsuited to their requirements and hateful to their religion. Baig (1974) points out that the traditional orthodox, supernaturalism of Islam was totally opposed to a scientific outlook and unless social reforms take place within the Muslim community, Muslims will continue to suffer from social degeneration, economic stagnation and educational backwardness. Sharma (1978) seems to endorse this hypothesis as he concludes that economic compulsions and constraints account only partially for the Muslim community’s educational backwardness; it is their preoccupation with religious teaching in the curriculum that to a large extent is responsible for their aversion to a modern system of education. In a nutshell, such arguments strongly potray Muslim religious orthodoxy as preventing the community from taking advantage of expanding educational opportunities, resulting in their underrepresentation in formal schools.

This explanation has met with severe criticism for several reasons. For one, even on the theological plane, the Qurän prescribes compulsory education for all [Ali 1950; Rafeda al Hariri 1987]. In the light of the Quranic references the Prophet exhorted the Muslims to seek ‘ilm’ (knowledge) even if they had to go as far as China. In fact, it has been documented that Muslims not only received knowledge from different countries, but also from anyone who was capable of imparting knowledge, and also contributed to different branches of knowledge including science [Arnold and Guillaume 1931; Levy 1962; Peacock 1978]. Further, neither the Qurän nor the Hadith6 enjoins upon Muslims to refrain from acquiring modern education [Peer 1990]. More importantly, such explanations tend to treat Muslims as a monolithic community and Islam as all pervasive and primarily prescriptive, governing every aspect of Muslim life from family structure, marriage, and divorce on the one hand to poverty, employment and education on the other [Hasan and Menon 2004]. However, several empirical studies suggest that Muslims are also a

Table 1: Socio-economic Profile of Sample Villages

Background Patna Villages Purnia Villages All
Total Hindu Muslim Total Hindu Muslim Total Hindu Muslim
Total population 837 332 505 1120 382 738 1957 714 1243
Male population 453 184 269 563 186 377 1016 370 646
Female population 384 148 236 557 196 361 941 344 597
Sex ratio 847 804 877 991 1048 960 927 929 924
Age structure
0-14 42.1 41.0 42.9 43.8 39.7 45.9 43.1 40.3 44.7
15-39 41.8 42.2 41.5 41.6 45.0 39.8 41.7 43.7 40.5
40-59 14.7 14.8 14.6 13.9 14.7 13.4 14.2 14.8 13.9
60 and above 1.4 2.1 1.0 0.7 0.5 0.8 1.0 1.3 0.9
Literacy rate
Overall literacy rate 71.4 67.8 73.7 44.8 55.8 39.0 56.1 61.4 53.1
Male literacy rate 72.8 69.6 75.1 55.7 64.0 51.6 63.3 66.8 61.4
Female literacy rate 69.5 65.5 72.0 33.8 48.0 26.0 48.4 55.5 44.2
Work participation rate
Overall 26.4 29.5 24.3 29.0 31.4 27.8 28.0 30.5 26.4
Male 43.7 44.0 43.5 50.4 51.6 49.7 47.4 47.8 47.1
Female 6.0 11.5 2.5 7.7 12.8 5.0 7.0 12.2 4.0
Farmer/cultivator 5.0 5.1 5.0 16.1 12.5 18.2 11.7 9.0 13.3
Agricultural labourer 10.5 15.3 6.6 40.0 41.0 38.0 27.5 29.4 26.3
Non-agricultural labourer 25.5 32.7 19.7 13.0 5.0 17.2 17.9 17.4 18.1
Self-employed 32.3 20.4 41.8 23.1 26.7 21.1 26.8 24.0 28.7
Services 26.8 26.5 27.0 9.1 15.0 5.7 16.2 20.2 13.6
Agricultural landownership**
Landless 80.0 63.3 91.4 55.6 55.0 56.1 65.3 58.3 70.1
Less than 2.5 acres 13.4 26.5 4.3 25.3 25.4 25.2 20.5 26.0 17.0
2.5-4.9 acres - - - 7.3 11.3 4.7 9.4 9.2 9.5
5.0 acres and above - - - .6 - .9 4.4 6.7 3.4
Notes: * As percentage of workers; ** As percentage of households.
Source: Field Survey 2002.
1614 Economic and Political Weekly May 5, 2007

differentiated community along class and caste-like hierarchies and in many ways they closely follow non-Muslim norms in a given region rather than Muslim selsewhere [Ahmad 1978; Engineer 2001; Hasan and Menon 2004].7

In addition to religious essentialisation, prevailing socioeconomic conditions among Muslims captured under “characteristics hypotheses”8 are often proposed to explain lower educational attainment among them. According to this proposition, differences in educational attainment across various segments of society result because of their differing levels of socio-economic status, which in turn impacts the (social) demand for education across social classes. As a result, education does not hold universal appeal for all social strata [Morrish 1972; Muller 1972; Astone et al 1991; Warren 1996]. In other words, the educational status of a group or a sub-group of population largely depends upon the size of the strata that usually aspires for education. If so, the educational backwardness of Muslims can be seen as an outcome of the small proportion of Muslims that traditionally values education [Ahmad 1981].

The characteristics hypothesis has been well received in academic circles and yet its substantiation against ground realities is inadequate. For example, it remains to be seen whether people belonging to the same religious groups and the same socioeconomic background within the groups (here, Hindus and Muslims) respond to the externalities in terms of educational opportunities comparably or with significant degrees of difference. It is possible to argue, for example, that the answer to this teaser, even as socio-economic endowments for the two groups are accounted for, could be either affirmative or otherwise. Further, it is also possible that the comparable strata of the two groups behave in an entirely different manner in one location vis-à-vis the other. White (1982) and Fejgin (1982) in their studies elsewhere found that socio-economic factors did not account for all the variations in educational attainment between racial and ethnic groups. What is being suggested is that instead of essentialising the so-called “homogeneity” of “Muslim characteristics”, more pertinent questions to be asked are: who are these Muslims? Where are they located? Who are the referent communities and how do they compare with these referent communities?

Within this overall conceptual framework however, gendered disparities need further probing as girls/women – Muslim or otherwise – carry an additional burden of subordination even as they do not occupy monolithic spaces within this class and castelike hierarchies introducing further complexities.9

II Educational Attainment of the Religious Groups: Findings from the Field

Table 1 provides an overview of some socio-economic parameters of the sample population. Even a cursory look at some of the socio-economic variables such as literacy, work status and land ownership confirms that Muslims in general and Muslim women in particular are disadvantageously placed vis-à-vis their Hindu counterparts, a situation widely prevalent in India. However, as the subsequent discussion reveals, several intersecting variables make the situation much more complex.

Although literacy rate10 is a crude indicator of educational attainment, it is vitally important as alphabetisation provides the basal stratum for the subsequent development of a multi-level educational pyramid [Raza 1990]. Table 2 presents the educational level and disparities therein of the sample population by sex, religion and location.

Withholding the overall literacy rate which is higher than “the state average” (47.5 per cent) for the sample population (56.1 per cent),11 the usually observed disparity between male/female literacy is seen here as well.12 Similarly, in keeping with observations elsewhere in the paper, literacy rates differ in terms of religious affiliation, skewing in favour of Hindus and in terms of gender, in favour of males. With an overall lower level of literacy and education females in general, at every level gender disparity in literacy is higher among Muslims as compared to Hindus. Although the subordinate location of females is common among both Hindu-Muslim groups, it is significant to note that the inter-religious disparities between Hindu-Muslim females are higher than the corresponding disparities for Hindu-Muslim males suggesting that Muslim females carry an additional burden of belonging to a minority community.

Age specific enrolment rate13 is a good measure of participation in and utilisation of educational opportunities available. It reflects not only the degree of keenness of the society towards education, but also the magnitude of value attached to it. Overall, about half of those between seven and 18 years were enrolled at different

Table 2: Literacy, Educational Attainment, Gender and Inter-religious Disparities

Educational Indicators All Hindu Muslim Gender Disparity (Inter-religious)
Intra-religious Disparity
Total Male Female Total Male Female Total Male Female All Hindu Muslim Total Male Female
Literacy rate 56.1 63.3 48.4 61.4 66.8 55.5 53.1 61.4 44.2 0.12 0.08 0.14 0.06 0.04 0.10
Age-specific enrolment rate
7-11 68.8 70.9 66.5 73.3 72.0 74.5 66.7 70.5 62.3 0.04 -0.02 0.08 0.60 0.01 0.11
12-14 55.0 63.0 46.0 72.0 78.6 63.6 47.3 55.2 38.5 0.18 0.14 0.20 0.26 0.23 0.29
15-18 27.7 39.0 15.3 34.0 46.0 19.4 24.6 35.2 13.4 0.46 0.45 0.47 0.16 0.14 0.17
All ages 53.5 59.6 47.0 59.6 65.2 52.3 46.2 53.6 38.0 0.14 0.13 0.17 0.12 0.08 0.17
Educational levels attained
Up to primary-middle 39.6 40.9 38.3 38.9 38.8 39.0 40.0 42.1 37.8 0.03 0.00 0.05 0.00 -0.02 0.01
Up to secondary 11.8 14.0 9.5 17.7 19.1 16.4 8.4 11.1 5.5 0.17 0.07 0.30 0.32 0.23 0.47
Above secondary 8.3 11.9 4.4 10.8 16.1 5.2 6.9 9.6 3.9 0.43 0.49 0.50 0.19 0.25 0.12

Notes: Disparity between the two religious groups is computed by using the (improved) disparity index as developed by Kundu and Rao (1985). The index is as follows: DS = log (X2/X1) + log (200-X1)/(200-X2), where X2>= X1and X2 and X1 are the literacy rates/enrolment rates of the two groups between which disparity is calculated. Larger the value greater is the disparity. In case of religion, the reference category is Hindus whereas in the case of gender the reference category is males. Negative value indicates that the reference category has lower value of disparity than that of the group with which comparison has been done.

Source: Field Survey 2002.

levels of education. For every five males three were enrolled while the corresponding ratio for females was less than one for every two. The enrolment rates are the highest in the 7-11 age group with consistent decline as well as widening disparities with increase in age. Although at each successive stage girls do lag behind boys, the disparity becomes acute only in the age cohort of 14-18 years.

In the age cohort of 7-11 years, about 69 per cent of the children were enrolled without much difference between boys and girls with a disparity value of 0.04 between the two. However, the enrolment rate came down to 55 per cent and the disparity value increased to 0.18 in the age group 12-14. This trends continues with the age cohort 15-18 when the children should be at the secondary level; for about every three children in this age group only one was continuing. One of the striking features, however, is very high disparity between boys and girls (0.46). The most striking disparity occurs at the secondary education level as the gender disparity among Hindus is very low (0.05) as compared to Muslims (0.30). The completion levels are much lower for Muslims – less than 10 per cent of boys complete secondary education (for Hindus, the percentage is 15).

It is neither surprising, nor unexpected to encounter girls approaching puberty dropping out of schools. As a daughter is to be married eventually the benefits of educating her would accrue to her in-laws, her success in education and employment is therefore not seen as critical or pressing to her natal family as that of the son. Privileging boys over girls as a universal construct cuts across religious locations but Muslim girls are far more precariously situated as compared to Hindu girls even as Muslim boys are not particularly advantaged.

The incidence of discontinuation/dropout relates to students who drop out at various levels of education after they are enrolled. Nearly 21.3 per cent of children aged 7-18 had discontinued study at one or another stage of schooling. The proportion of discontinued children was higher among girls than boys. As against 18 per cent boys who had dropped out, the proportion of dropouts/discontinuation among girls was 25.4 per cent (Table 3). As is evident from Table 3, the discontinuation rate among

Table 3: Discontinuation Rate by Religion and Sex (7-18)

Religion Total Discontinuation Rate Boys Girls
Hindus Muslims All 19.1 22.6 21.3 14.0 20.2 18.0 24.7 25.8 25.4

Source: Field Survey 2002.

Muslim children was higher than for Hindu children. Muslim boys were far more prone to dropout than Hindus. However, the likelihood of discontinuation of girls in both the religious groups was almost the same.

III Contextualising Educational Attainment

Scholars have pointed out the historical persistence of spatially entrenched literacy patterns, i e, literacy variation across Indian social space that has remained unchanged for decades [Sopher 1980; Raju 1991; Chatley 1995; Nuna 1993]. Such spatiality has also been shown to influence religious groups [Ahmad 1993]. We argue that in terms of literacy levels also Hindus or Muslims cannot be treated as contextually isolated groups and their literacy experiences need to be spatially embedded. In order to substantiate this proposition, we analyse the relative positioning of these two religious groups in two very different contexts

– villages of Patna and Purnia, which are located at the polar ends of developmental trajectories. The underlying concern is whether or not developmental externalities succeed in overcoming socially induced biases.

It is quite possible that despite high male literacy female literacy remains low, but it can also be argued that in the company of a highly literate population, the gains should filter down to less advantaged segments of population leading to distributive justice along with growth in general. It can indeed be seen that overall literacy rates are higher in villages in Patna than those of Purnia and so are the reduced gender disparities (Table 4). Significantly, female literacy – a stock variable – is quite high in Patna as compared to Purnia suggesting that the literacy attainments in Patna were somehow better distributed across various agecohorts over the years.14

Significantly, in Purnia where overall literacy was lower, Muslims were far behind the Hindus and the accompanying sex disparities were also higher in Purnia, which were almost double amongst the Muslims as compared to the Hindus. In Patna however, the Muslims had a slight edge over Hindus in terms of overall and male-female literacy (Tables 4 and 5). Part of the explanation for the divergence lies in the relatively higher proportion of Muslim workers in services in Patna compared to their Hindu counterparts. Thus, there were clear indications that the lower the overall level of literacy rates, higher was the disparity among different segments of the population in general and sexbased disparity in particular. It is with increase in female literacy that disparities go down rapidly [Sopher 1980].

Table 4: Educational Attainment by Religion and Gender: Patna

Educational Indicators All Hindu Muslim Gender Disparity Hindu-Muslim
Total Male Female Total Male Female Total Male Female All Hindu Muslim Total Male Female
Literacy rate 71.4 72.8 69.5 67.8 69.6 65.5 73.7 75.1 72.0 0.02 0.03 0.02 -0.02 -0.03 -0.02
Age-specific enrolment rate
7-11 77.5 73.2 81.7 75.6 70.0 80.0 78.4 74.1 82.3 -0.08 -0.09 0.04 -0.03 -0.04 -0.02
12-14 73.7 75.2 75.0 85.2 76.0 98.3 67.3 70.0 65.5 0.00 -0.02 -0.03 0.16 0.05 0.30
15-18 45.0 53.0 33.3 41.2 45.0 33.0 47.3 58.4 33.0 0.25 0.16 0.31 -0.08 -0.15 0.00
All ages 65.2 67.1 63.0 67.0 63.6 70.3 67.2 69.1 66.4 0.04 -0.07 0.03 0.00 -0.05 0.03
Educational levels
Up to primary-middle 45.9 40.2 52.4 42.6 39.5 46.4 48.8 40.7 56.2 -0.10 -0.19 0.13 -0.05 0.00 -0.07
Up to secondary 18.0 21.2 14.3 21.0 24.5 16.8 16.1 19.0 12.8 0.17 0.13 0.16 0.11 0.11 0.12
Above secondary 13.8 19.3 7.6 12.5 17.7 6.4 14.7 20.4 8.4 0.40 0.41 0.37 -0.06 -0.07 -0.1
Source: Field Survey 2002.
1616 Economic and Political Weekly May 5, 2007

A a closer look at inter-religious comparison reveals that in general the Hindus have a higher enrolment rate in every age group compared to the Muslims. While half the Muslim boys in the age group 7-18 were enrolled, the corresponding figure for Hindus was about three-fifths. It is evident that among older age groups the participation of Muslims is lower and the Hindu-Muslim disparity is larger. Within an overall dismal situation for Muslims, girls face an even worse situation. However, the enrolment rate varied in the districts under study. In Purnia, as expected, the enrolment rate was much lower than in Patna where less than half the children in the age group 7-18 were on rolls. In Patna the enrolment rates were comparatively much higher. The religious differential in age specific enrolment was too small to analyse and except for the age group 12-14, the enrolment rate among Muslims was marginally higher in all age groups. This marginal difference in enrolment rates between the two communities could be explained, however tentatively, by the insignificant socio-economic difference between the two groups in Patna (see Table 1).

IV Determinants of Disparity in Educational Attainment

Although limited in its coverage, the analysis supports our contention that membership of any religious group does not necessarily imply its backwardness partly because locational specificities do impact one group’s behaviour vis-à-vis the other. In this case, Muslims followed the referent Hindu group – the demonstration effect – and also because in a more developed environment, the access to education is enhanced for all to take advantage of. Conversely, if the context is one of restrictive opportunities, only a small segment of population is able to access limited resources while historically and socio-culturally marginalised sections continue to remain outside the orbit of development for lack of that extra push required in such circumstances. Certain structural vulnerabilities due to historical processes may still continue, but these vulnerabilities cannot be framed as being intrinsic to members of a particular community by virtue of their religious locations. The following section addresses some of these issues.

Literacy a social parameter acts both ways – as an input and output variable – in comprehending developmental outcome, which makes the task of positioning explanatory components in the literacy outcome somewhat difficult. This dual relation can only be explained partially and yet certain predictor variables can be picked up from existing literature. For example, apart from the demographic and social characteristics such as age and sex, economic attributes like the occupation of the head of the household and standard of living are crucial for suffusing aspirations and in allowing access to educational opportunities. In the ensuing analysis, age specific enrolment is taken as dependent upon the following household attributes. The discussion is based on logistic regression.15

Education of the Head of the Household

It has been observed that education is a priority with those who are already educated and the educational status of the head of the household affects the educational attainment of children both directly and indirectly – directly in the sense that the household has already developed a taste or desire for schooling. Also, an educated head of the family or more particularly educated parents have the first-hand experience of the advantage and benefits of being educated. The indirect effect of the educational status of the head comes in the form of motivation and inspiration, an indispensable ingredient for children’s educational attainment [Cohen 1965; Duncan 1967; Pearse 1985; Gore 1994; Haveman and Wolfe 1995].

The net effects of predictor variables on the likelihood of children getting enrolled are shown in Tables 6 and 7. Taking the two tables together, when all other variables are controlled, education of the head of the household, particularly those educated beyond the middle school seems to have a significant and positive impact on children’s chances to be in schools as compared to the children of illiterate heads of households. This stands true of both Purnia and Patna villages. However, while this association seems to operate for boys’ schooling, the same net effect is not observed in case of girls though education of the head did show a positive bearing on their chances to be enrolled as well. It is important to note that this deviation in the regression analysis may be because of other stronger (sometimes nonquantifiable) factors operating such as “distance of school from home” and “marriage of the girl child overtaking concern for her education” as previously discussed.16

Occupation of the Head of Household

Studies carried out in both developed and developing countries have demonstrated that educational attainment is the function of the type of work in which family members engage. It can be recalled from the preceding discussion that there exists a close link between occupation and aspiration for education. Occupation of the head is, therefore, expected to exert direct and immense influence on the participation of children in the sphere of education [Jain 1969; Ahmad 1981;

Table 5: Educational Attainment by Religion and Gender: Purnia

Educational All Hindu Muslim Gender Disparity Hindu-Muslim Indicators Disparity

Total Male Female Total Male Female Total Male Female All Hindu Muslim Total Male Female

Literacy rate 44.8 55.7 33.8 55.8 64.0 48.0 39.1 51.6 26.0 0.22 0.13 0.30 0.15 0.09 0.26

Age-specific enrolment rate7-11 62.3 69.3 54.4 71.7 73.3 70.0 58.0 67.6 46.7 0.15 0.03 0.22 0.14 0.05 0.25 12-14 38.1 54.3 18.4 56.5 82.0 33.3 31.1 46.0 11.5 0.57 0.54 0.68 0.33 0.37 0.51 15-18 14.5 25.5 4.8 26.5 46.7 10.5 9.6 17.5 2.3 0.77 0.74 0.91 0.48 0.50 0.67 All ages 43.0 54.0 31.1 55.6 68.0 44.3 37.5 48.6 24.8 0.30 0.26 0.36 0.22 0.20 0.30

Educational levels Up to primary-middle 34.7 41.4 27.9 35.7 38.2 33.3 34.2 43.1 24.8 0.17 0.06 0.24 0.02 -0.04 0.13 Up to secondary 7.1 8.1 6.0 15.0 13.8 16.0 2.9 5.3 0.4 0.13 -0.05 1.12 0.71 0.41 1.60 Above secondary 4.0 5.9 2.0 9.2 14.5 4.3 1.0 1.6 0.3 0.47 0.52 0.72 0.96 0.95 1.15

Source: Field Survey 2002.

Economic and Political Weekly May 5, 2007 Chernichovsky 1985; Burney and Irfan 1991; Connel 1994; Shariff 1999].

Occupational differentiation is reflected in educational differentiation in a variety of ways. There may be a distinction between farming and non-farming households with a household with agriculture as the sheet anchor of income not attaching priority to educational achievement since the occupation does not demand it as a prerequisite. In contrast, white-collar jobs and services may require some minimum level of education. Further, distinction between the self-employed and those employed in whitecollar jobs is arguably linked with educational attainment also since self-employed households may have a greater capacity to provide ready employment for children without requiring educational qualification [Ahmad 1981; Pearse 1985]. The class distinction between the working class and the middle class is also linked with educational attainment. Research shows that working class children, particularly those of unskilled and manual workers, are usually under-represented in terms of educational attainment [Doughlas et al 1968]. The class factor for which occupation can be taken as a proxy variable also works in terms of motivation, orientation and aspiration [Rose 1968; Fraser 1959; Jain 1969; Ahmad 1981].

It can be observed from Table 7 that the children of those employed in the service sector were most likely to be enrolled in schools (at 1 per cent level of significance) and those of the agricultural labourers were less likely to when compared with the children of farmers. Compared to the children of cultivators, children of agricultural labourers had lower chances to be enrolled in schools by about 70 per cent. These results match with observations by others in Bihar [Jabbi and Rajyalakshmi 2001].

However, differences could be seen in terms of the location and sex of the children (Table 7) as well. In Patna villages the difference in the probability of enrolment between cultivators and agricultural labourers was significant, albeit at a lower level of significance at 10 per cent. In comparison although the children of agricultural labourers have far fewer chances to be in schools as compared to those from cultivating households (significant at one per cent level) in Purnia, service households there also do not seem to exert much influence on children’s propensity to be in schools as compared to Patna. This suggests, as has already been said earlier, that in a situation where literacy levels are generally higher, even poorer segments of population envision the possibilities of being literate. There exists what can be termed as “latent social pressure” to “conform” to others. Withholding this, however, the differences between girl children of households engaged in agricultural labour and those belonging to cultivating households were highly significant (at 1 per cent level). Even as the boys’ propensity to be enrolled in schools was not significantly influenced by the households’ occupational status, girls seemed to bear the disproportionate brunt of their gendered location (Table 6).

Standard of Living

Standard of living17 is used here as a proxy for income of the household. The index is used because of well known difficulties in capturing income directly. In contrast, physical assets and other household amenities are visible markers and relatively easier to record. Also material possession does reflect the level of income of the household, as usually high incomes are associated with better asset holdings.

It is generally argued that the higher the level of income of the household, the higher is the demand for schooling [Wolfe and Behrman 1984; Hossain 1990; Knodel et al 1990; Haveman et al 1991; Ghosh 1991; Hill and O’ Neill 1994; Parish and Willis

Table 6: Logistic Regression on Current Enrolment

Predictor Variables Overall Patna Purnia Exp (β) Exp (β) Exp (β)

Religion – – Hindu (rc) – – – Muslim 0.914 .891 .992

Age ––– 7-11(rc) – – – 12-14 1.222 1.879** .906 15-18 0.180* .141* .201* Sex ––– Boy (rc) – – – Girl 0.584* .766 .445* Education of HH – – – Illiterate (rc) – – – Primary-middle 1.026 1.229 1.226 Secondary and above 1.993** 2.097** 2.104*** Occupation of HH – – – Cultivators/farmers (rc) – – – Agri-labourer 0.315* .195*** .346* Non-agri-labourer .095 .681 1.445 Self-employed 1.361 1.065 1.559 Services and others 3.670* 2.891** 1.889 Standard of living – – – Low (rc) – – – Middle 1.159 1.244 1.254 High 2.647* 3.098* 2.112*** Total sample 699 307 392

Notes: rc: Reference category.

* The coefficient in the underlying logistic regression differs significantly from zero at the 1 per cent level. ** The coefficient in the underlying logistic regression differs significantly from zero at the 5 per cent level.

*** The coefficient in the underlying logistic regression differs significantly from zero at the10 per cent level.

Table 7: Logistic Regression on Current Enrolment by Sex

Predictor Variables Overall Boys Girls Exp (β) Exp (β) Exp (β)

Age 7-11(rc) – – – 12-14 1.241 1.345 1.321 15-18 .193* .216* .114*

Education of HH Illiterate (rc) – – – Primary-middle 0.977 0.849 0.943 Secondary and above 2.035* 3.208* 1.638

Occupation of the HH Cultivator (rc) – – – Agricultural labour 0.344* 0.503*** 0.137* Non-agricultural labourer 0.91 .452*** 1.777 Self-employed 1.357** 1.714** 1.116 Services and others 3.486* 4.028* 5.044*

Standard of living Low (rc) – – – Medium 1.16 1.153 1.321 High 2.395* 2.676* 2.572*

Religion Hindus (rc) – – – Muslims 0.928 0.981 0.85

Total sample 699 366 333

Notes: rc: Reference category.

* The coefficient in the underlying logistic regression differs significantly from zero at the 1 per cent level. ** The coefficient in the underlying logistic regression differs significantly from zero at the 5 per cent level.

***The coefficient in the underlying logistic regression differs significantly from zero at 10 per cent level.

Economic and Political Weekly May 5, 2007

1994; Guha Roy et al 1995; Tansel 1997; Alderman et al 1997; Filmer and Pritchett 1999; Banerjee 2000]. In keeping with this observation, high standard of living appears to have a far more significant influence on the probability of getting children enrolled as compared to the households with low standard of living if all other variables are controlled. The odds are quite high for both boys and girls at one per cent level of significance (Table 7). This is not unexpected since low standard of living implies a certain status constraint, i e, of belonging to poor households for whom both the direct and opportunity cost of schooling their children is very high [Mehrotra 1995; Panchmukhi 1990; Tilak 1995] whereas the households enjoying a high standard of living might have higher aspirations and demand for schooling as not only can they support their children’s schooling but also aspire for white collar jobs for their children [Ahmad 1981; Parish and Willis 1994; Tansel 1997; Filmer and Pritchett 1999]. In a nutshell, a high standard of living seems to close the gender gap in matters of schooling. However, this association is not as straightforward as it seems at the outset as locational specificity does intercept the influence of standard of living on education. For example while in Patna a higher standard of living is observed to have profound and positive effects on children continuing schooling as compared to those with low standard of living, this association is significant, but very weak in Purnia villages, implying once again that for educational motivation the larger social environment matters more than individual household attributes or a few economic specifics.

Age and Sex

Age of the children also affects their chances of being in school. In general, the higher the age cohort, the lower is the proportion of children enrolled. From Tables 6 and 7, it can be observed that for the children in the age cohort 14-18 the chances are far less (at 1 per cent level of significance) to be enrolled when compared to the children in the age group 7-11. In the former age cohort (15-18), the chance was lower by more than 80 per cent. This observation holds true irrespective of gender and location, suggesting that if children are not in schools at appropriate ages, it becomes increasingly difficult to get them to schools at advanced ages, a phenomenon which is rather common in the study area and operates independent of developmental context. Several researches have highlighted the reasons why children, especially beyond the age of 14 tend to have lower propensity to schooling. One of the main reasons is that by that time most of households choose to employ their children so as to contribute to household income [Kanbargi and Kulkarni 1991; Leiten 2000]. This is particularly true for poor and self-employed households. While employment or opportunity cost has a more significant role to play in the case of absentee boys than girls, the negative association between schooling and age in the case of girls can largely be attributed to several socio-cultural barriers.

Overall, gendered location seems to have a profound effect on access to schooling. It can be observed that girls have a lower probability of getting enrolled in schools as compared to boys if all other variables including religion are controlled. Their probability is lower by about 42 per cent as compared to boys in general. However, once again Patna villages and Purnia villages have different profiles, which are reflected in a statistically insignificant gender difference in the two locations. Girls have better chances to be in school vis-à-vis boys in Patna in contrast with Purnia girls whose chances of schooling are far lower than that of Purnia boys. To reiterate the relatively widespread diffusion of literacy and education in Patna has meant that even the less privileged segments of population share the gains of literacy and education, whereas where the overall educational scenario is quite depressing as in Purnia and households’ priority for education is generally low, the education of girls gets even lower priority.


It is believed that religion plays a significant role in influencing and determining educational attainment of its followers as different religions provide different belief systems; prescribe different values, norms, attitudes and above all orientation towards life [Glenn and Hyland 1967; Greeley 1976; Muller 1980; Hirschman and Falcon 1980; Caspo 1981; Sanders 1992]. Very often it has been argued that Muslims withhold modern education and prefer Islamic theology and cultural ethos, which places them in the backwaters of educational development in India [Hunter 1869; Baig 1974; Sharma 1978]. However, when all other variables are controlled, the religious affiliation of boys and girls to Hindu or Muslim communities does not make a difference in their chances of getting enrolled. The odds for this are statistically insignificant (Tables 6 and 7). This is true of Patna as well as Purnia and also for boys and for girls, clearly suggesting that religious differences do not impact education levels independent of other socio-economic factors.

V Relative Educational Backwardness of Muslims: A Discussion

It is imperative in the light of the discussion so far that religious differences in educational attainment or educational backwardness among Muslims be placed in a much more complex and nuanced rather than simplistic framework.

To recall our earlier discussion, of the two hypotheses advanced to account for the educational backwardness of Muslims, particularised theology hypothesis deals with the educational backwardness of Muslims in terms of Islamic theology and the cultural ethos of the community. The hypothesised association was problematic even in its logical formulation and when tested against empirical evidence it becomes untenable as logistic regressions show no statistically significant contribution to educational disparities independent of other attributes. Characteristics hypothesis however provides a better anchor to advance our enquiry. As seen, certain socio-economic and demographic characteristics of the population such as age, sex, occupation and education of the head and standard of living of the household do determine educational development. If the occupation of household heads and the standard of living are taken into consideration, it becomes clear that higher educational attainment of a family (as well as a particular group) is associated with occupations such as services along with a high standard of living and regardless of religion. As a corollary to this, it may be proposed that the larger the number of people of a group or sub-group of population in the service sector enjoying a high standard of living, the higher would be the educational attainment in that group. Historically also, the spread of education has always shown such a class character both for Hindus and Muslims and across places as wide-ranging as Europe [Thapar 1966; Silver 1973; Ahmad 1981; Lareau 1987; Gore 1994; Gerwirtz et al 1995; Goldthrope 1996].18

Let us very briefly visit the base of the strata among Muslims vis-à-vis Hindus that usually enters the educational sphere. In our sample, a vast chunk of workers (about 42 per cent) among Muslims were self-employed including in agriculture (selfemployed in occupations other than cultivation 28.7 per cent; cultivators 13.3 per cent). The corresponding figure for Hindus stayed at 33.0 per cent (self-employed other than in cultivation

24.0 per cent and cultivators 9.0 per cent). Further, 70 per cent of Muslim households were landless as against 58.3 per cent Hindu households. Out of those Muslim households who owned land, about 3.4 per cent households had land amounting to 5 acres and above as against 6.4 per cent for Hindus (Table 1). Muslims in services including professional, managerial and executive constituted 13.6 per cent of the example, as against 20.2 per cent among Hindus. Thus, the stratum that in general appropriates the benefits of education has a narrower base amongst the Muslims as compared to the Hindus.

By and large certain socio-economic characteristics exert a profound influence on the literacy and schooling of children and religion remains rather insignificant, substantiating our claim that educational backwardness among Muslims is not due so much to religious orthodoxy as it is because of the specific stratum in which most of them are placed that renders education of little or no relevance. This observation can be applicable to other comparable social groups as well. However, this neat formulation gets teased once locational specificities enter the analytical framework in terms of the “spillover/demonstration” effect of other “class” locations. This is primarily because no community can/would function in contextual isolation. In this context, it is important to note that rather than positing religious groups – Hindus and/or Muslims in an undifferentiated category, we need to ask which strata (of the community) and which locations are being interrogated in terms of religious influences on various attributes rather than advancing meta-narratives or blue-prints of religious essentialism.

There is no doubt that Muslims of India as a whole lag behind other religious communities in the sphere of education. However, the reasons accounting for their educational backwardness are diverse and complex, which remain partially captured in available scholarship. Very often, we are confronted with impressionistic and speculative observations and unsubstantiated formulations, which are by and large one-sided and polemical. Moreover, studies dealing with educational backwardness of Muslim Indians seem to overlook the socio-economic and spatial contexts in which Muslims live. Contextually isolated studies on literacy and education fail to capture their relative standing vis-à-vis other communities whereby educational deficiency among Muslims may, in fact, be an articulation of an overall depressed milieu

– withholding their particular deprivation. Furthermore, within the religious location in a given space there are several fault lines along which communities are divided and Muslims are no exception. Seen in this light, any study intending to assess relative educational backwardness of Muslims cannot really afford to gloss over the fact that they do not constitute a homogeneous category. Otherwise, the analysis would remain partial at best. To conclude then, the complex issue of schooling as a social variable needs framing in multilayered and nuanced ways and has to be placed within a wider conceptual and analytical framework.




1 The National Policy on Education (1986, revised in 1992) emphasised on the removal of disparities and equalising educational opportunity by attending to the specific needs of those who have been denied equality so far. Scores of programmes aimed at addressing the problem of social inequalities in education followed. For example, during the Ninth Five-Year Plan, the programme of Universal Elementary Education was proposed, which aimed at universal access, universal retention, universal achievement of minimum level of learning and implementation of the constitutional right to elementary education by enforcing it through necessary statutory measures. In 1994, the DPEP programme was launched. The goals set by DPEP inter alia included (a) to reduce overall dropout rates for all students to less than 10 per cent; (b) to provide access for children to primary schooling; (c) to reduce differences in enrolment, dropout and learning achievements between gender and social groups to less than 5 per cent. In 2000, the ministry of Human Resource Development planned to launch an ambitious integrated programme called Sarva Shiksha Abhiyan (SSA) to achieve the goal of universalisation of education. The SSA covers the entire population with special focus on the educational needs of girls, scheduled castes and scheduled tribes.

2 According to the 2001 Census, amongst four major religious groups, Christians had the highest literacy rate (the overall literacy being 80.3 per cent) followed by Sikhs (69.4 per cent) and Hindus (65.1 per cent). With the literacy rate of 59.1 per cent, Muslims lagged behind all. In terms of educational levels, the findings of the NSS (55th round) report demonstrate that as the levels of education increase, the representation of Muslims falls sharply.

3 As Shariff and Razzack (2006:105) point out, by and large Muslims prefer the same kind of schooling facilities for their children that are available to other children.

4 As a sample, four villages (two each in Patna and Purnia districts) were selected. The four villages included Nohsa and Bhusola Danapur in Phulwari block of Patna and Kulla and Lakhna in Kasba block of Purnia district.

5 The term ‘Particularised Theology’ is borrowed from studies dealing with religious differentials in fertility [see Sydney 1969].

6 The term “hadith” refers to a tradition of the Prophet, a report about some saying or action of his, which, if recognised as authentic, is considered to be the fundamental source of law. Its authority, however, is subordinate to the injunctions contained in the Quran.

7 Kulkarni (2002) in his study found remarkable variation in the educational status of Muslims. While Muslims are at par with or slightly behind Hindus in the southern-western region, there is a wide gap between the two communities in the northern-eastern states in matters of education. Given these findings if we assume even for the sake of argument that religious orthodoxy has played greater part in holding Muslims back in the sphere of education, we are faced with some very fundamental questions. In the first place if religious orthodoxy can be held responsible for educational deficiency among the Muslims in the northern-eastern states what is it that explains good performance of the Muslims residing in the southernwestern states? If one were to say that the Muslims in the southern-western states might have responded to the processes of socio-economic development as positively as other social communities, by the same token, it can be argued that dismal performance of the Muslim community in the northern and the eastern states which are characterised by lower levels of socioeconomic development in general, except for a couple of states, is rather due to general socio-economic backwardness than because of religious orthodoxy. Secondly, as the statistics reveal [NSS-1999-2000] among those educated, the difference between Hindus and Muslims is quite marginal at least up till middle level education. After this stage however, disparities get widened. Why is there almost equal representation at the lower levels of education and dismal performance at the upper levels? Do Muslims live under the impression that school education at elementary levels is fine but higher education not, as it might take their children away from Islam or Islamic way of life? Or is it because after middle level

Economic and Political Weekly May 5, 2007

Muslim children find it hard to pursue education or have low aspiration for higher education due to the occupation of the household?

8 The characteristics hypothesis has been borrowed from studies on religious differentials in fertility. According to this hypothesis, fertility level of a group or sub group of population is the function of socio-economic characteristics thereof [Sydney G 1969].

9 Measuring disparity between two mutually exclusive sub sets of population has been an overriding problem before social scientists. Sopher (1974) proposed an index to measure disparity, which also tried to overcome the limitations of both absolute and relative measures. This index, however, has some limitations that Kundu and Rao (1985) were successful in overcoming. And we use their index to measure disparity.

10 We have followed the census definition of literacy, a person able to read and write in any language provided he/she was aged seven years or more is considered literate.

11 The overall literacy rate of sample population (56.1 per cent) is higher than that of the state average, which according to 2001 Census was 47.5 per cent. This difference can partly be attributed to the selection of sample villages. For the convenience and purpose of the study we could not select very interior and remote villages where literacy rate might be very low.

12 We have measured disparity by employing the index offered by Kundu and Rao (1985). It reveals that higher the value, larger is the disparity between two sets of population. In the present context, when we talk about gender disparity, the reference category is always boys while in terms of religious groups the reference category is Hindu. When there appears negative value it indicates that the reference category has lower value for which disparity is measured then.

13 It is expressed as: Ea/Pa*100, where Ea= Enrolment at specific age group ‘a’ and Pa= Population at specific age group ‘a’.

14 In Patna there has been a long tradition of education and this is one of the reasons of higher literacy whereas in Purnia entrenched feudalism did not let literacy and education spread across the social spectrum (Sachchidanand 1990).

15 Logistic regression (binary/binomial) is a form of regression, estimating the probability of a certain event to occur. It is used when the dependent variable is dichotomous and the independents are of any type. In brief, it can be used to predict a dependent variable on the basis of independents; to determine the per cent of variance in the dependent variable explained by the independents; to rank the relative importance of independents; to see interaction effects and to understand the impact of covariate control variables. In the present case, dependent variable (dichotomous) is currently enrolled children coded 1 if continuing and 0 otherwise. The exp (β) is the odds ratio representing proportional increase (if greater than 1) or decrease (if less than 1) for the likelihood of continuing education.

16 Although we have not examined this issue, other studies point out that distance to school is one of the reasons for steep decline in the enrolment of girls beyond the primary level. In states like Bihar and Uttar Pradesh middle/secondary schools are located at such a distance that parents choose not to send their daughters to school. In a study of Uttar Pradesh, Dreze and Gazdar (1996) argue that the far-flung location of middle/secondary schools discourages enrolment of girls particularly after primary level, which mitigates the effect of the education level of the head of the households.

17 The standard of living index was derived from the National Family and Health Survey (1999-2000). Scores were given to individual items of household assets. The sum of scores on household assets was classified as high, medium and low standard of living.

18 Education has a functional value and it requires a certain investment; only those who can see some tangible return accruing to them will invest in education. Seen in this perspective not all occupational classes value education equally. Literacy may be widespread and still the middle class tends to place a higher premium on education. Although the occupations that are included in the inventory of middle class vary a great deal in space and time continuum, in general, those who are in white-collar occupations, in big businesses and having large landholdings do constitute the middle class.


Ahmad, Imtiaz (1981): ‘Muslim Educational Backwardness: An Inferential Analysis’, Economic and Political Weekly, Vol 10, pp 1457-65.

– (1987): ‘Educational Development of Minorities’, Journal of Educational

Planning and Administration, Vol 1, (2), pp 191-209.

Ahmad, Imtiaz (1978) (ed): Caste and Social Stratification among Muslims, Manohar, New Delhi.

Alderman, Harold, Jere R Behrman, Shahrukh Khan, David R Ross and Richard Sabot (1997): ‘The Income Gap in Cognitive Skills in Rural Pakistan’, Economic Development and Cultural Change, Vol 46(1): 97-122.

Ali, Ameer (1950): The Ethics of Islam, Calcutta.

Ansari, Mehnaz (2001): ‘Traditional Occupations, Child Labour and Trade Policy: A Vicious Circle around Educational Backwardness of Muslims’, Journal of Educational Planning and Administration, 15(3), pp 333-53.

Archer, M (2003): Structure, Agency and the Internal Conversation, Cambridge University Press, Cambridge.

Arnold, T and A Guillaume (1931): The Legacy of Islam, Oxford University Press, London.

Astone, et al (1991): ‘Family Structure, Parental Practices and High School Completion’, American Sociological Review, Vol 56, pp 309-20. Baig, M R A (1974): Muslim Dilemma in India, Vikash Publications, Delhi.

Balagopalan, S and Subramanian (2003): ‘Dalit and Adivasi Children in Schools: Some Preliminary Research Themes and Findings’, IDS Bulletin, Vol 34, pp 43-54.

Banerjee, Rukmani (2000): ‘Poverty and Primary Schools’, Economic and Political Weekly, Vol 35 (10), pp 799-802.

Burney, Nadeem A and Mohammad Irfan (1991): ‘Parental Characteristics, Supply of Schools, and Child School-Enrolment in Pakistan’, The Pakistan Development Review, 30(1), pp 21-62.

Caspo, M (1981): ‘Religious, Social and Economic Factors Hindering the Education of Girls’, Comparative Education, Vol 17, pp 311-19.

Census of India (2001): Office of Registrar General, New Delhi.

Chatley, Y P (1995): Population, Education and Development, CIRRD, Chandigarh.

Chernichovsky, Dov (1985): ‘Socio-economic and Demographic Aspects of School Enrolment and Attendance in Rural Botswana’, Economic Development and Cultural Change, Vol 33(2), pp 319-32.

Cohen, E (1965): ‘Parental Factors in Educational Mobility’, Sociology of Education, Vol 38, pp 404-25.

Connel, R W (1994): ‘Poverty and Education’, Harvard Educational Review, 64 (2), pp 125-49.

Doughlas, J W G (1964): The Home and the School: A Study of Ability and Attainment in Primary School, MacGibbon, London.

Doughlas, J W G, J M Ross and W R Simpson (1968): All Our Future, Peter Davies, London.

Dreze, J and H Gazdar (1996): ‘Uttar Pradesh: Burden of Inertia’ in J Dreze and A K Sen (eds), Indian Development: Selected Regional Perspective, Oxford University Press, New Delhi.

Duncan, Beverly (1967): ‘Education and Social Background’, American Sociological Review, Vol 72, pp 366-71.

Engineer, Asghar A (2001): ‘Muslim and Education’, Economic and Political Weekly, August 25, pp 3221-23.

Fejgin, Naomi (1982): ‘Factors Contributing to the Academic Excellence of American Jewish and Asian Students’, Sociology of Education, Vol 68, pp 18-30.

Filmer, D and L Pritchett (1999): ‘The Effect of Household Wealth on Educational Attainment: Evidence from 35 Countries’, Population and Development Review, Vol 25, pp 85-120.

Fraser, E (1959): Home Environment and the School, University of London Press, London.

Gerwirtz et al (1995): Markets, Choice and Equity in Education, Open University Press, Buckingham.

Ghosh, Arun (1991): ‘Eighth Plan: Challenges and Opportunities’, Economic and Political Weekly, April 13, pp 941-46.

Glenn, N and R Hyland (1967): ‘Religious Preferences and Worldly Success: Some Evidence from National Surveys’, American Sociological Review, Vol 74, pp 74-85.

Goldthrope (1996): ‘Class Analysis and the Reorientation of Class Theory: The Case of Persisting Differentials in Educational Attainment’, British Journal of Sociology, 47 (3), pp 481-503.

Gore, M S (1994): Indian Education: Structure and Process, Rawat, New Delhi.

Greeley, A M (1976): Ethnicity, Denomination and Inequality, Sage, Beverley.

Guha Roy, Samir, Subir K Mitra and Surja Sankar Ray (1995): Achievement Level of Primary School Children at the End of Class IV, Indian Statistical Institute, Calcutta and State Council of Educational Research and Training, West Bengal.

Haveman, Robert, Barbara Wolfe and James Spaulding (1991): ‘Childhood Events and Circumstances Influencing High School Completion’, Demography, Vol 28(1), pp 133-57.

Haveman, R and B Wolfe (1995): ‘The Determinants of Children’s Educational Attainment: A Review of Methods and Finding’, Journal of Economic Literature, Vol 33, pp 1829-78.

Hasan, Zoya and Ritu Menon (2004): Unequal Citizens: A Study of Muslim Women in India, Oxford University Press, Delhi

Hill, M Anne and June O’Neill (1994): ‘Family Endowments and the Achievement of Young Children with Special Reference to the Underclass’, Journal of Human Resources, Vol 39(4), pp 1064-1100.

Hirschman and L M Falcon (1988): ‘The Educational Attainment of Religioethnic Groups in the United States’, Research in Sociology of Education and Socialisation, Vol 5, pp 83-120.

Hossain, Shaikh I (1990): ‘Interrelations between Child Education, Health and Family Size: Evidence from a Developing Country’, Economic Development and Cultural Change, Vol 38(4), pp 763-81.

Hunter, W W (1869): Indian Musalmans, Indological Book, New Delhi.

Hussain, Zakir (2005): ‘Analysing Demand for Primary Education: Muslim Slum Dwellers of Kolkata’, Economic and Political Weekly, Vol 40 (2), pp 137-47.

Jabbi, M K and C Rajyalakshmi (2001): ‘Access to Education of Marginalised Social Groups in Bihar’ in A Vaidyanathan and P R Gopinathan Nair (eds), Elementary Education in Rural India: A Grass Roots View, Strategies in Human Development in India-Vol II, Sage Publications, New Delhi, pp 395-458.

Jain, S P (1969): ‘Religion, Caste, Class and Parental Educational Aspiration in a North Indian Community’, Indian Educational Review, Vol 4, pp 64-72.

Jain, Sushila (1986): Muslims and Modernisation, Rawat Publications, Jaipur.

Jeffery, P and R Jeffery (2000): ‘Religion and Fertility in India’, Economic and Political Weekly, Vol 26, pp 3253-59.

Kamat, A R (1989): Studies in Educational Reforms in India, Himalaya Publishing House, Bombay.

Kanbargi, R and P M Kulkarni (1991): ‘Child Work, Schooling and Fertility in Rural Karnataka’ in Kanbargi (ed), Child Labour in the Indian Subcontinent: Dimensions and Implications, Sage, New Delhi.

Kareem, P A (1989): ‘Developmental Implications of Educational Backwardness: A Study with Special Reference to Indian Muslims’, Perspective in Education, Vol 5, pp 169-83.

Khalidi, Omar (1995): Indian Muslim since Independence, Vikas Publishing House, New Delhi.

Knodel, John, Havanon, Napaporn and Sittirai, Werasit (1990): ‘Family Size and the Education of Children in the Context of Rapid Fertility Decline’, Population and Development Review, Vol 26(1), pp 31-62.

Kulkarni, P M (2002): Inter-State Variations in Human Development Differentials among Social Groups in India, Working Paper Series No 80, National Council for Applied Economic Research, New Delhi.

Kundu, A and J M Rao (1985): ‘Inequity in Educational Development in India: Issues in Measurement, Changing Structure and Its Socio-economic Correlates with Special Reference to India’ in Moonis Raza (ed), Educational Planning – A Long Term Perspective, NIEPA, New Delhi.

Lareau, A (1987): ‘Social Class Differences in Family-School Relationship: The Importance of Cultural Capital’, Sociology of Education, Vol 60 (73), pp 73-85.

Learner, D (1964): The Passing of Traditional Society – Modernising the Middle East, The Free Press, Glencoe, IL.

Levy, R (1962): The Social Structure of Islam, Cambridge University Press, London.

Lieten, G K (2000): ‘Children, Work and Education’, Economic and Political Weekly, Vol 35 (2), pp 2171-78.

Mayer, A and H Sharp (1962): ‘Religious Preferences and Worldly Success’, American Sociological Review, Vol 27, pp 218-27.

Mehrotra, Nidhi (1995): ‘Why Poor Children Do Not Attend School: The Case of Rural India’, Department of Education, University of Chicago.

Mondal, Seikh Rahim (1997): Educational Status of Muslims: Problems and Prospects, Inter India Publications, New Delhi.

Morrish, I (1972): The Sociology of Education, George Allen and Unwin, London.

Muller, C W (1980): ‘Evidence on the Relationship between Religion and Educational Attainment’, Sociology of Education, Vol 53, pp 140-52.

Muller, W (1972): ‘Family Background, Education and Career Mobility’, Social Science Information, Vol 11, pp 223-55.

National Family Health Survey (1998-99): International Institute Population Sciences, Mumbai.

National Policy on Education (1986, 1992): Government of India, Ministry of Human Resource Development.

NSSO (1990): Servekshana: Results of the Fourth Quinquennial Survey on Employment and Unemployment, NSSO, Government of India.

– (2001): Employment and Unemployment Situation among Religious Groups in India, 1999-2000, 55th round, Government of India. Nuna, S C (1993) (ed): Regional Disparities in Educational Development, South Asian Press, New Delhi. Panchmukhi, P R (1990): ‘Private Expenditure on Education in India: An Empirical Study’ (mimeo), Indian Institute of Education, Pune. Parish, William L and Robert J Willis (1994): ‘Daughters, Education and Family Budgets’, The Journal of Human Resources, Vol 28(4), pp 863-97. Peacock, J L (1978): Muslim Puritans – Reformist Psychology in South Asian Islam, University of California Press, Berkeley. Pearse, R (1985): ‘Factors Related to Inequality in Participation in Schooling in Java’ International journal of Educational Development, Vol 5, pp 11-26. Peer, M (1990): ‘Muslim Approach to Education: An Overview’, Islamic Culture, Vol 64, pp 163-77. Rafeda, al Hariri (1987): ‘Islam’s Point of View on Women’s Education in Saudi Arabia’, Comparative Education, Vol 23, pp 51-57.

Raju, Saraswati (1991): ‘Gender and Deprivation: A Theme Revisited with a Geographical Perspective’, Economic and Political Weekly, Vol 26, pp 2827-39.

Raza, Moonis (1990): Education, Development and Society, Vikas, New Delhi.

Rose, G (1968): The Working Class, Longmans, London.

Sachchidanand (1990): ‘Disparity in Elementary Education: A Case Study of Bihar’ in Panchmukhi (ed), Reforms in Indian Education, Himalaya Publishing, Bombay.

Salamatullaha (1994): Education of Muslims in Secular India, CRRID, Chandigarh.

Sanders, W (1992): ‘The Effects of Ethnicity and Religion on Educational Attainment’, Economics of Education Review, Vol 11, pp 119-35.

Shariff, Abusaleh (1999): India Human Development Report, NCAER, New Delhi.

Shariff Abusaleh and Azra Razzack (2006): ‘Communal Relations and Social Integration’ in India: Social Development Report, Oxford University Press, New Delhi.

Sharma, K D (1978): Education of a National Minority: A Case of Indian Muslims, Kalamkar Prakashan, New Delhi. Silver, H (ed) (1973): Equal Opportunity in Education: A Reader in Social Class and Educational Opportunity, Methuen, London. Singh, J P (2005): ‘Universalisation of Education in India: Promise and Performance’, Man and Development, Vol 27, 1, pp 103-24. Sopher, D E (1974): ‘A Measure of Disparity’, Professional Geographer, Vol 26, pp 389-92.

– (1980): ‘Sex Disparity in Indian Literacy’ in Sopher (ed), An Exploration of India: Geographical Perspective on Society and Culture, Cornell University Press, New York.

Sydney, G (1969): ‘Socio-economic Differentials among Religious Groups in United States’, American Journal of Sociological Review, Vol 74, pp 612-31.

Tansel, Aysit (1997): ‘Schooling Attainment, Parental Education and Gender in Cotì d’Ivoire and Ghana’, Economic Development and Cultural Change, Vol 45(4), pp 825-56.

Thapar, Romila (1966): A History of India, Penguin, Baltimore.

Tilak, J B G (1995): ‘How Free Is Free Primary Education in India?’ Occasional Paper 21, National Institute of Educational Planning and Administration, New Delhi.

  • (2004): ‘Public Subsidies in Education in India’, Economic and Political Weekly, Vol 39 (4), pp 343-59.
  • (2006): ‘Education: A Saga of Spectacular Achievement and Failure in India’ in India: Social Development Report, Oxford University Press, New Delhi.
  • Versoff, J et al (1962): ‘Achievement, Motivation and Religious Background’, American Sociological Review, Vol 27, pp 205-18.

    Warren, John Robert (1996): ‘Educational Inequality among White and Mexican Origin Adolescents in the American South-West: 1990’,Sociology of Education, Vol 69, pp 142-58.

    White, K R (1982): ‘The Relationship between Socio-economic Status and Academic Success’, Psychological Bulletin, Vol 91, pp 461-81.

    Wolfe, Barbara L and Jere R Behrman (1984): ‘Who Is Schooled in Developing Countries? The Roles of Income, Parental Schooling, Sex, Residence and Family Size’, Economics of Education Review, Vol 3(3), pp 231-45.

    Economic and Political Weekly May 5, 2007

    To read the full text Login

    Get instant access

    New 3 Month Subscription
    to Digital Archives at

    ₹826for India

    $50for overseas users


    (-) Hide

    EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

    Back to Top