WOMEN'S POSITION, CONJUGAL RELATIONSHIPS AND FERTILITY BEHAVIOUR AMONG THE YORUBA

The aim of this paper is to examine, as part of the ongoing research, the influence of women's position on conjugal relationships within the home and how this, in turn, influences fertility behaviour. The society being studied are the Yoruba who live in the South Western part of Nigeria.

. A number of sociological and anthropological studies of African traditional and modern societies have discussed in detail women's position within the home and among kinship groups (Fadipe 1970;Fortes 1978).Fadipe, in his study of Yoruba society noted the low position accorded women relative to men in domestic, kinship and farm related activities.This was especially so for new wives in polygynous setting (Fadipe 1970).Studies of contemporary situations however reveal that changes have begun, to take place which have improved the status of women in such societies (Caldwell 1982;Omideyi 1987).

22.
A few other sociologists have also discussed the influence of type of marriage and education on conjugal relationships (Oppong 1974(Oppong , 1977(Oppong , 1987;;Lloyd 1967;Marris 1961).Findings of such studies reveal the existence of conflicts which individuals in monogamous marriages experienced (especially in matrilineal elite societies in Ghana) in having to cope with cqmpeting claims of their own kin, of spouses and of children on their loyalties and support.Probing further with data from focussed biographies in a polygy nous marriage diminished the status of a wife '... in terms of economic and social rewards, and emotional satisfaction', although it has some form of rewards for infertile and divorced women.Another interesting finding iri her study was that the highest incidence of polygynous marriage occurred among the highest educated women.Her explanation was that highly educated women who married late had narrowed options, forcing them to accept polygynous marriages, In the study, only few women (mostly G a women) pooled their incomes with, shared leisure pursuits or had assistance from husbands with housework or child-minding.More importantly, it was found that only few of the educated, married women had personal satisfaction from marriage.More satisfaction was gained from work, parenthood and kinship roles.Conjugal role deprivation, w hich was reflected in strained relation ships within marriage, was felt among the women either because they did not secure the kind of husband or the type of marriage they wanted or because of divorce or an unhappy current marriage (Oppong 1987).
The description of conjugal relationships among Ghanaian women may ba an eye opener to what obtains among the Yoruba, although there are a few differences between the tribes described and Yoruba women in terms of societal typology.The Yoruba are of patrilineal descent and a lot of impor tance is attached by wives to what the husband's family members think and want.Yoruba women believe that this is a way of securing their position as wives.In spite of these differences however, what comes out from various studies about conjugal relationships in developing countries is the emotional distance observed among couples.Spouses rarely show affection openly.They seldom spend leisure time together and evidence of joint decision making between husband and wife on a wide scale is lacking.Explanations may be found in the fact that couples who still depend on the wider/extended family for economic support may be physically, though not economically, nucleated and so may still be unable to make important decisions without the consent some members of the extended family.Also, strong kinship bonds can achieve the same effect even when economic nucleation has been achieved.Other cultural or sociological reasons may be responsible for the observed emotional distance e.g. the parlance among the Y oruba that "a woman's place ls .ln .
^ome may preclude a wife being seen outside with her husband enjoying leisure hours.Cultural barriers may also prevent show of emotion between spouses to outsiders (Caldwell 1977).
The focus of interest in this paper is the implication of this emotional distance for fertility in these countries.It is possible that the high fer tility recorded for them is a reflection of the kind of conjugal relationships existing.Emotional distance between couples could result in high fertility through mechanisms which can take two forms : a) the lack of show of emotion may be a cause of the desire for more children on the part of the wife as a means of attracting her h u s b a n d s affection more (because of the high value placed on children), or as a means of finding alternative emotional outlet.b) the lack of show of emotion may be the effect of the large family size already had by the couple.Due to the demands of the large number of children which the wife has to cater for, more time is spent on the children and less on the husband, especially if, in addition, the wife is fully or partially engaged in some form of occupation.This may cause a drifting apart or loosening of emotional ties.
Since it is difficult in a high fertility oriented society to separate emo tional distance as a cause or as an effect of high fertility as described above, the position of the wife in terms of her educational and work status will be used to create the dividing line.Hence it is expected that educated and working women who are presumed to have closer conjugal ties will have fewer children than uneducated and non-working women who are presumed to have greater emotional distance between them and their spouses.(See schematic diagram below).

Indicators of
Conjugal Fertility Women's Position Relationships Two other status variables will be used to examine the relationships between women's position, conjugal n-iati^nchipc and fertility among Yoruba women.These are residence status and type of marriage.The assumption is that women who live in urban areas and women who are in monogamous marriages will ex p e rie n c e closer conjugal relationships and hence have fewer children than women in rural areas and women in polygamous unions.In other words, as !he diagram shows, conjugal relationship in the form of closeness or 'jointness' in decision making between husband and wife is important in the female slatus-iertili y relationsbip.
The data used for the study derived from information collected from 2,170 currently married women and their husbands during a survey on women's status and fertility among the Y oru ba in 1986.The types of information collected related to issues of marriage, fertility and work histories, wife's purchasing power, role activity and status change, husband/wife decision making, knowledge, attitude and practice of contraception among others.The survey location was Oyo State and samples were drawn from two towns -Ibadan and Ilesa -and four intervening rural areas namely Ipetumodu, Ode-Omu, Asejire/Ikire and Moniya.A stratified random sampling procedure was used to arrive at the final set of respondents.Three sets of questionnaires were administered using the direct interviewing method.These were the household, the female and the male questionnaires (Omideyi 1985).In addi tion, a subset of the women were studied in-depth to give an insight into their own perceptions about their position in relation to their husbands and in the society as a whole.The main body of data are used in the analyses while information from the focussed biographies will be used to corroborate the findings.
The variables considered as indicators of status in this study are education, employment status, urban/rural location and type of marriage.Many more indicators have been identified (Dyson and Moore 1983;Mason 1984) but restriction is necessary here to preserve focus.The merits and demerits of each cjhosen indicator as a good measure of status are also not discussed here since this has been discussed elsewhere (Omideyi 1985, Jeejebboy 1985, Omideyi 1988).Variables indicating "closeness" between husband and wife were computed from responses to questions about wife's habitual eating, sleeping, having leisure together, sharing confidences/secrets and holding discussions with her husband.For example, to the question "Do you usually eat together with your husband ?", a "Y es" will be scored 2 and a "No" will be scored 1.Hence, to the five questions the maximum score for each woman will be 10 and the minimum 5.The total score for each respondent is then divided by 5 to arrive at the average.Indices indicating "jointness" are an ordered information on w ho usually decides on certain selected issues namely where to live, whether wife should work, choice of family friends, how many children wife should have and whether couple should practice family planning.Responses and scores were: no one decides/relatives (0); I always decide (1); I decide more than my husband (2); Both of us decide (3); My husband decides more than I (4); and My husband always decides (5).
It is important to point out that the indices of closeness and jointness relate to behavioural questions which may be influenced to some extent by what is culturally permitted or expected.In this case, however, it is hoped that the manner of computation should give a balance of opinion between the educated w ho are likely to give candid responses, and the non-educated who may be more prone to adherence to expected cultural norms.

FINDINGS Background Information on coniugal relationships
Based on information on the proportion of all women who reported that they usually ate, slept together in the same room, had leisure, shared confidence-/secrets and held discussions a lot with their husbands, it is clear that educated women usually have closer conjugal relationships with their husbands than uneducated women (Table 1).The reason for this may be the fact that educated women are more likely to be married to educated men, making it easier to enjoy conjugal relationships because of their common exposure to modern facilities for relaxation and their common appreciation of close husband/wife interaction.Slightly higher proportion among the uneducated (89.9%), however, report that they shared confidences with their /husbands.
In the case of employment status, there seem to exist closer conjugal relationship among unemployed women than those who are employed, except in the area of sharing confidence.It may be reasonable to expect this pattern because the time for playing the conjugal role for most working women is taken up mostly by work activities.
Women in urban areas and monogamously married women also tend to have closer conjugal relationships with their husbands.But rural women a.id polygynously married women shared more confidence with their spouses.
Overall there seems to exist a compensatory mechanism whereby inability to do things together with spouses results in sharing more confidence with them.
With respect to jointness in decision making, again more educated women seem to take joint decisions with spouses than uneducated women, especially on issues relating to number of children wife should have and whether wife should practice family planning.Working women also take more joint deci sions with spouses than non-working women, especially on the issue of whether wife should work.Greater jointness in decision making on the selected issues are also reported for women in urban areas and among women in monogamous marriages, except that in rural areas and among women in polygamous marriages, more joint decision is taken on wife s employment intentions than in urban areas and in monogamous marriages.It will be noticed that the proportion of women reporting jointness on any specific issue does not exceed 61% which is for educated women on how many ch1dren wife should have.This shows that decision making is not usually joint to a large extent among Yoruba couples.From information gathered it was found that next to joint decision making on issues, the husban d was influen tial in deciding on where to live, children's school and whether wife should work while relatives were next in importance in deciding how many children to have and whether to practice family planning.Apart from joint decision by spouses, decision on choice of family friends was equally shared y husband and relations.

Effect of conjugal relationships on fertility
Correlation coefficients between indicators of closeness and mean children ever born are shown in Table 3.The values are negative and significant below 99% confidence interval (C.I.) for indicators of closeness between spouses.Among indicators of jointness, however, correlation coefficients are significant below 95% C.I. only in the case of decision about children's schooling (where the relationship is positive) and whether wife should practice family planning (where the relationship is negative).Examination of the values tend to confirm that where conjugal relationships are close, there is a tendency for fertility to decrease.The negative relationships are strongest where spouses usually sleep together in the same room, share confidence and hold discussions.
Effect of Indicators of women's position on "Closeness" and "Jointness" Indices of closeness were computed by combining scores on eating, sleeping, sharing confidence and having leisure with husband and then finding the mean.Indices of jointness were likewise computed for joint decision on where to live, children's school, whether wife should work, choice of family friends, how many children wife should have, and whether couples should practice family planning.The beta coefficients in Table 4 were derived from multiple classification analysis, showing the extent to which indicators of women's position determine closeness and jointness in decision making.
-* As already shown above, there exists greater closeness the higher the educational level, among non working women, among women in urban areas and among monogamously married women.Jointness in decision making (as measured by the indices) however decreases with higher education after primary education level has been attained.It reduces for working women compared with non-working women, increases for urban women over rural women and is greater among women in polygamous than in monogamous marriages.An examination of the beta coefficients show that education is the most powerful determinant (beta coefficient .34) of closeness followed by residence in urban or rural areas.Education also has the strongest, though weak, influence (beta .07) on jointness in decision making by spouses.Overall, the multiplier square values show that indicators of women's position are more important in determining "closeness" between spouses than "jointness" in decision making.

Women's position, conjugal relationships and fertility
Improvement in women's position, in general, is expected to reduce fertility.
A number of studies have shown that although this is so, the mechanisms of achieving the reduction are not as clear as one would wish especially if proximate determinants are considered as the main intervening variables (Jeejebhoy 1985;Omideyi 1988).In this study, proximate determinants have been ignored because the aim is to explore the relationship between women's http://aps.journals.ac.za position and fertility, using conjugal relationships as the main intervening variable.Table 5 shows the effect of controlling for "closeness" between husband and wife in the relationship.
When "closeness" is not controlled for, of all the indicators of women's position, education is the most powerful determinant of mean children ever born with beta coefficient of .12,and a 17.1% reduction in fertility between secondary and University levels.Other indicators were not significant in determining children ever born.The predictive power of the indicators of women's position in determining fertility as revealed by the multiple r square value is .16.When "closeness" is cohtrolled for, the influence of education in determining CEB is reduced as expected since education has already been shown to be powerful in determining closeness.But no dif ference occurs to the beta values of other indicators and yet the predictive power of the indicators of women's position on fertility becomes greater (r square = .20).Moreover, the mean values of CEB for secondary level reduces by one point and that of university education level increases by one point, showing a lower percentage reduction (12.5%) in fertility o fthe two groups.
Hence, in the absence of closeness between spouses fertility is likely to converge to a higher vklue for women with post secondary education level.
The suggestion here is that for Yoruba women, emotional distance is in part responsible for tjie high fertility experienced: women who are uneducated and who exhibited less closeness and jointness in husband/wife relationship show higher mean CEB than the highly educated.Also, where closeness to be controlled for, fertility of the highly educated would increase.
Evidence that some women do have children in order to attract their hus band's attention is revealed in a statement by a respondent who was a grade two teacher in the focussed biographies.In reply to a question on why she wanted children she said: "... if a woman is with a man and she doesn't have a child, it is as if they are just together and there is no love.So the reason I want a child is for our (husband and self) love to grow more.
The role of conjugal relationships in determining fertility among Yoruba women in different status groups was assessed.The study revealed closeness between couples with respect to educated, non working, urban based and monogamously married women and jointness in decision making among the same groups except non-working women.Of all the indicators of women's position, education was most powerful in determining "closeness", and "jointness" in decision making.The fertility of the highly educated was lower than for the uneducated.When closeness was controlled for, fertility of the highly educated converged to higher values, and indicators of wo men's position became more important in determining fertility.Hence while emotional distance may be responsible for the high fertility of women of low status (education), absence of "closeness" would definitely lead to increases in the lower fertility of the highly educated. http://aps.journals.ac.za

Fig. l .
Fig.l.Schematic diagram showing connection between women's position, conjugal relationships and fertility in a developing country.

TABLE 2 :
Proportion of all women w ho make joint decision with husband on selected issues by indicators of women's position, Yoruba women, 1986 H u s b a n d a n d w i f e d e c id e t o g e t h e r o n : http://aps.journals.ac.za

TABLE 3 :
Pearson's Correlation Coefficients between mean Children EverBorn and indicators of "closeness", and "jointness" in decision making between husband and wife, Yoruba women, 1986.

TABLE 5 :
Effect of indicators of women's position on mean Children Ever Born a) controlling for age b) controlling for age and index of "closeness" between husband and wife, Yoruba women, 1986