 |
| About Bioline | All Journals | Testimonials | Membership | News |
|
||||||
|
||||||
African Population Studies/Etude de la Population Africaine, Vol. 19, No. 1, April 2004, pp. 81-98 Household Socioeconomic Status and Sexual Behaviour among Nigerian Female Youth Uche C. Isiugo-Abanihe1, Kola’A. Oyediran2 1Department of Sociology, University of Ibadan, Ibadan, Nigeria Code Number: ep04005 ABSTRACTThis paper examines the determinants of sexual behaviour with special reference to the effect of household socio-economic status as a proxy for poverty. The data are derived from the Nigeria Demographic and Health Survey; the analysis is restricted to 1,831 never married females age 15-24. Both descriptive and analytical methods are used to assess the effects of each poverty-related factor when the effects of other demographic and socio-cultural factors are controlled statistically. The results show that 31.5% of the respondents have had sexual intercourse, and more than half of these had an affair in the month preceding the survey. The median age of sexual debut is 17 years, and there is little variation among socio-demographic and poverty-related indices, indicating a generally early initiation of sexual activity. Condom use is low among the youth; only 15% have ever used condom, and about 22% of the sexually experienced used condom the last time they had an affair. The results indicate that those who have access to the media and those of high socio-economic status are more sexually exposed than their counterparts who do not have access to media or have less household facilities. Exposure to premarital sex increases with access to newspapers, radio and television. The data do not support the view that the high level of sexual activity is a function of household poverty, although poverty exposes the youth to reproductive health risks as a result of low usage of condom among those from less well-off homes. BACKGROUND There has been considerable concern in many countries about the sexual and reproductive health of young people, in part because of their perceived increased vulnerability to the risk of sexually transmitted infections (STIs), including acquired immune deficiency syndrome (AIDS) (Knode-lule et al., 1997; Preston-Whyte, 1994; Scommegna, 1996; Twa-Twa et al., 1997), the potential risks to their health due to early pregnancy (AbouZahr and Royston, 1991; Barreto et al., 1992; United Nations, 1989), and the negative consequences of early and non-marital childbearing to young people’s life prospects (Hayes, 1987). Public concern over the reproductive health problems among Nigerian youth has drawn the attention of researchers, non-governmental organizations (NGOs) and policy makers to examining the driving force behind the upsurge in adolescent sexual activity. For instance, Isiugo-Abanihe (1994) has shown that more than 38 % of female adolescents in Nigeria, age 15 to 24 years, had initiated sexually activity in 1990, with a mean age of sexual debut less than 17 years. Nearly 60 % of those who had initiated sex had affairs within four weeks of the survey. Despite the near universal adult and policy makers’discomfort with the subject, consensus has begun to build in Nigeria that young people need expanded information, skills, and services with respect to sexual and reproductive health. Clearly, entry into reproductive life is a key transition in a person’s life and the choices and behavioural patterns acquired during this early stage typically shape the subsequent life course, a transition that is marked by critical life events: puberty, sexual initiation, marriage and childbearing (UN, 2002). Also, the growing concern about the reproductive health of adolescents and young adults derives in part from the sheer size of their cohorts. According to estimates by the National Population Commission (NPC, 1998), about 84 million Nigerians are currently under age 25, or about 63 % of the population; nearly 59 million or about 44 % are under age 15. The population classified as youth in this paper, those between ages 15 and 24, is estimated to be about 19 % of the Nigerian population or about 25 million in 2005, more than the population of any other country in West Africa. The overall health and development of the youth are shaped by many factors of which health programming is only one. These factors range from the social, economic, cultural and political conditions of the wider society, to those that characterize the living situation of an individual adolescent, including the family situation with respect to education, and income levels. Interest in the health of adolescents has grown, particularly in sub-Saharan Africa countries that are characterized by an economic environment of endemic poverty. The downturn in Nigeria’s economy over the last two decades, associated with increased unemployment, retrenchment of workers and reduction in family income, has adversely affected living standards, as parental resources have become inadequate to meet the various needs of household members, especially those of adolescents and young people. As a result, many families had to withdraw children from schools, giving rise to early entry into the informal labour force and increased contribution by children to family income through various means. By the mid-1980s, the level of the economic crisis had worsened and led to the introduction of structural adjustment programme (SAP) in Nigeria in 1986 this was launched to create a self-reliant and vibrant economy that will effectively reshape and restructure consumption and production patterns. In reality, however, SAP led to a lower standard of living for most Nigerians as a result of the devaluation of the naira, privatisation and commercialisation of public enterprises and substantial reduction in agricultural and petroleum subsidies, among other economic measures. Obadan and Odusola (2001) have argued that though the GNP per capita stood at $963 in 1982 and fell to $499 in 1986, it terribly worsened since the beginning of the 1990s. Among other consequences, SAP has engendered widespread poverty, strengthened people’s survival strategies and coping mechanisms and has tended to weaken moral values that moderate sexual behaviour especially among adolescents. Mass poverty during this period and people’s subsequent adjustment activity aggravated several adolescents’antisocial behaviour, particularly indis-criminate and clandestine social involvement. Female adolescents had to contend with the allurement of financial gratification and sexual overtures by relatively richer peers and adults. Consequently, poverty or the need to survive became the driving force and motivation behind the sexual activity of adolescent girls (Isiugo-Abanihe, 1993). Indeed, evidence from many developing countries suggests that poorer women are more likely to have non-regular partners and that condom use with non-regular partners is significantly lower among poorer women (Filmer, nd.). Booysen and Summerton (2002) have observed for South Africa that poverty increases the vulnerability of women to HIV infection by resulting, among other things, in unsafe sexual practices, often due to a lack of knowledge, lack of access to means of protection, and inability to negotiate condom use with sexual partners as a result of entrenched gender roles and power relations. In this respect, attention has been drawn on the ‘sugar-daddy’syndrome, whereby schoolgirls enter into sexual relationships with older, wealthy men who can assist them with school related expenses or the purchase of material goods (De Bruyn, 1992; Gorgen et al., 1993; Isiugo-Abanihe, 1993; Schoepf, 1994). The rising level of adolescent female sexual activity is also a function of the need to achieve or maintain an upscale life-style or for the longer-term objectives of establishing contacts with wealthy or prestigious people, and of obtaining assistance with finding a good job (Meekers and Calves, 1997); others use sex as a bargain for marriage or to prove their fecundity as a prelude or prerequisite for marriage (Isiugo-Abanihe, 1993). This paper examines the linkages between poverty and sexual activity among unmarried female youth in Nigeria. In the absence of direct measures of poverty, two proxy measures are adopted namely, access or non-access to means of information and essential household facilities such as radio, television, refrigerators, motorcycle, and bicycle among others. DATA AND METHODSThe analysis presented in this paper uses data collected in the 1999 Nigeria Demographic and Health Survey (NDHS), conducted between March and May 1999. It was a national representative sample survey covering both urban and rural households. The area-sampling frame for the survey was based on the enumeration areas (EAs) map prepared by the National Population Commission (NPC) for the conduct of 1991 census. Sample selection was done in two stages: First, 400 EAs were selected with equal probability; second, within each of these 400 EAs, a complete listing was done of all residents in sampled households, from which male and female respondents were interviewed. A total of 9,810 women aged 10-49 and 3,082 men aged 15-64 were interviewed. Among the female respondents, 1,831 of age 15-24 years have never married or lived together with a man. The analysis here is restricted to this sub-group. Past studies have found some association between household or family economic status and risky reproductive health behaviour among adolescents. Two indicators of household socio-economic status (SES) are included in the analysis: an index of seven household assets, and an index of access to information. The index of household facilities measures whether the household had electricity, a radio, television, telephone, refrigerator, and whether any member of the household owned a bicycle and a motorcycle. The index of access to information is derived from respondents’accessibility to newspaper, radio and television. The scale for household assets ranges from zero to seven, while that of access to information ranges from zero to three, with higher scores indicating higher household-socio-economic status and low scores for low SES. These two indices are used as proxy measures of household well-being, which is hypothesized to be directly related to that of the adolescent within. In addition, type of toilet facilities available and floor materials are also considered as possible predictors of household socio-economic status that could influence adolescent sexuality. Both bivariate and multivariate analyses are conducted to ascertain the association and net effect of the key independent variables (poverty proxies) on the dependent variables (sexual activity and condom use) when selected background characteristics are controlled. Logistic regression is used to assess the predictors of the four binary outcomes: whether respondents have had sex, whether they had sex in the last four weeks preceding the interviews, whether they had ever used condom, and whether condom was used at the last sexual intercourse. Socio-demographic characteristics of respondents are included in the analysis to identify factors for possible intervention and to act as control variables in the analytical models. These include age, highest level of education completed, place and region of residence, and religious affiliation. RESULTSDemographic ProfileA total of 1,831 never married (both legally and consensually) youth age 15-24 years, interviewed during the 1999 NDHS, are analysed in this study. The basic-demographic characteristics of the study population are provided in Table 1. Nearly 70% of them are 15-19 years of age, while the remaining 30.3% are aged 20-24 years. The average age of the respondents was 18.2 years. Sixty-one percent of the respondents reside in rural areas, although fairly equal numbers of them had childhood place of residence in city and the countryside (more than 40% apiece). Spatial distribution of the study population shows that about two in three of them reside in the south, while the remaining one-third reside in the northern part of the country. This lopsided distribution is a clear indication of early marriage in the north, which renders many female adolescents ineligible for inclusion in the sample. Data on educational attainment indicate that the majority (about 70%) of the youth under study have had some secondary education. Only 6% of them have tertiary education; slightly more than this proportion have no formal education at all. Christians constitute an overwhelming majority of the sampled population (76.3%); Muslims constitute 22.4%, and only 1.4% belong to indigenous religions. The result indicates that a large proportion of the study population live in poverty, as measured by available household assets (mean of 2.6, with a maximum of 7). Only one-third of the youth are categorized as coming from homes of high socio-economic status. Table 1 indicates that more than one-quarter of the adolescents come from the household where the floor materials are made of natural and rudimentary materials such as “dung, earth/sand, wood planks and palm/bamboo.”Furthermore, 25.3% of the respondents live in homes where there is no toilet facility, and the majority of them live in homes that use pit toilet. It is clear that the majority of the respondents come from home environments that can be described as poor. Sexual behaviourOverall, 31.5% of respondents had ever had sexual intercourse and the median age at first sex is 17 years. About 17 % of them had sexual relation in the month preceding the survey; among those who have ever had sex, about 54 % had intercourse during that period. These figures are consistent with those reported from the 1990 NDHS by Isiugo-Abanihe (1994). Table 2 displays the distribution of sexually experienced unmarried women, current sexual activity among the sexually experienced, and median age of sexual debut by selected socio-economic factors. Table 1: Percentage Distribution of Unmarried Female Youth age 15-24 by Selected Socio-demographic Characteristics
Source: Computed from 1999 NDHS Data - For this and Subsequent Tables CASES As would be expected, the proportions of respondents who are sexually experienced as well as those who had a relation in the month preceding the survey increase with age. For instance, whereas 21 % of unmarried female respondents age 15-19 years had ever had sex, about 56 % of their older sisters age 20-24 have done so. The corresponding figures for sexual activity in the past one month are 50.4 % and 56.5 % respectively. Table 2 also shows regional differences in the level of sexual experience, with respondents in the southern regions exhibiting higher levels of sexual activity, and initiating sex earlier relative to those in the north. To a large extent this finding is due to selection problem whereby those who are sexually exposed had been given away in marriage in the prevailing culture of early marriage in northern Nigeria. Rural residents tend to initiate sexual relation earlier than those in urban areas (with a two-year difference in median age of sexual debut), and are slightly more sexually experienced; however, current sexual activity is higher among urban residents. Childhood place of residence supports the notion of lower sexual activity among those who have some urban exposure. The folk life of the rural areas provides ample space for sexual mixing more than the busy and crowded urban environment. Table 2 clearly shows that increased educational attainment is associated with an elevated likelihood of being exposed to sexual relation. About three-fifths of respondents who have tertiary education had ever had sex, compared with 16.9 % among those who did not attend any formal school. However, the age at sexual debut is generally higher among those with higher level of education relative to their sisters with low levels of education. The finding indicates that female adolescents with no formal education started sexual activity three years earlier than those with tertiary education. It would seem that female youth who are not in school have more time and opportunity to engage in extracurricular activities that predispose them to early sexual relations. Level of access to media information is positively related with the level of sexual activity, although the median age of initiation of sex is inversely related to access to information. Respondents with low access to media information started sexual activity almost 2 years earlier than those with greater access to information. Table 2 also shows that household socio-economic status is inversely associated with sexual experience. First, those from low SES homes started sexual relation one year earlier than those from medium or high SES. By the same token, sexual experience tends to increase as SES declines. Current level of sexual activity, however, increases with SES, which may be symptomatic of the increasing abdication of parental care and responsibility by many well-to-do Nigerians in pursuit of professional careers, wealth and the so-called greener pasture beyond the shores of Nigeria, often with the children left behind. Table 2: PercentageDistribution of Unmarried Female Youth Who Have Ever Had Sex, Current Sexual Activity, and Median Age of First Sexual Debut by Selected Socio-economic Characteristics
We examined the effect of personal income on sexual relationship among those adolescents who earned some income. Even though the result should be taken with caution because of the characteristic incompleteness of income data as well as the small number of youths who earn income, the finding indicates a rising age of sexual initiation as income increases. For instance, adolescents with annual income of less than N6,000.00 started sexual activity at about 16 years compared with 18 years among those with annual income of N45,000.00 and above. Current level of sexual activity also suggests that those with lower income are more sexually active. The two other variables that depict the socio-economic status of the respondents, namely, the type of toilet facility and floor materials, support the general pattern of lower age of initiating sex by those who are less well-off. Condom UseDespite a high level of sexual activity among unmarried adolescent Nigerian women, the level of condom use amongst them is low, an indication of their high level of exposure to pregnancy and sexually transmitted infections. Only 15 % of unmarried female youths reported ever using condom; the reported level of condom use at last sexual intercourse is higher at 21.6 %. Results of the bivariate analyses of factors associated with condom use (Table 3) reveal multiple influences, with the background and poverty factors having more marked effects than they did on sexual activity. Table 3: Percentage Distribution of Unmarried Female Youth Who Are Have Ever Used Condom, and Current Use of Condom by Selected Background and Poverty Related Factors
The poverty-related factors (household socio-economic status and access to information) are positively associated with the two condom-use variables under examination. For instance, 24.5 % of the adolescents with a high level of access to information have ever used condom, as against 7.3 % of their counterparts with low access to information. About 20 % of the adolescents who are living in household with high SES reported ever used of condom while 12.8 % of their counterparts in low SES households had ever use condom. The results also indicate that significantly higher percentages of adolescents residing in households with modern toilet facility (flush system), and those whose houses have finished floor materials (tiles, marbles) reported higher condom use (see Table 3). The Table shows that the use of condom increases with age, education and exposure to an urban environment. Considering the fact that older and more educated youths are usually better informed about the consequences of unprotected sex, and the fact that accessibility to contraceptives is higher in the urban centres, the observed relationships between age, education, place of residence and condom use are not unexpected. It is also important to note that ever use of condom varies markedly according to the region of residence, being lowest in the north-east and north-west, even though the two regions recorded usually high level of current use, which is clearly related to the low cell size evident in the Table. It is noteworthy also that condom use among female Nigerian youth cuts across religious affiliation, with those who belong to the Roman Catholic denomination having the highest level of ever use and current use of condom. Given the opposition of the Roman Catholic Church to the use of condom, the above result is somewhat surprising, and suggests that younger members of the Church may not be strongly committed to the doctrines of the Church with respect to sexual conduct. LOGISTIC REGRESSION ANALYSISLogistics regression analysis was carried out to assess the independent effect of the poverty indicators and other socio-demographic variables on the sexual behaviour of the female youth. All the variables in the descriptive table were tried, but only those with some significant categories are retained in the final analysis. Table 4 presents the result of the logistic regression models that estimate the net of the poverty indices on sexual activities (past sexual experience and sexual activity in the previous four weeks). In model 1, we included only the principal explanatory variables, level of access to information and household socio-economic status, in order to determine their separate association with the sexual behaviour variables. In model 2, we included five background variables (age, education, place of childhood residence, religion and region) in order to determine the net effect of the poverty indicators when other variables are statistically controlled. The analysis revealed that respondents with moderate and high access to information are significantly less likely to have ever had sex compared with those who have low access to media information. For instance, compared to female youths with low access to information, those who reported moderate access to information are twice less likely to have had sexual relations. The effect remains consistent in model 2 after controlling for other variables that might have influence on the underlying relationships. The Table reveals that respondents’socio-economic status exerts significant influence on past sexual experience; those who are from high SES households are more likely to have had sexual intercourse relative to their counterparts from low SES household. The relationship remains consistent in model 2, with those from households of moderate socio-economic status being 83 % more likely to have been sexually exposed compared with those from low socio-economic status household. Of five socio-demographic variables considered in the model 2, three were significant at 5 % level or better; these include region of residence, education and religious affiliation. For instance, region of residence indicates that female youths from the north-east region are almost five times less likely to have had sexual experience relative to those from the south-east who are nearly as likely as those from the south-west. As stated earlier, this result may be affected by the high probability of self-selection in the north. Female youths with higher education are nearly five times more likely to have had sexual intercourse relative to those with no formal education. Protestants are less likely to have initiated sex compared with other Christians. Older female youths are more likely to have had sexual exposure than their younger counterparts. Place of childhood residence seems to have some influence on sexual exposure though the relationship is not significant. Female youths who were raised in the countryside are 26 % more likely to have had a sexual act relative to their counterparts whose place of childhood residence was the city. Table 4 revealed that the poverty-related and other socio-demographic variables are not significantly related with current sexual activity (sexual exposure in the previous four weeks), apart from education and childhood place of residence that show only marginal effect (at 10 % significant level). Nevertheless, the relationship between poverty-related factors and recent sexual activity is in the expected directions. Respondents from medium to high socio-economic status homes are less likely to be sexually active in the previous four-weeks relative to their counterparts from low socio-economic status household. Also, the result revealed that those with moderate to high access to information are less likely to have engaged in sexual activity in the previous four-weeks compared with those who have low access to information. Table 4: Odds Ratios for Regression Analysis Assessing the Likelihood of Ever Having Sex, and Having Sex in the Previous 4-weeks, by Poverty Indicators and Selected Characteristics
*p<.10. **p<.05. ***p<.01. Categories with Odds Ratio of 1.00 Are Used for Reference In Table 5, logistic regression analysis was also used to assess the independent effect of the poverty-related indicators on the use of condom. Of the two poverty-related variables in model 1, access to media information exerted significant effect on ever use of condom. For instance, compared with female youths with low access to information, those who reported high access to information are two times less likely to have ever used condom. Similar findings are also observed for the youth whose household socio-economic status was categorized as high (odds ratio of 0.76), though the association was not significant. The relationship remains consistent in model 2, with the effect of coming from high socio-economic status household being significant at 5 % level. Table 5: Odds Ratios for Regression Analysis Assessing the Likelihood of Ever Using Condom, Using Condom During the Last Sexual Relation, by Poverty Indicators and Selected Characteristics
*p<.10. **p<.05. ***p<.01. Categories with Odds Ratio of 1.00 Are Used for Reference The analysis in Table 5 further revealed that between the two poverty-related indicators, only access to media information exerted significant effect on the use of condom during the last sexual activity, though the relationship remain not in the expected direction. For instance, youths who have moderate access to information are three times less likely to have used condom at the last sex compared with their counterparts who have low access to information. Similar findings are observed in model 2. The analysis revealed that the household socio-economic status is not significantly related to use of condom at the last sexual intercourse. As expected, female youth from high socio-economic homes are more likely to have used condom at the last sexual act relative to those from low socio-economic status homes. Of the socio-demographic variables considered, current age, age at sexual debut, region of residence and religious affiliation are significant predictors of condom use at the last sexual activity. The analysis shows that current age and age at sexual debut are inversely related to condom use in the last sexual relation. Female youth from the south-west region are significantly less likely to use condom during the last sexual intercourse compared with their counterparts from the south-east, and those from the north-east region are more than two and one half % as likely, a rather counterintuitive finding given the predominance of the Muslim religion in the latter. The small cell size may have biased the result. The table also shows that female youth whose place of childhood residence was the city are more likely to have used the condom during their last sexual encounter, although the relationship is not significant. DISCUSSIONThe findings of this study confirm the high level of premarital sexual activity among Nigerian female youth. Age of initiation of sex remains low and current sexual activity is high, even though the personal and sensitive nature of questions on sexual activity makes them prone to deliberate misreporting. Also, the results suggest that a large proportion of the youth engage in risky sexual behaviour by not using the condom to protect themselves from infection with sexually transmitted diseases and in preventing unwanted pregnancy. The results reveal that initiation of sexual activity and use of condom varied considerably by a number of poverty-related and socio-demographic variables. The bivariate analyses suggest that sexual experience and age at initiation of sex are directly related with access to media information, implying that those from more affluent homes are more sexually active, even though they initiated sex at a later age. The other proxy measure of poverty (that is, household socio-economic status), is inversely related with sexual experience, and positively related with age of sexual debut. This finding is in the expected direction and supports the hypothesis that the poor not only start sex earlier but also are more sexually active than those from more affluent homes. However, the relationships fail to hold up in the multivariate analysis. Thus, there is little evidence that poverty, as measured here maybe imperfectly, is associated with sexual behaviour in Nigeria. On the contrary, the findings suggest that those who have access to media information and those from high socio-economic status homes are more sexually active than their counterparts who do not have access to media information or those from low socioeconomic homes. This finding supports the pattern of high levels of moral decadence among children from affluent homes which pervades university campuses in Nigeria as well as the major cities. In part, this may be explained by abdication of parental responsibility emanating from parental preoccupation with their professions and other pursuits of daily existence in the country. In fact, many professionals and wealthy Nigerian parents do not physically live with their children any more; many are overseas while their children remain in Nigeria, at best with their mothers or some other surrogate parents, otherwise they are alone; many parents have relocated to the new capital city of Abuja or other cities while their children remain in Lagos or the other cities from which they migrated. Perhaps as a way of compensating for their absence, such parents erroneously equate parental love and care with lavish provision of money and other durables for their children, including automobile, television, web-browsing facilities, mobile telephones, etc. The high level of sexual activity among such children can therefore be seen in the context of an opulent and permissive environment which favours the attraction and interaction of youths of opposite sex, and which, therefore, inevitably predisposes them to express themselves sexually. Although the findings of the present study do not support the view that the rising sexual activity is a function of household poverty, it has some important policy implications with respect to this. First, in the era of HIV/AIDS moral instructions are imperative in homes, schools and religious institutions, to enable the youth make responsible decisions with respect to relationship with members of the opposite sex. Second, the fact that the use of condom is very low even among those from affluent homes suggests that the promotion of condom use should be strengthened among sexually active youth with special emphasis on how to use them effectively and consistently as well as ensuring its availability and accessibility to potential clients. Behavioural change communication activities should be strengthened with extensive education on sexual abstinence and safe-sexual behaviour through culturally appropriate messages, as is currently undertaken by various NGOs in the country. In particular, more effort should be made to reach the youth in the countryside, who typically have earlier sexual debut and also tend to be more sexually active, with adequate information on behavioural change, prevention of unwanted pregnancy and protection from sexually transmitted infections, including HIV/AIDS. In addition, more holistic and realistic gender-sensitive programmes and interventions are needed to address the issues of gender inequality that is deeply rooted in the society, and which exposes the female youth unduly to sexual exploitation. Further research is required to establish how poverty and socio-economic status are related to sexual behaviour in Nigeria. It is problematic to define poverty where data on income are hard to collect, more so among the youth most of who do not earn income as such. Parental or household income status may not adequately and consistently reflect children’s well-being, since some wealthy parents may not liberally transfer money to their children, and some children have the adroitness of making money from sources other than parental sources. It is nevertheless important to research into and fine-tune non-income measures of children’s well-being within specific cultural contexts on the basis of which their sexual behaviour and other phenomena could be more realistically predicted. REFERENCES
Copyright 2004 - Union for African Population Studies | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||
