Version classiqueVersion mobile

Gender discriminations among young children in Asia

 | 
Isabelle Attané
, 
Jacques Véron

Part III - Gender discriminations among young children in Korea, Taiwan and Pakistan

11. Son Preference in Pakistan: Its Effects on Sex Ratio, Preferential Treatment of Boys and Sex Differentials in Infant Mortality

Mehtab S. Karim

Résumé

Au Pakistan comme dans de nombreux autres pays en Asie, il existe traditionnellement une forte préférence pour les fils. En utilisant les données du recensement de 1998 et d’enquêtes conduites au cours des dix dernières années, l’auteur étudie les tendances récentes du rapport de masculinité à la naissance et chez les enfants de moins de cinq ans, de même que les différentiels de mortalité infantile et juvénile selon le sexe. Il examine également la possibilité de traitements préférentiels selon le sexe, pouvant favoriser les garçons.
Etant donné que la fécondité est encore élevée au Pakistan, les avortements sélectifs selon le sexe sont encore très peu pratiqués. Les couples ne renoncent pas pour autant à un fils, mais acceptent de faire plusieurs tentatives, c’est à dire d’avoir plusieurs filles, avant d’y parvenir. Par conséquent, le rapport de masculinité à la naissance au Pakistan est conforme à la norme biologique. L’auteur démontre également que, en dépit d’une forte préférence pour les fils, les naissances de filles ne sont pas malvenues, et que les filles ne sont pas particulièrement discriminées après leur naissance. Il constate cependant que, ces dernières années, la mortalité infantile des garçons a diminué plus vite que celle des filles.

Texte intégral

Introduction

1In most countries in East, South and West Asia a strong desire among parents for a male against a female child is reported (Arnold 1992). In most cases, sons are desired for social reasons (such as carrying the family name) and economic reasons (for old age security). Table 1 suggests that among parents in all the countries of South Asia, the next child to be a son is prevalent, which increases when there are more daughters in the family than sons. In Pakistan such desire is more prominent as compared to other countries in the region.

Table 1: Percentage of currently married, non-pregnant women 15-49 who want another child, by number and sex composition of living children, South Asian countries

Table 1: Percentage of currently married, non-pregnant women 15-49 who want another child, by number and sex composition of living children, South Asian countries

Source: Arnold (1997).

2With this background, utilizing data from the 1998 population census and sample surveys conducted during the past decade, this paper examines: evidence of sex preference in Pakistan; sex ratios at birth; sex ratios among children under five; sex differentials in mortality at infancy and in early childhood; and, whether any preferential treatment is given to male (over female) children.

Evidence of Son Preference

3At the dawn of Islam, female infanticide was widely practised in the Arabian Peninsula; therefore a special verse was revealed in the Qur’an that forbids the killing of children. Followers of Islam were reminded by the Prophet (as reported in Hadith, sayings of the Prophet) that daughters are as revered as sons. Although Muslims constitute about 97 per cent of Pakistan’s population, sons are much preferred over daughters. Earlier studies conducted in Pakistan (Khan and Sirageldin 1987, Ali 1989, Sathar and Karim 1996) have shown not only preference for sons, but have also demonstrated its effects on the desire for additional children. Sotoudeh-zand (1988) contends that Pakistani parents are reported to attach great value to sons as economic assets and old age security. Similarly, Farooqui (1990) reports a strong preference for sons over daughters, whereby 87 per cent of the women interviewed considered three as the ideal number of children, showing a desire for two sons and one daughter.

4As shown in Table 2, son preference in Pakistan is, however, somewhat declining, although it has remained high during much of the past decade. For example, among women with no children, about one-third desired the first child to be a male. While none in 1990-91 wanted this child to be a girl, in 2000-01 only 2 per cent desired a female child. Further, among women who had two children both of whom were girls, 93 and 87 per cent in the respective surveys wanted a male child, while none in 1990-91 and only one per cent in 2000-01 wanted another female child. Among those who had one son and one daughter, still 79 per cent in both the surveys wanted another son and none wanted a daughter. Preference for the next child to be a male is nearly universal when the number of children is three or more and among those who have no sons. On the other hand, even among those who have two living sons, the desire for a daughter is still not that prominent.

Table 2: Per cent distribution of currently married women wanting another child, by their preferred sex of the next child by number of living children and living sons

Table 2: Per cent distribution of currently married women wanting another child, by their preferred sex of the next child by number of living children and living sons

* Less than 25 observations.
Sources: 1990-91 Pakistan Demographic and Health Survey, 1990-91.
2000-01 Pakistan Reproductive Health and Family Planning Survey 2000-01.

Sex Ratio at Birth

5The sex ratio at birth could differ from the accepted biological sex ratio at birth-normally 105 males for 100 females-for three reasons. If the ratio is higher, then it could be due to (a) abortion of female foetuses, (b) intentional or unintentional infanticide of girls immediately after birth or (c) systematic under-reporting of female children. Since Pakistan has yet to reach the stage of the two-child family, the abortion of female foetus is rarely practised. Instead, those couples that desire to have male children do so by having successive female children until they achieve their desired goal. Even though ultrasound has become an important tool to predict the sex of the foetus in large cities, in a personal communication with the secretary general of the society of ultrasound practitioners, it was pointed out that whatever abortions take place in Pakistan, it is irrespective of the sex of the child to be born. Consequently, the sex ratio at birth of 107 during the past five years as reported in the Pakistan Demographic Survey (Federal Bureau of Statistics 2001) is consistent with the accepted biological norm.

Table 3: Sex ratios at birth by birth order (males per 100 females)

Table 3: Sex ratios at birth by birth order (males per 100 females)

Source: Pakistan Integrated Household Surveys.

6Table 3, presents sex ratios at birth for various birth orders and suggests that, while the overall sex ratio at birth has increased slightly from 107 in 1995-96 to 109 in 2000-01 (only in rural areas), for the first birth order the ratios were fairly high during both the periods, especially in rural areas. Is the high sex ratio at the first birth order in Pakistan “real” or a product of reporting bias? This may be possible in view of the strong desire for male children–particularly in the rural setting-, when, as a reflection of the desire that the first born be a boy, the first born daughters are omitted and reported as younger than the later-born male sibling.

7Arnold (1992), found high sex ratios for the first birth order in Indonesia, Sri Lanka, Thailand, Egypt, Brazil, Ecuador and Trinidad and Tobago. He therefore contends that, “if there is a strong preference for sons, one would expect to see relatively high sex ratios at the early parities and relatively low sex ratios for larger families, because couples would be more likely to discontinue childbearing if their early children were predominantly males”. On the other hand, it is also likely that in societies with preference for sons, childbearing continues as long as a son is not born. For example, in Pakistan, for second and onwards birth orders, normal sex ratios are reported. Similarly, much higher sex ratios at birth in rural areas are apparently due to reporting biases, where desire for sons is more prevalent because of their utility as farm labour. However, in the absence of any technology to determine the sex of the unborn child, sex-selective abortions are unknown in rural areas of the country.

Sex Ratios among Children under Five

  • * The first census was conducted in 1951 followed by the 1961 census. The 1971 census was delayed fo (...)

8Since independence in 1947, five censuses have been conducted in Pakistan*. As shown in Figure 1, while overall the sex ratio has been declining gradually from 120 in the 1951 census to 104 in the 1998 census, the sex ratio among children under 5 has fluctuated somewhat. It was 105 in the 1951 and 1961 censuses, declined to 101 and 97 in 1972 and 1981, respectively and in 1998 was reported as 104. This fluctuation is perhaps not real but is most likely due to the poor quality of data collected in censuses conducted in Pakistan. In a recent exercise conducted to evaluate the 1998 census data, Ali and Sultan (2003) however maintain that as compared to the 1972 and 1981 censuses there is a deficit of children aged 0 and 1 in the 1998 census. They point out that this inconsistent pattern is an indication of faulty quality of census data in Pakistan.

Figure 1: Overall sex ratio and sex ratio at age below 5 years reported in censuses of Pakistan: 1951-1998

Figure 1: Overall sex ratio and sex ratio at age below 5 years reported in censuses of Pakistan: 1951-1998

9Sex ratios in Pakistan based on the 1998 census for single years below age five for urban and rural areas and for each province are presented in Table 4 and indicate that sex ratios are within the range of 104 to 108 at each age in both rural and urban areas. Lower sex ratios in rural areas of Balochistan are somewhat unusual, which could be related to the poor quality of age-sex data in the 1998 census. Balochistan is not only the smallest province of the country, but is the least developed and some of its rural areas are difficult to access. Sex ratios calculated for each of the districts within each province (see Appendix, Tables 4A to 4D) again do not suggest any imbalanced sex ratios below age five. However, several districts within each province report sex ratios which are lower than 100. This could again be due to the poor quality of census data. A fairly high sex ratio of 138 is reported in only one district of Sindh Province (Appendix, Table 4B). Tharparkar is mainly a desert and rural area bordering the Indian state of Rajasthan. Although it is highly unlikely that sex-selective abortion is widely practised in this remote area, intentional or unintentional infanticide of female children can not be ruled out due to cultural reasons, which needs further investigation.

Table 4: Sex ratios at ages under five by rural urban residence: Pakistan and provinces, 1998

Table 4: Sex ratios at ages under five by rural urban residence: Pakistan and provinces, 1998

Sources: Census of Population, 1998.

Sex Differentials in Infant Mortality

10As a consequence of son preference, there is likely to be discrimination against female children, which could be reflected in the patterns of mortality in infancy through the neglect of female children. Although Alam and Cleland (1984) found no evidence of "selective neglect" of female children in Pakistan, they do report that in families with no previous sons, when a boy is born, the infant mortality rate is lower. This suggests that “a particularly precious child” receives special parental care. They however recognized the fact that although infant mortality among males is always higher than among females, it is offset by substantially higher post-neonatal mortality among females, as was suggested by the Pakistan Fertility Survey conducted in 1974-75.

Figure 2: Recent trends in infant mortality rate by sex Pakistan: 1990-2000

Figure 2: Recent trends in infant mortality rate by sex Pakistan: 1990-2000

11Figure 2 clearly demonstrates that the pace of mortality decline at infancy has been substantially faster for males than females. This could be due to the incidence of wider neglect of female children-or preferential treatment being given to male children-as greater value is attached to sons, which is perhaps gaining ground. Table 5 shows that the sex differentials in the infant mortality rate have narrowed substantially during the past decade, and consequently the male-female ratio has declined substantially from 1.19 to 1.00. Similarly, the female advantage even at an early period of life is also disappearing. To what extent this more rapid decline in male infant mortality has to do with the preferential treatment being given to the male child is examined in the following paragraph.

Table 5: Mortality rates (per 1000 live births) by sex 1990-01 and 2000-01

1990-01

2000-01

Infant Mortality Rate:

Male

102

77

Female

86

77

Male: Female Ratio

1:19

1:08

Neonatal Mortality Rate:

Male

60

47

Female

46

45

Male: Female Ratio

1:30

1:08

Post-neonatal Mortality Rate:

Male

42

29

Female

39

32

Male: Female Ratio

1:07

0:88

Sources: 1 Pakistan Demographic and Health Survey 1990-91.
2 Pakistan Reproductive Health and Family Planning Survey 2000-01.

Evidence of Preferential Treatment for Sons

12Strong son preference could also be assessed by whether daughters get less attention than sons in terms of breastfeeding by mothers and treatment provided in case of illness. Furthermore, the nutritional status of male and female children could provide important insights in this respect.

13Table 6 provides no evidence for the hypothesis that sons get preferential treatment over daughters in terms of breastfeeding practices and the age when they are given solid food. While there is fairly universal breastfeeding practice for both (97 per cent were breastfed), a higher percentage of female children is given solid food at early ages. An explanation of this could be that since girls are considered to be the weaker sex, mothers may be giving them solid food earlier than to boys.

Table 6: Breastfeeding and weaning patterns and nutritional status of children by sex (all figures are in per cent)

Male

Female

Mothers Reported Breastfeeding1

97.0

97.0

3-12 months old given solid food1

76.8

78.5

Under 5 years underweight2

41.5

40.5

Under 5 years stunted2

29.9

27.5

Under 5 years wasted2

11.6

11.8

Sources: 1 Pakistan Integrated Household Survey 2000-01.
2 Pakistan National Nutrition Survey 2001-02.

14Since parents may not want to admit that they treat sons and daughters differently, a comparison of the nutritional statuses of male and female children could give some insight into the differential in infant mortality by sex. However, Table 6 again does not show any significant difference in their nutritional status. Indeed in a society where sons are preferred over daughters, a slightly higher percentage of male children is found to be malnourished as compared to female children, which is rather surprising. This finding perhaps suggests that once a female child is born, there is no systematic neglect of her in evidence.

15When it comes to treatment of diarrhea -one of the major illnesses common among children-a preferential treatment given to sons is prevalent. For example, a slightly higher percentage of males is taken for treatment and more of them are taken to a private physicians than girls, who are taken to unqualified doctors, while some do not receive any treatment at all. These discriminations against girls may appear minor, but in the long run it could be a factor indicating more concern to save the life of a boy than of a girl.

Table 7: Prevalence of diarrhoea and its treatment by sex of child (all figures are in per cent)

Male

Female

Children 12-23 months immunized

96

98

Children <5 suffered from diarrhoea during the past month

13

11

Whether taken for treatment to a clinic/health centre

87

87

Type of practitioner consulted:

Private physician

61

56

Government facility

26

24

Unqualified person

8

12

Others/No one

5

7

Sources: Pakistan Integrated Household Survey, 2000-01.

Conclusion

16Although preference for sons is fairly strong in Pakistan, there is no evidence of higher than normal sex ratios at birth or at early ages of life. Male and female children are not treated differently by mothers, as reflected in similar patterns of breastfeeding and weaning. Indeed malnutrition among male children under five is slightly higher than among females. Similarly, overall no differences in health-seeking behaviour are noted for male and female children. However males do seem to receive preferential treatment when a higher percentage of mothers report taking them to a private doctor, while more females are taken to unqualified doctors. Although in the past infant mortality among males was higher than expected, female advantage is disappearing, suggesting that saving the male infant’s life is apparently becoming a norm.

17It is a well-established fact that there is a strong desire among parents in Pakistan to have more sons than daughters. Although this could be one of the many reasons for constantly high fertility in Pakistan, recent census and survey data do not suggest an increasing sex ratio at birth or a general neglect of daughters. Thus, it appears that although there is a strong desire for sons, daughters are not unwanted or grossly neglected after birth, as would be expected. Perhaps Pakistani society is somewhat unusual, whereby once girls are born they get as much attention as boys. It could be due to a strong belief in fate among parents, which does not encourage them to go for sex-selective abortion or to neglect the girl child, which would be considered tantamount to interference in the God’s will. However, the reasons for a more rapid decline in infant mortality among males than among females need to be thoroughly investigated. Furthermore, since the fertility level in Pakistan has yet to decline considerably, especially among the group where son preference is more prominent (e.g., in rural areas and among lower socio-economic strata), it is unlikely that son preference per se, will have any bearing on creating a sex imbalance in the country in the near future.

Bibliographie

References

Alam, I. and Cleland, J. (1984), ‘Infant and Child Mortality: Trends and Determinants’, in Alam I. and Dinesen B. (eds.), Fertility in Pakistan, A Review of Findings from Pakistan Fertility Survey, Voorburg (Netherlands): International Statistical Institute.

Ali, S.M. (1989), ‘Does son preference matter?’, Journal of Bio-social Science, Vol. 21, No. 4.

Ali, S.M. and Sultan, S. (2003) ‘Age and Sex Distribution of 1998 Census: An Evaluation’, in Kemal A.R. et al. (eds.), Population of Pakistan: An Analysis of 1998 Population and Housing Census, Islamabad: Pakistan Institute of Development Economics.

Arnold, F. (1992), ‘Sex Preference and its Demographic and Health Implications’, International Family Planning Perspective, Vol. 18, No. 3.

Arnold, F. (1997), ‘Son Preference in South Asia’, Proceedings of Seminar on Comparative Perspectives on Fertility Transition in South Asia, Rawalpindi, Pakistan, December 1996, IUSSP.

Farooqui, M.N.I. (1990), ‘Sex Preference and Contraceptive Use: Some Evidence from Pakistan NGO Survey’, Proceedings Fourteenth Research Seminar on Population Welfare, Karachi: National Research Institute of Fertility Control.

Federal Bureau of Statistics, (2000), Pakistan Demographic Survey, Karachi: Statistics Division, Government of Pakistan.

Khan, M.A. and Sirageldin, I. (1977), ‘Son preference and the demand for additional children in Pakistan’, Demography, Vol. 14, No. 4.

National Institute of Population Studies (1992), Pakistan Demographic and Health Survey 1990-1991, Islamabad: National Institute of Population Studies.

National Institute of Population Studies (1992), Pakistan Reproductive Health and Family Planning Survey 2000-01, Islamabad: National Institute of Population Studies.

Rukunuddin, A.R. (1982), ‘Infant Child Mortality and Son Preference as Factors Influencing Fertility in Pakistan’, The Pakistan Development Review, Vol. XXI, No. 4.

Sathar, Z. and Karim, M.S. (1997), ‘Son Preference in Pakistan’s High Fertility Setting’, in Sex Preference for Children and Gender Discrimination in Asia, Seoul: Korea Institute for Health and Social Affairs.

Sotoudeh-Zand, M. (1988), ‘The Economic Value of Children’, Islamabad: National Institute of Population Studies, (mimeograph).

Yi, Z. (1993), ‘Causes and Implications of the Recent Increase in the Reported Sex Ratio at Birth in China’, Population and Development Review, Vol. 19, No. 2.

Annexes

Appendix Table 4A. Sex ratios in population under five in districts of Punjab Province by urban and rural areas 1998

District

All Areas

Rural

Urban

Attock

104

104

105

Bhakkar

106

106

106

Bhawalnagar

104

105

103

Bhawalpur

104

104

105

Chakwal

105

105

105

Faisalabad

104

104

104

Gujranwala

104

104

104

Gujrat

105

105

105

Hafizabad

105

105

105

Jehellum

104

105

102

Jhang

105

105

105

Kasur

104

104

105

Khanewal

104

104

104

Khushab

106

106

104

Lahore

104

105

104

Layyah

105

104

105

Lodhran

105

105

104

Mandi Bahauddin

105

105

102

Mianwali

105

106

104

Multan

104

104

104

Muzaffar Garh

103

103

104

Narowal

104

104

104

Okara

103

103

104

Pakpattan

104

104

106

Rajanpur

102

102

105

Rawalpindi

105

104

105

Sahiwal

104

105

104

Sargodha

105

105

106

Sheikhupura

104

104

104

Sialkot

105

105

104

Tobatek Singh

105

105

102

Vehari

106

106

104

Sources: Census of Population, 1998.

Appendix Table 4B. Sex ratios in population under five in districts of Sindh Province by urban and rural areas 1998

District

All Areas

Rural

Urban

Badin

104

105

102

Dadu

102

101

106

Ghotki

103

104

101

Hyderabad

104

103

104

Jaccobabad

98

96

103

Karachi

105

104

105

Khirpur

102

102

103

Larkana

102

100

105

Mirpur Khas

104

103

104

Naushahro

105

106

103

Nawabshah

102

101

105

Sanghar

103

103

104

Shikarpur

102

102

101

Sukkur

104

103

105

Tharparkar

138

138

116

Thatta

104

104

107

Umerkoth

108

108

108

Sources: Census of Population, 1998.

Appendix Table 4C. Sex ratios in population under five in NorthWest Frontier Province by urban and rural areas, 1998

District

All Areas

Rural

Urban

Abbott Abad

105

105

107

Bannu

103

103

109

Batagram

104

104

-

Buner

104

104

-

Charsadda

106

105

106

Chitral

104

104

106

D. I Khan

107

107

106

Hangu

107

106

107

Haripur

105

105

104

Karak

107

107

104

Kohat

107

107

109

Kohistan

93

93

-

Lakki Marwat

106

106

109

Lower Dir

105

105

115

Malakand

105

106

99

Mardan

106

106

105

Masehra

113

113

105

Nowshehra

106

105

107

Peshawar

105

105

106

Shangla

104

104

-

Swabi

105

105

103

Swat

104

104

107

Tank

105

105

105

Upper Dir

102

102

103

Sources: Census of Population, 1998.

Appendix Table 4D. Sex ratios in population under five in Balochistan Province by urban and rural areas, 1998

District

All Areas

Rural

Urban

Awaran

108

108

-

Barkhan

91

90

109

Bolan

104

103

110

Chagai

104

104

107

Chagai

108

108

107

Dera Bagti

88

86

109

Gawadar

113

115

111

Jaffar Abad

96

94

103

Jhal Magsi

100

100

98

Kallat

90

89

97

Kech

114

115

109

Kharran

100

98

112

Khuzdar

98

96

103

Killa Saifullah

113

112

116

Kohlu

92

90

111

Labela

102

102

101

Loralai

93

92

106

Mastung

98

98

101

Mussa Khel

73

72

83

Naseerabad

99

99

101

Pishin

113

113

106

Punjgor

112

112

107

Quetta

107

111

105

Sibbi

101

98

107

Zhob

93

90

113

Ziarat

119

119

92

Sources: Census of Population, 1998.

Notes de fin

* The first census was conducted in 1951 followed by the 1961 census. The 1971 census was delayed for a year and was followed by the 1981 census. The 1991 census was not conducted and the last census was conducted in 1998.

Table des illustrations

Titre Table 1: Percentage of currently married, non-pregnant women 15-49 who want another child, by number and sex composition of living children, South Asian countries
Légende Source: Arnold (1997).
URL http://books.openedition.org/ifp/docannexe/image/4561/img-1.jpg
Fichier image/jpeg, 76k
Titre Table 2: Per cent distribution of currently married women wanting another child, by their preferred sex of the next child by number of living children and living sons
Légende * Less than 25 observations.Sources: 1990-91 Pakistan Demographic and Health Survey, 1990-91.2000-01 Pakistan Reproductive Health and Family Planning Survey 2000-01.
URL http://books.openedition.org/ifp/docannexe/image/4561/img-2.jpg
Fichier image/jpeg, 170k
Titre Table 3: Sex ratios at birth by birth order (males per 100 females)
Légende Source: Pakistan Integrated Household Surveys.
URL http://books.openedition.org/ifp/docannexe/image/4561/img-3.jpg
Fichier image/jpeg, 71k
Titre Figure 1: Overall sex ratio and sex ratio at age below 5 years reported in censuses of Pakistan: 1951-1998
URL http://books.openedition.org/ifp/docannexe/image/4561/img-4.jpg
Fichier image/jpeg, 81k
Titre Table 4: Sex ratios at ages under five by rural urban residence: Pakistan and provinces, 1998
Légende Sources: Census of Population, 1998.
URL http://books.openedition.org/ifp/docannexe/image/4561/img-5.jpg
Fichier image/jpeg, 151k
Titre Figure 2: Recent trends in infant mortality rate by sex Pakistan: 1990-2000
URL http://books.openedition.org/ifp/docannexe/image/4561/img-6.jpg
Fichier image/jpeg, 89k

Auteur

Head Reproductive Health Program and Professor of Demography
Department of Community Health Sciences
Aga Khan University
Stadium Road, P.O. Box 3500
Karachi 74800, PAKISTAN
mehtab.karim@aku.edu

© Institut Français de Pondichéry, 2005

Conditions d’utilisation : http://www.openedition.org/6540

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search