Women’s Autonomy, Women’s Status and Nutrition in India
123Sandip Chakraborty, Kaushlendra Kumar and Faujdar Ram
The constitution of India makes no distinction between the sexes. But this politically granted equality has not been very evident in practice and the social and economic status of women has not been on par with that of men. There is a need to study the differences in status of men and women in India, and, the changes that have occurred in these differences over time.
Several indicators like expectation of life at birth, adulthood literacy, workers in the modern sector, singulate mean age at marriage etc. were selected to study the relative status of men and women at several points of time, at a macro level. The data for the present study have been taken from different Censuses (particularly from 1971 to 2001) and Sample Registration System. Taxanomic method is used to classify and compare the status of men and women in the major states of India for the several periods. Cluster analysis has been performed to show the homogeneity among the several states in India in terms of status of men and women both.
The result shows that the gap between status of men and women are closer over the period and particularly in the period between 1991 and 2001. The status of men and women are high in the states like Kerala, Punjab etc. Though there is no difference in the clustering of states in terms of status for the period 1971 and 1981, but for the period 1991 and 2001, position of some states have changed.
1 Research Scholar, International Institute for Population Sciences, Mumbai-400 088, email: firstname.lastname@example.org
2 MPS Scholar, International Institute for Population Sciences, Mumbai-400 088, email: email@example.com
3 Prof. and Head, Department of Fertility Studies, International Institute for Population Sciences, Mumbai-400 088, email: firstname.lastname@example.org
Measuring women‟s status and autonomy can be problematic. Women‟s status has traditionally been measured using education and employment status variables. In a study of female autonomy in India, Dyson and Moore (1983) stated that autonomy represents the „capacity to manipulate one‟s personal environment,‟ and that „equality of autonomy between the sexes…implies equal decision-making ability with regard to personal affairs.‟ Autonomy
has thus increasingly been defined as a woman‟s „ability or lack thereof to make decisions in
the household‟ (Hindin, 2000b). Higher levels of women‟s autonomy, though context-
specific and therefore measured slightly differently in different studies, have been associated with nutritional status (Hindin, 2000a), maternal health care utilization (Beegle, Frankenberg & Thomas, 2001; Bloom, Wypij, Das Gupta, 2001), and fertility behaviors and contraceptive use (Balk, 1994; Hindin, 2000b; Govindasamy & Malhotra, 1996; Al Riyami, Afifi & Mabri, 2004; Moursund & Kravdal , 2003), lower rates of child mortality (Castle, 1993). Malhotra et al. (2002) provide an overview of women‟s status, empowerment, and decision-making
autonomy, and a review of the literature linking these variables to health outcomes.
In developing countries females are in disadvantageous position with regard to health and well being (Santow, 1995). The cultures of South Asia are largely gender stratified, characterized by patrilineal descent, patrilocal residence, inheritence and succession practices that exclude women, and hierarchical relations in which the patriarch or his relatives have authority over family members (Jejeebhoy and Sathar, 2001). Patriarchal kinship and economic systems limit women‟s autonomy and as a result the health status of both women
and children, particularly female children, suffers in relation to that of males (Caldwell, 1986).
Autonomy and Nutrition:
Although women have tended to be producers for the family in many agricultural settings, their lack of access to the income from this labour leaves them resource-poor (Abbas, 1997). There has been some evidence to suggest that women who have lower levels of autonomy and status within in the household are more likely to experience under nutrition (Hindin, 2000) or have a lower BMI (Bindon & Vitzthum, 2002; Baqui et.al, 1994).
Aim of the Study:
The purpose of the study is to explore the extent of women‟s autonomy and its relationship with the nutritional status of the women in India. The core hypothesis behind the paper is that, women with low autonomy and status will be less likely to obtain adequate food resources and will be more likely to experience under nutrition or Chronic Energy Deficiency (CED).
The nationwide data from India‟s National Family Health Survey (NFHS-2)
conducted during 1998-99 was used for this study. This survey covered a representative sample of 90,303 ever married women in the age group of 15-49 years, from 27 states of India. The sample comprised more than 99 percent of India‟s population (IIPS, 2000). The
survey used uniform questionnaires, sample designs, and field procedures to facilitate comparability of the data within the country, so as to achieve a high level of data quality.
For the purposes of the study, the sample was limited to non-pregnant married women who had not given birth in the last three months. These constraints led to a sample of 74, 391 women in India.
Measure of Nutrional Status:
National Family Health Survey provides the information on height and weight of the
woman. Based on these two information, Body Mass Index (BMI) is calculated to the
restricted population. Lastly, a dichotomous measure, Chronic Energy Deficiency (CED), 2based on the standard BMI cutoff of<18.5 kg/m was generated. This measure was used as a
nutritional status of the individual.
Measures of Sociodemographic characteristics, Women’s and Partners’ characteristics:
The Sociodemographic characteristics of the sample are divided into two groups: household level characteristics and women characteristics. The household level characteristics include: i) Residence ii) Caste iii) Religion iv) Standard of living v) Size of the household and vi) Husband living in the household.
The women characteristics included in the analysis: i) Age ii) Number of births iii)
Education and iv) Occupation. Education (for both the respondent and her partner/husband) was divided into four categories viz. Illiterate, Up to Primary, Up to secondary and higher.
Occupation (for both the respondent and her partner/husband) was coded into five categories viz. Unemployed or not-working, working in agricultural, unskilled/skilled manual, non-manual and professional.
Partners‟ characteristics include: i) Education and ii) Occupation.
Measures of Women’s relative status, Women’s status in society and Decision-making Autonomy:
Women’s relative status: Women‟s relative status is conceptualized as their status relative to their partner‟s status in terms of age, education and occupation. For age, three categories
were used based on the continuous measures of age:
a) respondents were four or more years older than their partners
b) respondents were six or more years younger than their partners
c) everyone else who was near the same age as their partners Relative educational status was calculated as a difference between the partners‟ schooling levels with three categories:
a) respondent has more
b) the couple has same level
c) the partner has more
A relative occupational difference was calculated using the five occupational levels with three categories:
a) respondent has more
b) the couple has same level
c) the partner has more
Women’s status in society: In National Family Health Survey, women were asked about their attitudes toward wife beating. The women were asked to give their opinion about the justification of wife beating by their husband in the following situations:
a) if she is unfaithful
b) if her family does not give money
c) if she shows disrespects
d) if she goes out without permission
e) if she neglects children and house
f) if she does not cook properly
From these dichotomus variables (yes/no), an index was created based on whether women think it is justified for a husband to beat his wife, under any of the circumstances. This variable is used as a proxy to measure women‟s status.
Measures of decision making: In the National Family Health Survey, women were asked
about the person who has the final say or she has to need any permission over the following aspects:
a) Final say over what to cook
b) Final say over health care
c) Final say about purchase jewelry
d) Final say about staying with family
e) Permission to visits relatives and friends house
f) Permission to go to market
g) Allowed to have money set aside
For each of these questions, the women were given the following response options:
c) respondent and husband/partner jointly
d) someone else
e) respondent and someone else jointly
A set of dichotomous variables was created for each of the decision making dimension to reflect patterns of decision making. For each domain, the variable was coded as 1 if the women had the final say over that decision alone and 0 if the women did not have the final say. An index of autonomy was constructed on the basis of the decision taken by the women alone. Index of autonomy is a simple measure by summing up all the domains where the women (alone) had the final say. This index is categorized into three categories:
a) low (if the index lies between 0-2)
b) medium (if the index lies between 3-4)
c) high (if the index value is 5 or more)
Statistical analysis: Univariate and Bivariate analyses were used to study the level of autonomy, women‟s status and nutritional status of women. Multivariate analyses were used to explore the determinants of Chronic Energy Deficiency (CED) and the level of autonomy. The multivariate analyses were done by Binary and Multinomial logistic regression analysis.
Binary Logistic Regression: The basic form of the logistic regression is
Where is constant, are the coefficients of bb,b,............,bx,x...........x12k12k0
is the estimated probability of having Chronic Energy Deficiency (CED) p
Multinomial Logistic Regression: In this model the response variable is mutually
exclusive and exhaustive. The Multinomial Logit model can be given as:
p，(1 = +, i= 1, 2, 3……………n; balogx！；？1i1ip3，，
p，(2 = +, i= 1, 2, 3……………n; bxloga！；？2ii2p3，，
p + p + p = 1. Where a and a are constants and b, b are the coefficients 123121i2i
of x‟s. i
For example in the present analysis, p is the estimated probability of medium level of 1
autonomy p denotes the estimated probability of high level of autonomy and p is the probability 23
of low level of autonomy. Here p is the reference category. 3
Table 1 represents the prevalence of Chronic Energy Deficiency (CED), Sociodemographic
characteristics, Husbands characteristics, Women’s autonomy in decision making and Womens status in society. In India the percentage of women with Chronic Energy Deficiency (CED) is about 32
percent. Most of the women are rural resident (68 percent), Hindu (78 percent), general category (42 percent), having at least one child (91 percent), from large family (55 percent), representing medium standard of living (48 percent), higher age group (79 percent), illiterate (49 percent) and unemployed (63 percent). At the time of survey only 5 percent of women reported that their husband were not living with them, their husband‟s were having at least secondary level of education (55 percent) and they were from agricultural and unskilled/skilled work group (66 percent).
In terms of women‟s relative status fewer women (less than 1 percent) were older than their wives, 47 percent of the husbands were older than their wives and 53 percent of the couples were in same age. In India, women and their partner attain same level of education or men have more education than their partners. Only 6 percent of women have higher status job than their partners.
In India, most of the women have low status of autonomy (58 percent). Women from Western and Southern part have substantially more autonomy than the other parts of the country. In different domains of decision making, only in what to cook, women have more final say. Even in health care she has depend on others (mostly husband) decision. Most of the women need permission to go elsewhere. In terms of women‟s status in society
54 percent of women in India believe it is OK for husbands to beat their wives in at least one of the six domains posed in the questionnaire. Interestingly it was found that even in Western and Southern part where the autonomy is high as compared to rest of the country, most of the women (55 percent and 66 percent) have the believe about the justification of wife beating by their husbands.
Level of autonomy by different background characteristics:
From table 2 it is clear that woman who belongs to rural area, low standard of living, childless, residing in large family and younger in age have less autonomy in decision making. They are unable to take decisions (alone) if husbands live with them. Education and occupation have positive relationship with the level of autonomy, as the level of education and occupation increases the level of autonomy also increases. If the woman has higher level of education and higher status of job as compared to her husband, she has more autonomy in decision making.
Factors associated with women’s autonomy in decision making:
To find out the influential factors on autonomy, the multinomial logistic regression has been applied by taking low autonomy as reference category. The coefficient under
p，(1 represents the effects of predictor variables on medium level of autonomy over log！；p3，，
p，(2low autonomy and represents the effects of predictor variables on high log！；p3，，
autonomy over low autonomy. The results in table 3 show that when the womens education level and occupation level are increasing the relative risk ratio of having high autonomy is also increasing, particularly for those women who are in non-manual, professional and higher level of education as compared to unemployed and illiterates. Women who are in professional jobs , their odd values is almost 2.5 times more as compared to unemployed and who are possessing higher level of education, their odd values is 1.7 times more as compared to illiterate. With respect to medium level of autonomy the corresponding ratio is about 1.8 times more as compared to unemployed and 1.5 times as compared to illiterate.
The standard of living index of the household does not play any significant role in autonomy whereas place of residence has some significant effect on autonomy, as the results show.
Compared to Hindus, the ratios of high autonomy over no autonomy for Muslims is reduced to 78 percent but a significant increase is noticed among others (69 percent) in the same. The ratios of medium level of autonomy over no autonomy for Muslims is reduced to 75 percent as compared to Hindus.
As compared to General caste, Schedule Castes, Schedule Tribes and Other Backward Castes category women are more likely to have high autonomous status over no autonomus status.
Other variables like household size, husbands living in household, educational and occupational level difference also play significant roles in women‟s autonomy.
There exists a significant age group and regional differentials in high autonomy over no autonomy. Women from higher age groups have the high autonomy as compared to 15-19 age groups. Women who are in 40-44 and 45-49 age group , their odd values are almost 10 and 11 times more as compared to 15-19. Western and Southern Zone women have high autonomy as compared to North. Their odd values are 3.1 and 2.5 times more than the reference zone (North). The others zone have less autonomy than the North.
Estimated probabilities for categories in Autonomy through Multiple Classification Analysis (MCA) table
Estimated probabilities of Low Autonomy (p), Medium Autonomy (p) and High 12
Autonomy (p) based on elicited parametric estimates of the multinomial logit models are 3
shown in table 4. The adjusted values are based on the complete model including all predictor variables simultaneously.
Education level of the mother a significant role in the high autonomy of women. In India, the high level of autonomy has increased from 11 percent among those women who are illiterate to 15 percent among those who have the higher level of education, when the other variables are controlled.
Interestingly, the religious differentials in high autonomy is negligible among Hindus and Muslims. Women belongs to general category are less likely to be highly autonomus.
From the table it is clear that place of residence play an important role in the level of autonomy. Women from the urban areas are more autonomus than the women live in villages
As the family size increases the level of autonomy decreases. From the MCA table (table 4) one can say that standard of living, educational level difference and occupational level difference may not have that much of effect on women‟s autonomy.
There exists a huge difference in the magnitude of high autonomy depending upon the partner‟s presence in the household. The magnitude increases from 9 percent among those whose husband stays in the household to 44 percent among those whose husbands stay elsewhere.
Among the different age groups the women belong to last age group i.e. 45-49 have the high level of autonomy and the percent of high autonomy increases as the age increases.
If we compare the regions, the level of autonomy differs from one region to another. The women from West and South region are in advantageous position in terms of autonomy than the others. The women belong to Eastern and Central zone have the lower level of autonomy.
Chronic Energy Deficiency (CED) with different background characteristics:
Sociodemographic, Women’s and Partner’s Characteristics:
In table 5, the unadjusted associations with CED are explore using cross tabulation and chi-squared tests. In India more rural women have CED than urban women. Fewer literate women having CED than the illiterate. Women in agricultural work have CED than women in other occupations or unemployed women. More women with partners have less education had more CED and fewer women with partners employed in occupations others than agriculture has CED.
Women’s Relative Status, Women’s status in Society and Decision Making Autonomy:
CED is high among those women who are older than their husbands and less when the age is same, though the magnitude is less. If the women are in higher status of job and higher level of education as compared to their husband, they are in advantageous position i.e. the CED among those women is low. Decision making autonomy is also associated with the CED as the level of autonomy increases the level of CED decreases. It is clear from the table that all individual measures of wome‟s attitudes towards wife beating are associated
with CED in India.
Factors associated with CED:
To identify the factors associated with the CED, three models were run for CED using logistic regression. Where the first model includes only Sociodemographic and woman‟s characteristics, the second model adds couple characteristics and women‟s relative status in society, the third model adds women‟s status in society and women‟s decision making autonomy. Since the variables did not substantially change in the presence of others, the final model with all the variables is presented for India and regions.
From the model it is clear that the women from rural area are more likely to have CED than the women in urban area and as the level of SLI increases CED decreases and both the results are statistically significant. Age is associated with CED. In India more younger women have CED. The schedule castes women are more likely to have CED than the others. CED is less among the Muslims and the others as compared to the Hindus. Higher levels of education and occupation are associated with lower rate of CED. Among
the relative status variables only education level variables and occupational level variables are considered. It is found that the women who have higher level of education than their husband are less likely to have CED. The relationship between occupational differences is in favour of women but this is not statistically significant. In terms of women‟s status in society, women who feel that wife beating is justified in more domains, are less likely to have CED though this result does not attain significance at the p<0.05 level. The level of autonomy in decision making has an positive impact on CED. A woman who takes more decisions is less likely to have CED.
Despite different policies regarding women‟s empowerment in India, women are still behind in terms of autonomy in decision making and in nutritional status. Almost in every domain of decision making they have to depend on others (husband and in laws) in the family, even in the sphere of health care also. She has to take permission or depend on decision about financial matters in the household. For example in most of the household in India if a woman feels seek she has to depend upon her husband or others in laws house for buying medicines and other necessary things to recovery soon. In a patriarchal society like India, women take their regular meals at the end. Most of the time they take under nutritious food like sukhi roti or only chawal because of insufficient amount of foods. Since they have
financial restrictions and limitations to go outside, most of the time they can not buy whatever she wants to consume to keep her healthy. This may pilot many of the womens into the dearth of physical energy.
From the analysis it is clear that level of education and occupation plays a vital role in womens autonomy as well as in nutritional status. If the education level is higher a woman can know about her nutritional rights. She is aware about her physical needs. She explores herself through the increasing level of knowledge and as a result she tries to put his decision in the household. If a woman is in professional jobs they have more decision making power in the family as compared to the unemployed. She need not take permission to go outside or to buy anything.
Urban women are in advantageous position in terms of autonomy and nutritional status also. It is well known fact that urban women are much more advanced in terms of education, employment and in other spheres of development as compared to rural women. More education and higher level of employment means higher probability of autonomy and as a result less CED.
If the husband stay elsewhere than woman alone has to take most of the decisions about the family matters. She has to take decisions to care of all the members of the household.
In general women become more autonomus as they aged. They can rely on the relationship with their husbands, older children and friends – as direct sources of power and
security in the household. At the beginning of married life, however, women need the external support of natal kin in order to realize their needs and desires (Bloom et.al, 2001).
This may be a possible reason of less CED among the aged women.
In terms of women‟s status in society, more women who feel that wife-beating is
justified in more domains are less likely to have CED. This surprising result may have to do how widely accepted wife beating in India.
Our findings suggest that there are several factors which affect the level of autonomy as well as the nutritional level of women. Now the question is besides these factors what are other factors that may affect the level of autonomy and as a result the nutritional status of the women. For example education and employment do not necessarily enhance women‟
autonomy and some traditional factors conferring status on women remain strong (Jejeebhoy and Sathar, 2001). So, the strategies to enhance women‟s autonomy need to expand beyond education and employment. The issues like women‟s gender consciousness,
enabling women to mobilize and access community resources and public services, providing support for challenging traditional norms, enhancing women‟s access to and control over economic resources, issues relating to women‟s right should be emphasized.
Abbas, J.D. 1997. Gender asymmetries in intrahousehold resource allocation in Sub Saharan Africa: Some policy implications for land and labour productivity. In L.Haddad, J.Hoddinott and H. Alderman (eds), Intrahousehold resource allocation in developing countries: Models, methods and policy (249-262). Baltimore: The Johns Hopkins University Press.
Al Riyami,A., Afifi,M., & Mabry,R.M. (2004). “Women's autonomy, education and employment in Oman and their influence on contraceptive use.” Reproductive Health Matters,
Balik, D. 1994. “ Individual and community aspects of women‟s status and fertility in rural Bangladesh.” Population Studies, 48, 21-45.
Beegle,K., Frankenberg,E., & Thomas,D. 2001. “Bargaining power within couples and use
of prenatal and delivery care in Indonesia.” Studies in Family Planning, 32(2), 130-146.
Bindon, J. R. and Vitzthum, V.J. 2002. “Household economic strategies and nutritional anthropometry of women in American Samoa and highland Bolivia.” Social Science & Medicine,
Bloom,SS., Wypij,D., & Das Gupta,M. (2001). “Dimensions of women's autonomy and the influence on maternal health care utilization in a North Indian city.” Demography, 38(1), 67-78.
Caldwell, J.C. 1986. “Routes of low mortality in poor countries”. Population and Development
Review, 12, 171-220.
Castle, S.E. 1993. “Intra household differentials in women‟s status: Household function and focus as determinants of childrens illness management and care in rural Mali.” Health
Transition Review, 3, 137-157.
Dyson,T., & Moore,M. 1983. “On kinship structure, female autonomy, and demographic behavior in India”. Population and Development Review, 9(1), 35-60.
Govindasamy,P., & Malhotra,A. (1996). “Women's position and family planning in Egypt.”
Studies in Family Planning, 27(6), 328-340.
Hindin,M., & Adair,L. 2002. “Who's at risk? Factors associated with intimate partner violence in the Philippines.” Social Science & Medicine, 55(8), 1385-1399.
Hindin,M.J. 2000a. “Women's power and anthropometric status in Zimbabwe.” Social Science
& Medicine, 51(10), 1517-1528.
Hindin,M.J. 2000b. “Women's autonomy, women's status and fertility-related behavior in
Zimbabwe.” Population Research and Policy Review ,19(3), 255-282.
International Institute for Population Sciences. 2001. National Family Health Survey-2, 1998-99, IIPS, Mumbai.
Jejeebhoy, S. and Sathar, Z. 2001. “ Womens autonomy in India and Pakistan : The influence of religion and region.” Population and Development Review, 27(4), 687-712.
Malhotra,A., Schuler,S.R., & Boender,C. Measuring Women's Empowerment as a Variable in International Development. 2002. Washington, D.C., The World Bank, Gender and Development Group and Social Development Group.
Moursund,A., & Kravdal,O. 200). “Individual and community effects of women's education and autonomy on contraceptive use in India.” Population Studies, 57(3), 285-301.