African Population Studies Union for African Population Studies ISSN: 0850-5780 Vol. 12, Num. 2, 1997

African Population Studies/Etude de la Population Africaine, Vol. 12, No. 2, September/septembre 1997

Tiisetso MAKATJANE

Department of Statistics and Demograph, National University of Lesotho, Roma

Code Number: ep97010

ABSTRACT

It is often assumed that women in female-headed households are less sexually active and, therefore, are expected to make less use of modern methods of contraception than their counterparts in male-headed households. The data from a demographic and health survey carried out in 1991 was used to examine the effects of type of households in which women live on their contraceptive behaviour. The sex of the household head was found to make a difference on the level of contraceptive use in Lesotho. However, the effect is dependent on the marital status of the woman and whether the method in use is modern or traditional. The results show that women in male-headed households reported higher rates of modern contraceptive prevalence than those in female-headed households. There is no support for the hypothesis that women in female-headed households use modern contraceptives more than women in male-headed households except for never married women.

RÉSUMÉ

On présume souvent que les femmes chefs de ménage sont moins actives sexuellement et donc moins susceptibles de recourir aux méthodes contraceptives modernes que les femmes dans les foyers dont le chef de famille est un homme. Les données issues d’une enquête démographique et de santé menée en 1991 ont été utilisées pour examiner les effets du type de ménage dans lequel les femmes vivent sur leur comportement sexuel. On a trouvé qu’au Lesotho le niveau de la pratique contraceptive était différent suivant le sexe du chef de famille. Cependant, l’effet dépend de la situation matrimoniale de la femme ou du fait que la méthode utilisée est moderne ou traditionnelle. Les résultats montrent que les femmes vivant dans des foyers où le chef de famille est un homme faisaient plus grand usage des méthodes contraceptives modernes que celles dans des foyers dont le chef de famille est une femme. Il n’y aucune preuve pour étayer l’hypothèse selon laquelle les femmes dans des ménages dont le chef est une femme recourent aux méthodes contraceptives modernes plus que celles vivant dans des ménages dont le chef est un homme, à l’exception des femmes qui ne se sont jamais mariées.

INTRODUCTION

Available literature suggests that, Basotho men, generally, have a negative attitude towards use of modern contraceptives (Poulter et al., 1981; Makatjane, 1987; Clarke, 1984; Hall and Malahleha, 1989; Shale and Makatjane, 1988). The most commonly cited reason for Basotho men's displeasure with modern contraceptives is fear of a high likelihood of infidelity by their wives. Literature further shows that in men's view, spouse separation stemming out of labour migration invalidates the need for use of modern contraceptives as there is no risk of pregnancy in their absence (Makatjane, 1987).

Some studies have documented the husband’s non-approval of contraceptive use as a reason for non-use of contraceptives among married women (Schuster, 1979; UNFPA, 1991; Wener, 1983; Hall and Malahleha, 1989; Sembajwe  and Makatjane, 1987). These studies suggest that women residing in male-headed households are less likely to use modern contraceptives than women residing in female-headed households. That is, since men are generally against use of modern contraceptives, households headed by men are not a conducive environment within which women can be encouraged to use modern contraceptives. It is arguable, therefore, that men's attitude towards use of modern contraceptives might be influencing women's use of modern contraceptives.

The majority of households are headed by men on a de jure basis in Lesotho. Less than a third of households in Lesotho are headed by women on a de jure basis (Sembajwe and Makatjane, 1987a: 1; Makatjane, 1990: 9; Bureau of Statistics, 1988: 7; Lawry, 1986: 8). Generally, women acquire household headship late in life mainly through widowhood and to a lesser extent through separation and divorce. According to figures of the Lesotho 1986/87 Household Budget Survey, in about nine out of ten households headed by women, the head was either widowed, divorced or separated while among male-headed households, 90 per cent of the heads were currently married (Makatjane, 1990). The other categories of households that are headed by women on a de jure basis are those of never married women constituting 7 per cent of female-headed households. On the basis of these sex differences on household headship rates, the rate of acceptance of the use of contraceptives would, on the average, be expected to be low in Lesotho based on the hypothesis that male-headed households are not necessarily a conducive environment for adoption of contraceptives. That is, if it is correct that since men are reported to have a negative attitude towards contraceptive use, women residing in male-headed households would report lower rates of contraceptive prevalence, acceptance of modern contraceptive use would be low because the majority of households are headed by men.

Modern contraceptive use is not only a relatively new concept in developing countries, but is also deemed contradictory to most traditional norms which are characteristically pro-natalist. The introduction of modern contraceptives in the majority of developing countries has been militated against by resistance influenced by the belief that since children are a gift from God, it is therefore ungodly to limit such a gift (see for example, Poulter et al., 1981). Although contraceptive use is being accepted over time, a lot of resistance against it would still be found among old women who are also less educated than the younger lot. This is because older people should have more influence on tradition and culture than young individuals. Since on the average female household heads are older than male household heads (Makatjane, 1990), the sex of the household head alone might not determine whether a particular household creates a conducive atmosphere for increased acceptance of contraceptive use but age of the household head could also be an influencing factor. This suggests that female-headed households might not be as conducive for increased use of contraceptives as suggested in the preceding paragraphs since female heads are older and less educated and might perceive modern contraceptives as ungodly if they are meant to limit the number of children other than help in their spacing.

Notwithstanding that women in general support use of modern contraceptives, women residing in female-headed households should be using modern contraceptives more than their counterparts residing in male-headed households. Since marriage is not only an affair between husband and wife but also includes parents as well as relatives of both parties, marriage dissolution through either separation or divorce does not necessarily mean that the relationship between the wife and the husband's parents and relatives is severed. If separated or divorced women are still in good terms with their in-laws irrespective of their marriage situation, such women might not want to sever their relationship with their in-laws by giving birth to a child not fathered by their son. This would also be the case in a situation of someone contemplating that the marriage could still be salvaged because giving birth to a child by another man would jeopardise the chances of reconciling the marriage. Such women would have to protect themselves against unwanted pregnancies.

On the contrary, if the relatives of the husband think that the wife is in the wrong in a case of divorce or separation, children born after the marriage dissolution are not likely to be accepted in the husband's lineage and would be made to suffer a number of humiliations during customary and traditional family rituals such as burial formalities. A woman who is separated or divorced could also use contraceptives to avoid giving birth to children who might not be accepted in any lineage particularly if she has a partner with whom she is living on a more or less permanent basis. These factors would seem to encourage women in these circumstances to use contraceptives more than would be normal.

Although, as indicated above, women generally acquire de jure household headship through marriage dissolution, it is also worth noting that a widow would not normally assume household headship if her son is old enough to assume household headship (see Makatjane and Botana, 1990; Murray, 1981). This would suggest that female-headed households have a low overall sex ratio. This would further suggest that the majority of women in female-headed households are either never married, separated, divorced or widowed. It can therefore be argued that, due to their marital status, the contraceptive prevalence rate of women residing in households headed by women would be low because they might be sexually inactive as they might not be staying with a man. Moreover, even if these women do acquire partners despite their marital status, they would still use contraceptives less because of their infrequent sexual intercourse unless they are permanently staying with their newly acquired partners.

It is also arguable that spousal separation cannot necessarily be taken to mean sexual inactivity on the part of the woman. Instances where women fell pregnant while husbands were away in the South African mines (Mueller, 1977: 208) do suggest that absence of the husband does not necessarily mean that the wife is sexually inactive. On the other hand, the relatively low fertility of Lesotho by African standards, which is explained in terms of spousal separation stemming from male labour migration, implies that women falling pregnant in the absence of their husbands are isolated instances which cannot be generalised for the whole country. Since of late migrants visit their homes almost every fortnight, spousal separation cannot be a factor to consider now. However, frequent but unpredictable home visits of married men working as labour migrants in the South African mines might influence wives of labour migrants to use contraceptives more than their counterparts where husbands are not labour migrants. Due to the unpredictable home visits of their husbands, wives of labour migrants are more likely to use contraceptives to avoid unwanted pregnancies.

The preceding discussion has demonstrated that our knowledge is far from complete concerning the role of the sex of the household head in influencing use of modern contraceptives. The discussion has further presented evidence which largely suggest that women residing in female-headed households would report higher rates of contraceptive prevalence although information which would suggest the contrary is also presented. That is, since women generally support use of modern contraceptives, female-headed households are conducive for increased use of contraceptives while the opposite might be the case for male-headed households. On the other hand, it is equally plausible that women in female-headed households can report low rates of contraceptive use due to their marital status which suggests that they should be sexually inactive. Only empirical data can help to answer the questions raised so far; hence the importance of this study.

Since it is in the interest of the Government of Lesotho for the majority of couples to use contraceptives (Lesotho, 1992), knowledge about factors influencing contraceptive prevalence would facilitate action in removing obstacles in the way of increased contraceptive prevalence. Hence this study does not only investigate the role of sex of household head on contraceptive prevalence for its own sake, but the results of the study would also provide valuable information for decision and policy makers to further perfect their strategies of creating a conducive atmosphere for increased use of modern contraceptives as a means of bringing population growth rate in Lesotho to an acceptable level. The results of the study will further provide a better understanding of female-headed households.

The Data

The 1991 Lesotho Demographic and Health Survey is the source of data for the present analysis. The survey was a national representative sample survey covering both urban and rural households. The survey instrument was divided into several sections and this study is using information mainly from the contraception section. Sex of household head, marital status, education and work status information from other sections have also been linked to data on contraception.

Only information from women in their reproductive ages has been used to measure contraceptive prevalence. A total of more than 3,000 women form the sample size for the analysis.

Method of Analysis

Contraceptive prevalence rates are calculated for women disaggregated on the basis of the sex of household head. Comparisons of observed prevalence rates are made on the basis of marital status of the women. In all cases a test is made to find out whether differences in the prevalence rates are statistically significant. To establish whether differences are not due to differences in the characteristics of women as indicated in the introduction but as a result of the influence of the household head, the effects of background characteristics are later controlled for in the observed prevalence rates.

Characteristics of the Population

The characteristics of the study population are presented in Table 1. Generally, respondents in male-headed households are similar in characteristics with those residing in female-headed households except with respect to few characteristics. Major differences concern marital status where the majority of currently married women reside in male-headed households while the never married and previously married women are concentrated in female-headed households. Work status is another exception. Women residing in male-headed households are more likely to report that they are not working. They are also slightly more educated and younger than women residing in female-headed households.

Table 1: Characteristics of the study population by sex of the household head; Lesotho 1991

Since households in Lesotho are headed by men except in situations where the woman is widowed and does not have an elderly son who would assume headship (Murray, 1981), the presence of currently married women in female-headed households needs an explanation. Out of 131 currently married women residing in female-headed households, 125 had their husbands elsewhere in Lesotho or in South Africa. The most probable reason for the presence of currently married women in female-headed households is return to parental home for child delivery or related activities. This is further supported by the fact that currently married women residing in female-headed households are mainly daughters of the household head.

RESULTS

Generally speaking, the differences in contraceptive prevalence depend on the marital status of the women and the method used. Currently married women use contraceptives more if they are residing in male-headed households while the opposite is the case with respect to traditional methods. That is, currently married women residing in female-headed households are using traditional methods than their counterparts residing in male-headed households. As regards never married women, those residing in female-headed households use modern contraceptives more than their counterparts residing in male-headed households. For traditional methods, there are no differences in prevalence among never married women by sex of the household head. Among previously married women there are no significant differences in contraceptive use by sex of the household head for both modern and traditional methods (Table 2).

Table 2 : Contraceptive prevalence by marital status and sex of household head

Note: * differences are significant at 10% level ** differences are significant at 5% level

Controlling the Effects of Other Factors

It was observed in Table 1 that currently married women are concentrated in male-headed households while never married and previously married women are concentrated in female-headed households. Moreover, factors such as rural-urban residence, educational attainment, religion and work status are known to influence use of contraceptives. Labour migration has also been shown to influence the demography of southern African countries involved in labour migration (Makatjane, 1994; Tuoane, 1995; Chimere-Dan, 1996) and, in particular, it has been suggested that labour migration is responsible for the high use of modern contraceptives among never married women in some parts of South Africa (Chimere-Dan, 1996). It is therefore important to establish whether the differences in contraceptive prevalence in Table 2 are genuine or due to differences in the characteristics of the respondents. Table 3 presents the results.

As was the case in Table 2, among currently married women, those residing in male-headed households use modern contraceptives more than their counterparts residing in female-headed households and the opposite is the case for never married women where women residing in female-headed households use contraceptives more than their counterparts residing in male-headed households. For all the background factors in Table 3 where differences in contraceptive prevalence are statistically significant, among currently married women those residing in male-headed households are more prone to use contraceptives than their counterparts residing in female-headed households while the opposite is the case among never married women. For all the factors included in Table 3 there is at least one group where differences in contraceptive prevalence between women residing in female and male-headed households are statistically significant for currently married and never married women. Previously married women are still an exception as was the case in Table 2 in that contraceptive use among them does not differ by sex of the household head. It is only with respect to age where previously married women aged less than 30, and residing in female-headed households are using modern contraceptives more than their counterparts in male-headed households.

DISCUSSION AND CONCLUSION

According to the results in Tables 2 and 3, it can be concluded that the sex of the household head does matter in the use of contraceptives. However, the effect is dependent on the marital status of the woman and whether the method in use is a modern method or a traditional method. For modern methods, currently married women residing in male-headed households reported higher contraceptive use than their counterparts residing in female-headed households while never married women residing in female-headed households are more likely to use modern contraceptives than their counterparts residing in male-headed households. For traditional methods, currently married women residing in female-headed households use traditional methods more than their counterparts residing in male-headed households.

Results of this study have not supported the hypothesis that women in female-headed households use modern contraceptives more than women in male-headed households except for never married women. Contrary to the argument made in the introduction, women in male-headed households reported higher rates of modern contraceptive use. This was generally true irrespective of the background characteristics of the women. With respect to traditional methods, currently married women residing in female-headed households are more inclined to use these methods than their counterparts in male-headed households.

It was hypothesised in the introduction that, due to their marital circumstances, women in female-headed households should use modern contraceptives less than their counterparts in male-headed households. The major underlying assumption for this hypothesis was that women in female-headed households are sexually inactive due to their marital circumstances. The results of the study do support this hypothesis with respect to currently married women.

Table 3 : Contraceptive prevalence (any method) by sex of the household head and marital status for selected background factors.

Note: * differences significant at 10% level ** differences significant at 5% level *** differences significant at 1% level
- information is not presented as group has less than 20 cases

This is based on the lower prevalence rate obtained for currently married women in female-headed households. It is, however, noted that part of sexual inactivity among currently married women residing in female-headed households is probably a result of having returned to their natal home for child delivery and having to rely on sexual abstinence as argued earlier in the paper. This is further supported by a higher use of traditional methods of contraceptives for these women.

It was indicated in the introduction that contraceptive use is expected to decline with age as the frequency of sexual intercourse declines. Since previously married women are generally older (70 per cent of previously married women are aged 30 or more compared with 49 and 12 per cent for currently married women and never married women respectively) coupled with their expected sexual inactivity resulting from their marital circumstances, it is probably in order that their use of contraceptives should not be influenced by sex of the household head. The significant differences in contraceptive use among previously married women aged less than 30 are indicative of high sexual activity due to young age. However, because we grouped the separated, divorced and widowed women together due to few numbers of divorced and separated women, it has not been possible to address some of the arguments raised in the introduction, particularly contraceptive use of separated or divorced women. This could have been revealing since separated and divorced women are not likely to be as old as the widows and their relations with their in-laws might have an impact on their contraceptive use as indicated in the introduction.

Migration status of the husband was included in Table 3 as a factor for currently married women and it has appeared as an important factor. The high contraceptive prevalence among wives of labour migrants in male-headed households is interpreted as a protection on the part of the women against unwanted pregnancies which might result from unpredictable but frequent home visits of their husbands who are working as labour migrants in the South African mines. Bearing in mind that wives of labour migrants residing in female-headed households might have returned to their natal home for child delivery, coupled with the fact that sons-in-law are not by tradition expected to visit their in-laws without prior arrangements, wives of labour migrants residing in female-headed households should be using contraceptives less than their counterparts residing in male-headed households.

The figures in Table 3 have shown that the age of the woman has an effect on contraceptive use for all women by marital status. However, the effect is not the same. For both never married and previously married women, significant differences are for women aged less than 30 while for currently married women significant differences are among women aged 30 or more. Based on the argument that couples are likely to use contraceptives less when fecundity and frequency of intercourse decline due to advancement in age (Njogu, 1991), significant differences among women aged less than 30 for never married and previously married women are according to expectation while significant differences among women aged 30 or more for currently married women are not. It is arguable that currently married women are using contraceptives because they are highly sexually active due to the presence of husbands and they are probably doing this to avoid unwanted pregnancies.

REFERENCES

• Bureau of Statistics 1988. Income, Expenditure and Consumption of Basotho Households Methodological Report from the Household Budget Survey 1986/87. Lesotho: Bureau of Statistics.
• Chimere-Dan, O. 1996. "Contraceptive Prevalence in Rural South Africa", International Family Planning Perspective 22 (1):4-9.
• Clarke, E. T. 1984. Male Attitudes to and Responsibilities in Family Planning in Lesotho. Institute of Southern African Studies, National University of Lesotho.
• Gay, J. 1988. "Wage Employment of Rural Basotho Women: A Case Study", South African Labour Bulletin, 6 (4) 40-53.
• Hall, D and G. Malahleha 1989. Health and Family Planning Services in Lesotho: The People's Perspectives, Morija: Morija Printing Works Lesotho.
• Lawry, S. 1986. Livestock and Range Management in Sehlabathebe: A Study of Communal Resource Management, Lesotho: Ministry of Agriculture, Range and Management Division.
• Lesotho 1992. First Five Year Development Plan; Lesotho: Government Printer.
• Makatjane, T. 1985. Internal Migration and Fertility in Lesotho, Working Papers in Demography No 6, Lesotho: Demography Unit Department of Statistics National University of Lesotho.
• ____________. 1987. Child-Spacing in Lesotho. Working Papers in Demography No. 9 Demography Unit National University of Lesotho.
• ____________. 1990. Lesotho's Household Structure: Some Reflections from the Household Budget Survey. Working Papers in Demography No. 18 Demography Unit National University of Lesotho.
• ____________. 1992. Dynamics of Females Activity Rate Differentials in Lesotho: Results from the 1986/87 Household Survey. Working Papers in Demography No. 20 Demography Unit National University of Lesotho.
• ___________. 1994. "The Future of Labour Migration: Its Implications for Lesotho", Lesotho Social Sciences Review vol. 1 (2):26-38.
• Makatjane, T. and Toziba Botana 1990. "Female-Headed Households: Some Methodological Issues", Paper Presented at the Thirteenth Annual Conference of the African Studies Association of Australia and the Pacific, held at Deakin University.
• Mueller, M. B. 1977. ‘’Women and Men in Lesotho: Periphery of the Periphery’’. PhD Thesis Presented at Brandies University.
• Murray, C. 1981. "The Effect of Migrant Labour: A Review of Evidence from Lesotho" South African Labour Bulletin, 6 (4) 21-39.
• Njogu, W. 1991. "Trends and Determinants of Contraceptive Use in Kenya", Demography 28, (1): 83-99.
• Poulter, S.; W. Maclain; J. B. Kaburise; J. Mugambwa and D. Milazi 1981. Law and Population Growth in Lesotho, Lesotho: National University of Lesotho.
• Schuster, I. M. G. 1979. New Women of Lusaka, USA: Mayfield Publishing Co.
• Sembajwe, I. and T. Makatjane 1987. Fertility and Mortality Survey: Information on Females Aged 15 to 59 Years, Working Papers in Demography No. 12, Lesotho: Demography Unit Department of Statistics National University of Lesotho.
• ____________________________. 1987a. Fertility and Mortality Survey: Information on Household Characteristics, Working Papers in Demography No. 11, Lesotho: Demography Unit Department of Statistics National University of Lesotho.
• Shale, M. and T. Makatjane. 1988. Evaluation Report, Lesotho: Planned Parenthood Association Male Motivation Project.
• Tuoane,'M. 1995. Socio-economic Status of Women and Fertility in Lesotho. Working Papers in Demography No. 22, Lesotho: Demography Unit Department of Statistics National University of Lesotho.
• UNFPA 1991. The State of World Population, London: Nuffield Press.
• Wener 1993. Use of Birth Control by the Rural Population, USA: Wiley and Sons.

Copyright 1997 - Union for African Population Studies.