National Academies Press: OpenBook

Factors Affecting Contraceptive Use in Sub-Saharan Africa (1993)

Chapter: 6 Regional Analysis of Contraceptive Use

« Previous: 5 Family Planning Programs and Policies
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 170
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 171
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 172
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 173
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 174
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 175
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 176
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 177
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 178
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 179
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 180
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 181
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 182
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 183
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 184
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 185
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 186
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 187
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 188
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 189
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 190
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 191
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 192
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 193
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 194
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 195
Suggested Citation:"6 Regional Analysis of Contraceptive Use." National Research Council. 1993. Factors Affecting Contraceptive Use in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2209.
×
Page 196

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

6 Regional Analysis of Contraceptive Use The preceding three chapters describe the main factors affecting contra- ceptive use in sub-Saharan Africa. In Chapter 3, several socioeconomic factors are shown to be associated with high fertility: low levels of female education and income per capita, rural residence, and high infant and child mortality. Although the associations between fertility and these socioeco- nomic characteristics are not always as strong as in other regions of the world, they do suggest that changes in these characteristics would have some effect on fertility, as in Botswana, Kenya, and Zimbabwe. We also suggest that changes in costs of living, due to economic reversals, may increase the acceptability of smaller family sizes in certain population sub- groups, but the response to such changes is very much a country-specific matter. In Chapter 4, we identify the aspects of African social structure that support high fertility norms, specifically, the importance attached to per- petuating the lineage, the linkage between number of children and access to resources, the use of child fostering to spread the costs and benefits of having numerous children, and the weak conjugal bond. Although these social characteristics are changing, it is clear that they have exerted consid- erable positive influence on historical fertility levels and have not disap- peared today. In addition, the penetration of major religions (Christianity and Islam) has affected contraceptive use, primarily through female educa- tion and social organization, such as marriage patterns. The contribution of family planning policies and programs to changes 170

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 171 in contraceptive use is discussed in Chapter 5. The changes in policy to promote family planning, as well as expansion of programs, have increased access to modern methods of contraception, fulfilling latent demand and possibly creating additional demand. This chapter uses statistical analysis to examine the relative importance of those factors that can be measured by using World Fertility Survey (WFS) and Demographic and Health Survey (DHS) data. Whereas Chapter 2 high- lights bivariate relationships based on individual-level data, this chapter examines these relationships using multivar~ate analysis and regional-level data. The factors described in earlier chapters-socioeconomic develop- ment, patterns of social organization, the influence of major religions (Christianity, Islam), and access to family planning services clearly differ among re- gions within a given country. Our analysis is based on the WFS and DHS data sets described in Chapter 2. The countries included are shown on a map in Figure 6-1, and a list of the weighted and unweighted sample sizes for each country is pro- vided in Appendix B. The sample consists of 37 regions in the WFS data set, 55 regions in the DHS set, and 92 regions in the pooled (WFS and DHS) set. The full set of WFS and DHS data considered are shown in the appendices to this chapter. We are limited by the variables collected under the two survey pro- grams.i Accordingly, we focus on four variables that were shown to have a marked effect on patterns of reproduction in an earlier analysis by Lesthaeghe (1989b): two socioeconomic variables-the level of female schooling and the degree of urbanization and two variables that reflect social organiza- tion the extent of polygynous marriage and the proportion Muslim. Al- though by no means exhaustive measures of socioeconomic development and community/kinship relations, these four variables nonetheless capture important dimensions of the contextual variables discussed in earlier chap- ters. This current analysis confirms the relationships documented in the earlier analysis (Lesthaeghe, 1989b), which used a smaller, less diverse data set based primarily on the WFS. The WFS data were collected mainly in countries in West Afnca. Notably lacking in this analysis is a measure of the strength of the family planning supply environment.2 Although national-level indicators 1It should be stressed that socioeconomic variables for which data was not collected under the WFS and DHS, such as income, are plausibly of importance in explaining differentials in contraceptive use. 2In addition, infant and child mortality were not included in the analysis. Because mortality may be influenced by high fertility or vice versa, we decided that it was not statistically sound to include mortality as an explanatory variable. We did explore this option, however, and found that mortality showed no significant independent effect on fertility, possibly because

72 Mali )= n I - ~ / ~= ~_ / \ FA CTORS AFFECTING CONTRACEPTIVE USE ~ Sudan ~ a_ ~ ~ ~ = | ~ 2N Ugan\~~ J Ghana Togo I - Benin _ _ ..... Nigeria · WFS REGIONS ° DHS REGIONS ~ WFS + DHS x5' Kenya ~ Burundi Zimbabwe I f$~ l IJ opt ~ Botswana / Lesotho FIGURE 6-1 Location of countries and regions that participated in the World Fertility Survey and Demographic and Health Survey programs. are available to measure this dimension, there are no analogous measures at the regional level that would reflect local differences in political commit- ment, access to services, variety of methods available, and related aspects of the family planning supply environment. This major shortcoming should be addressed in future research, especially in light of the evidence presented in Chapter 5 on the progress in improving family planning services in Af- nca.3 female education influenced both contraceptive use and mortality. Although we highlight the important positive association between mortality and fertility in Chapter 3, we cannot defini- tively test mortality's relative effects on contraceptive use in this statistical analysis. 3The multivariate model described below was also tested with a national measure of family planning effort (using scores developed by Mauldin and Ross, 1991). The t-statistic for the positive coefficient was 1.93. There is too little variation in national-level scores to make this an effective measure to explain regional differences. By introducing family planning effort measured at the national level, we are probably underestimating its true effect. There is a great likelihood that family planning has had a significant role to play in increasing contraceptive use in sub-Saharan Africa, but its contribution is impossible to assess precisely.

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 173 In anticipation of the results of the multivariate analysis, which indicate the overriding importance of female education with respect to contraceptive use, we begin our regional analysis by focusing on the bivanate association of education with contraceptive practice. FEMALE EDUCATION AND CONTRACEPTIVE PRACTICE As we demonstrate in Chapters 2 and 3, education is positively associ- ated with contraceptive use at the individual level and negatively associated with fertility at the national level. In our consideration of the relationship between female education and contraceptive practice, we examine not only modern contraceptive use, but also its precursors: ideal family size and knowledge of contraceptive use. Ideal Family Size Figure 6-2 depicts the relationship between the average length of schooling in the WFS and DHS regions and the percentage of married women with ideal family sizes not exceeding four children. One measure of association between two variables, the Pearson correlation coefficient, r, indicates the expected positive relationship (r = .73~.4 The percentage stating a prefer- ence for four or fewer children, here called "small ideal family size," com- monly rises above 30 for regions with schooling averages of four years or more, and above 50 for several regions in Kenya and in urban centers of Botswana and Zimbabwe. By contrast, very few (about one-fifth) of the low-education regions reach the level of 30 percent stating small ideal fam- ily sizes. Furthermore, these tend to be national capitals Khartoum, Lome, Cotonou, Bamako indicative perhaps of an effect of urban residence. Areas with fewer than 10 percent of women stating small ideal family sizes have been concentrated predominantly in West Africa. Analysis of the WFS showed that such low percentages were observed in most of Cute d'Ivoire, Nigeria, and Cameroon. In Ghana and Benin, such low percent- ages were found in the northern regions. Also the WFS data for Senegal show percentages around 10 for regions other than Dakar and Thies. The DHS data confirm the persistence of these low percentages of women with small ideal family size in West Africa. The 15 percent level is not ex 4The Pearson correlation coefficient, r, is often used to measure the extent of a linear relationship between two variables. A positive r indicates that the slope of a regression line fit using data on the two variables is positive. (A negative r indicates that as one variable increases, the other decreases.) When r is statistically different from zero, it is determined that there is an association between the two variables. /

174 He . . 90 B0 70 ~ 60 a) .,. En ~40 ~ ., - A) H 50 30 20 10 O - FACTORS AFFECTING CONTRACEPTIVE USE o o o o o o o of o o o o o ~ :~ o ~ a o° o o o o o o ~ o o Wo /°o ° o o o o oo o / o o o o l 6 7 ~9 I 1 1 1 1 0 ~2 3 4 5 Education: years FIGURE 6-2 Relationship between percentage of currently married women ages 15- 49 with ideal family sizes of four or fewer children and mean number of years of female education WFS and DHS regions (~--.73~. SOURCE: Demographic and Health Survey and World Fertility Survey reports. ceeded in Northern and Upper Ghana; the Since and Grand Gedeh regions of Liberia; the Mopti, Gao and Tombouctou, Kayes, and Koulikoro regions of Mali; in most of Senegal; in Ondo State (Nigeria); and in the Savanna region of Togo. The only regions outside West Africa with such low per- centages are the Kordofan and Darfur provinces of Sudan. Knowledge and Use of Modern Contraceptive Methods Knowledge of modern methods of contraception is also strongly related to average female schooling levels in the ' pooled data set, as shown in Figure 6-3 (r = .77~. At an educational average of less than four years, knowledge of at least one modern method among women in a union varies widely from close to 0 to almost 90 percent. Regions with schooling aver- ages of more than four years generally exhibit knowledge levels of 50 percent or better.5 SScatter plots relating average female schooling levels to urbanity (positive effect) and to the

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 100 90 ~0 . . to A) lo: c Go ~0 A 70 60 50 40 20 10 O 175 o o o CD oO ° ° / o a/ o/° o o ~o on o oo oo Oo io° o o~ o o o~ o V o o ° o° oo o oo 69 - ko ° ° no ° ° 8 Oo ° O ~ a o o a I I ~I 1 1 ' 4 5 6 7 ~9 1 1 1 1 0 1 2 3 Education: years FIGURE 6-3 Relationship between percentage of married women ages 15-49 who know at least one method of contraception and mean number of years of female education WFS and DHS regions (r = .77~. SOURCE: Demographic and Health Survey and World Fertility Survey reports. Figure 6-4 displays the relationship, in the pooled data set, between the use of modern methods and the knowledge of such methods. The relation- ship is strong (r= .80), but its pattern deviates markedly from linearity. Where the regional knowledge level is less than 80 percent, the use of modern methods remains less than 10 percent among women currently in a union. Only at very high levels of contraceptive knowledge is there a sharp increase in the use of such methods. As indicated in Figure 6-3, knowledge levels of 90 percent or better emerge only in regions with female schooling averages of four years or more. As a consequence, one should expect levels of current use of modern contraception in excess of 15 percent to emerge only in regions with both high knowledge levels (80 percent or more) and high female educational levels (schooling averages of more than four years). proportion Muslim or adhering to traditional religions (negative effect) also show strong rela- tionships.

176 50 45 40 ~ . . o a' ~25 Q' o ~5 30 20 a' 15- 10 O - FACTORS AFFECTING CONTRACEPTIVE USE of 1 o / o lo (;D /o ° o o 9 o used,,, ~ ~0 ~ 0w ~ O ~ O / TOO o V O oOO oO O o O o o o ~I I I I I I I ~~~-r--~- rl 0 10 20 30 40 50 60 70 80 90100 Know One Method: % FIGURE 6-4 Relationship between percentage of married women ages 15-49 using modern methods of contraception and percentage of married women knowing of at least on modern method WFS and DHS regions (r = .80~. SOURCE: Demograph- ic and Health Survey and World Fertility Survey reports. This expectation is borne out in Figure 6-5, which shows the link be- tween the percentage of users of modern methods and the regional female schooling averages in the WFS and DHS regions. Contraceptive prevalence greater than 10 percent is virtually never reached in regions with mean female education durations of less than four years. Beyond this schooling level, the scatter widens considerably and the average percentage of users rises much more rapidly. An early inkling of increases in contraceptive use in the regions with more female education can be found among the WFS data for the late 1970s and early 1980s, as indicated by the circles in Figure 6-5. The additional information gathered by the DHS in the late 1980s, shown as triangles, further confirms this relationship. It should be stressed, however, that the regions that score highly on both modern method use and female education stem largely from Zimbabwe and Botswana (see Figure 6-6, which graphs the same data points as Figure 6-5, but in this case, the circles represent regions of Zimbabwe and Botswana). Conversely, Islamic regions (more than 75 percent Islamic or traditional religions) contribute disproportionately to the set of WFS and DHS regions

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 177 o WFS ~ DHS 50 45 40 . . ~35 o a' 30 C 25 a' o 20 C 15 . - 10 5 - O - 2 l\ An :///'a 0~ ~ 0 no A ^~ at 0 o / 0 o 1 1 1 1 1 6 7 ~ 0 1 2 3 4 5 Education: years FIGURE 6-5 Relationship between percentage of married women ages 15-49 using modern methods of contraception and mean number of years of female education- WFS and DHS regions (r = .80~. SOURCE: Demographic and Health Survey and World Fertility Survey reports. characterized by the combination of low female schooling and low modern method usage (see Figure 6-7, circles). The change in ideal family size and contraceptive knowledge and use can be demonstrated for the 1980s in countries that participated in both the WFS and the DHS, as shown in Table 6-1. In Kenya, knowledge levels were very high in 1978 (greater than 80 percent), and this knowledge base in combination with a rapid rise in the percentage of women preferring four or fewer children is reflected in substantial increases in users of modern methods between the two surveys. Knowledge levels in 1979-1980 were much more heterogeneous across regions in Ghana than in Kenya, and the increase in proportions with small ideal family sizes during the 1980s is more modest as well. Between the two surveys there was no change or even a decline in modern contraceptive use in Ghana. Hence, these two countries, both of which started after their independence with relatively high educational levels for women, have followed divergent paths with re- spect to family planning success. Senegal and Northern Sudan are typical examples of Sahelian Islamic

178 FA CTORS AFFECTING CONT^CEPTI VE USE o Zimbabwe and Botswana ~ Other regions 50 45 a ~40 . . 35 an 30 25 Q) to Cal .,. ~10 20 15 _ O o o o o ~To / A ~A A O ~ O O O / o a /o / ~a ~ d~ ~ ~ ~ .tys ~^ 1 1 1 1 1 1 0 1 2 3 4 5 6 Education: years l 7 ~9 FIGURE 6-6 Relationship between percentage of married women ages 15-49 using modern methods and mean number of years of female education (r = .80~. SOURCE: Demographic and Health Survey and World Fertility Survey reports. societies with lower levels of female education. In both countries there was a noticeable increase during the 1980s in the knowledge of modern methods of contraception. Several areas, other than the capitals, now reach knowl- edge levels between 70 and 80 percent. Yet the proportion of women with small ideal family sizes has hardly changed, and the increase of modern contraceptive use remains insignificant. In the regions of northern Sudan, a decline in current use of modern methods may have occurred as well. In summary, we find that: · there is a strong relationship between the regional levels of female schooling and the proportion of women who have small ideal family sizes, who have knowledge of modern contraception, or who use a modern method (refer to Chapter 2 for individual-level confirmation); · current use of modern methods increases above the 10-percent level only in regions that have a mean length of female schooling of four years or more; and · these conditions have been met in Zimbabwe and Botswana, as well as several regions of Kenya. In these areas, there has been a corre

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 179 o Regions >75% Islamic ~ Other regions 50 45 40 . . to o rat a) o en c .,' ~10 35 30 25 20 15 5 - O - ret ,\ l\ / / ~ o ^^ / ran o ~c~ a I, ~ I ~ ~ ~1 1 1 1 01 2 3 4 5 6 7 8 ~ Education: years FIGURE 6-7 Relationship between percentage of married women ages 15-49 using modern mettods and mean number of years of education WFS and DHS regions (r = .80~. SOURCE: Demographic and Health Survey and World Fertility Survey reports. spending rise in actual contraceptive use. However, there are several areas in other countries with a mean length of female schooling of four years or more (e.g., Imbo province in Burundi, all regions in Ghana except the two northern ones, Montserrado in Liberia, Ondo State, Khartoum, and Kampala) in which current use has remained low or may have decreased in the last decade (Ghana and Sudan). This diversity of experience within regions with relatively high female schooling levels suggests that other variables may be equally or more important in influencing contraceptive use. All of the regions that would be expected to have a prevalence of modern methods of at least 10 percent and did not were located in countries with weak family planning programs, which suggests that contraceptive supply may be a factor. MULTIVARIATE ANALYSIS OF MODERN CONTRACEPTIVE USE We now consider the relationship between female education and con- traceptive practice at the regional level in a multivariate model. We present

180 FACTORS AFFECTING CONTRACEPTIVE USE TABLE 6-1 Comparison of Ideal Family Size, Modern Contraceptive Knowledge, and Modern Contraceptive Use for Women Currently in Union in Sub-Saharan Regions Included in Both WFS and DHS Surveys (percent) Ideal Family Know at Least One Use Modern Size ~ 4 Modern Method Method Region and Date WFS DHS WFS DHS WFS DHS Kenya (1977-1978, 1988-1989) Nairobi 32 79 93 95 19 28 Central, Eastern 16 68 92 94 9 25 Rift Valley 12 50 81 85 5 18 Coast 15 36 83 92 5 15 Western, Nyanza 17 49 88 92 3 10 Ghana (1979-1980, 1988) Central, Western 27 33 76 79 5 4 Greater Accra, Eastern 37 51 81 90 11 8 Volta 35 46 62 78 6 4 Ashanti? Brong Ahafo 30 45 59 80 9 6 Northern, Upper ~7 13 40 1 1 Senegal (1978, 1986) Central 11 11 13 70 0 1 Northeast 12 11 16 39 0 1 South 7 13 14 54 0 2 West (Dakar, Thies ) 14 26 38 86 2 6 Sudan (Northern, 1978-1979, 1989-1990) Central 37 20 63 80 7 4 Khartoum 41 41 82 96 1 ~16 Kordofan, Darfur 21 11 27 45 2 1 North, East 28 21 50 74 3 4 the results of multiple regressions applied first to the DHS data alone and then to the pooled VVFS and DHS data (see Tables 6A-1 and 6A-2~. The least squares regression model for the pooled data is presented in Figure 6-8. (The results for the DHS data alone are similar to the results for the pooled data set.) In these models we consider four independent vari- ables: . the average length of schooling for women ages 15-49, · the presence of a large and dominant urban area in a region (dummy variable), · the percentage of married women ages 15-49 that are in a polygy- nous union, and the proportion of respondents that are Muslim or subscribe to traditonal . . · . religions.

REGIONAL ANALYSIS OF CONTRACEPTIVE USE -.20 | % Muslim |. + 02 ~~ Urban ~-.05 + +.22 \ /~\\ / ,/~ i~| XYears education ~3 ~1 \ / ., Dependent variable Y 181 -.03 .__ · · Years since 1978 · ~5 Dependent Variable y l3l 132 03 04 05 R adj US % Use Modern Method +.6sa _.2Sa +.12 -.06 0.16b .62 .62 % Know 1+ Modern Method +.S4a _.20a -~03 -~04 0.29a .64 .60 % Ideal Family Size <4 + 42a _ 24a +.01 +.1sb +.34a .61 .62 x Breastfeeding _ 60a + 12 _ 26b -.15 +.1gb .38 .79 x Amenorrhea -.68a +.06 -.13 _ 20b + 29a 54 .68 x Postpartum Abstinence -.10 + 54a _.36a _ Do -.12 .32 .82 x Postpartum Nonsusceptible _.54a +.32a _.33a -.08 +.10 .43 .75 aSignificant at 01. bSignificant at 05. FIGURE 6-S Regression model for various demographic dependent variables- WFS and DHS regions (n = 924. In the regressions based on both the WFS and the DHS data' a calendar time variable (years since 1978) has also been introduced to allow for varia- tion with respect to year of measurement. It should also be noted that the regions of Senegal, Kenya, northern Sudan, and Ghana contribute two data points in the pooled WFS and DHS sample. These measurements are ap- proximately 10 years apart. Relationships Among Background Variables There are a number of relationships among these four background vari- ables that have implications for the subsequent analysis. Urbanity and the proportion Muslim have important but obviously op- posite associations with the level of female education. These associations

182 FACTORS AFFECTING CONTRACEPTIVE USE were found in both the DHS and the pooled regional samples: The effect of urbanity on education, measured by standardized regression (beta) coeffi- cients, is +.47 and +.41, in the respective samples; by contrast, the effect of the proportion Muslim is -.49 and -.51, respectively. When we tested the relationship of the proportion Muslim, urbanity, and female education to the percentage of married women in polygyny in both data sets, only the negative effect of female education was significant and large: beta coefficients are -.50 and -.62, respectively. Other factors, not specified in the model but subsumed in the residual variable U2, exerted a major influence on the incidence of polygyny (effects of U2 are .87 and .86~. Polygyny is associated positively with other features of sub-Saharan culture and social organization (e.g., involvement of women in trade and survival of female secret societies or of institutionally structured economic and political networks for women; see Chapter 4~. Also the negative asso- ciation between polygyny and matrilineal systems of descent is another example of a historical connection that is not explicitly considered in the model. The polygyny variable can therefore be considered a proxy for a broader spectrum of features of social organization that is particularly rel- evant for the West African regions in our samples. Relative Importance of These Background Variables in Contraceptive Practice Of the four background variables included, level of education had by far the strongest effect (~) on modern contraceptive use in the two analyses (0.71 and 0.65~. Next in influence was the percentage of women in po- lygyny, which was negatively related to modern contraceptive use (B2) . The two other background variables urbanity and the proportion Mus lim influenced level of schooling, but had little or no direct effects on modern contraceptive use (although urbanity is significant in the individual- level analysis of contraceptive use in Chapter 2~. Similar patterns held when the knowledge of a modern method of con- traception and small ideal family size (i.e., less than five children) were regressed on the background variables. The effect of female education is again strongest (+0.54 and +0.42, respectively). The effect of polygyny remains negative, as expected. The proportion Muslim has no effect after controlling for female education.. Urbanization (i.e., living in a major city) has a positive effect on smaller family size ideals, but not on the knowledge of at least one modern method.

REGIONAL ANALYSIS OF CONTRACEPTIVE USE Conclusions 183 These results for the pooled WFS and DHS sample are strikingly simi- lar to the regression results of the WFS regional sample (Lesthaeghe, 1989b). However, there are no observations in either analysis for central Africa, or for regions that were in the past strongly affected by high levels of infecundity (e.g., Gabon, southern Sudan, northwest Zaire, and the Central African Re- public). In addition to this regression analysis, we also tested these relationships using LISREL models6 applied to the two data sets (results not shown).7 Those results confirmed the role of female education in the adoption of contraception. However, in the context of sub-Saharan Afnca, it is essen- tial to examine the dynamics of contraceptive use in the larger context of other birth spacing practices. In the next chapter we analyze the relative contribution of modern contraceptive use versus the traditional postpartum means of child spacing to changes in fertility. 6LISREL is a statistical package that allows for the construction of complex variables on the basis of highly correlated indicators. In this analysis, LISREL allowed for the simultaneous modeling of all the dependent variables presented in Figure 6-8 and gave a measure of the goodness of fit of the model. The results from the LISREL models are presented here to confirm the results of our simpler analysis. 7In the LISREL model, we assumed a causal linkage among a series of demographic vari- ables including length of postpartum nonsusceptible period, small desired family size, knowl- edge of modern contraception, and current contraceptive use. The level of female education was then measured as the common component of the average female schooling duration and the percentage of illiterate women aged 15-49 in the regions. Finally, the percentage of users of modern contraception was complemented by a broader index of use-effectiveness (i.e., a weighted average of the method-specific use-effectiveness levels with the weights equal to the proportion of women using a given method; adapted from Bongaarts and Potter, 1983). The two LISREL models were characterized by an excellent goodness of fit, uncorrelated residuals, and a strong similarity for the two data sets between the respective sets of standard- ized coefficients.

184 TABLE 6A- 1 WFS Regional File-Part A FACTORS AFFECTING CONTRACEPTIVE USE Married Married Use Contraceptive Who Know Who Want Modern Use- at Least 1 ~ 4 Method Effectiveness Modern Children Region (%) Index Method (%) (%) Benin Atacora, Borgou 0.6 2 5.4 5.9 Central, South 0.4 4 10.6 12.0 Cotonou 4.1 8 52.4 36.1 Cameroon Center-South, East 0.6 3 48.9 11.4 Littoral, Southwest 1.3 1 38.7 2.6 North 0.1 0 6.4 18.4 West, Northwest 0.1 1 22.3 2.7 Douala, Yaoundd 4.9 6 62.9 6.8 Cole d'lvoire Abidjan 1.9 2 34.3 9.7 Rural forest 0.1 0 13.5 5.1 Rural savanna 0.1 0 4.0 4.9 Urban forest 1.2 2 25.5 9.4 Urban savanna 0.8 1 21.2 10.2 Ghana Central, Western 5.3 4 75.5 26.7 Greater Accra, Eastern 11.3 9 81.4 37.0 Volta 6.0 6 61.9 35.0 Ashanti, Brong Ahafo 8.5 6 59.4 29.9 Northern, Upper 1.1 1 13.1 5 .1 (East, Upper West) Kenya Nairobi 18.8 15 93.2 31.5 Central, Eastern 9.0 7 92.4 15.5 Rift Valley 4.7 5 80.6 11.6 Coast 5.1 4 83.4 15.1 Western, Nyanza 2.6 3 88.1 16.7 Lesotho Lowlands 5.4 6 66.4 37.1 Other (Foothills, 2.1 3 54.5 28.7 Orange River Valley, Mountains)

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 185 Average Dummy Duration Vanable of Illiterate (1 = major Married in Religion (%) Education or Koranic urban, 0 = Polygamous (years) Only (%) not urban) Union (%) Muslim Christian Traditional 0.6 89 0 40.8 41 8 51 0.7 88 0 37.8 7 28 65 4.2 43 1 32.0 8 67 25 2.2 53 0 31.5 4 94 2 2.3 53 0 27.6 1 96 3 0.4 91 0 44.3 54 19 27 1.5 69 0 56.7 14 73 13 3.8 29 1 23.0 4 93 3 2.5 65 1 33.S 38 40 22 0.9 82 0 42.7 23 37 40 0.3 95 0 41.5 44 6 50 2.2 68 1 43.8 45 36 19 1.6 76 1 47.9 49 25 26 3.2 60 0 31.8 9 78 13 5.6 37 1 27.1 7 75 18 4.5 46 0 43.2 9 67 24 4.8 43 0 29.9 11 75 14 0.6 93 0 49.9 19 11 70 6.2 20 1 21.6 1 96 3 3.7 37 0 18.3 0 96 4 3.3 44 0 24.9 1 90 9 2.2 69 0 32.6 37 40 23 3.2 46 0 40.5 3 96 5.6 4 0 8.4 0 86 14 4.6 11 0 8.7 0 78 22

186 FACTORS AFFECTING CONTRACEPTIVE USE TABLE 6A-1 WFS Regional File Part A (continued) Married Married Use Contraceptive Who Know Who Want Modern Use- at Least 1 c 4 Method Effectiveness Modern Children Region (%) Index Method (%) (%) Nigeria Northeast 0.0 0 8.9 11.7 Northwest 0.3 0 7.3 7.7 Southeast 1.4 1.2 32.9 3.0 Southwest 1.5 1.5 34.6 7.9 Senegal Central 0.4 0 12.7 10.7 Fleuve, Oriental 0.0 0 16.1 11.7 Casamance 0.0 0 14.3 7.0 Dakar, Thies 2.3 2 37.9 14.3 Sudan (northern) Khartoum 17.6 13 82.7 41.1 North, east 2.9 3 49.8 28.2 Central 7.1 7 63.3 36.7 Kordofan, Darfur 1.5 2 26.5 20.9

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 187 Average Dummy Duration Variable of Illiterate (1 = major Married in Religion (%) Education or Koranic urban, 0 = Polygamous (years) Only (%) not urban) Union (%) Muslim Christian Traditional 0.6 86.8 0 42.3 69 23 8 0.3 94.5 0 49.3 93 1 6 3.8 52.0 0 39.3 0 84 16 4.0 52.5 1 35.5 36 57 7 0.4 93 0 50.0 98 1 1 0.5 91 0 45.0 98 1 1 0.6 89 0 54.4 87 13 0 2.0 71 1 45.1 93 7 0 2.7 57 1 9.8 96 4 0 1.0 79 0 12.6 98 1 1 0.9 79 0 13.3 98 1 1 0.6 89 0 27.2 98 0 2

188 _ 1 ~ ~ ._ i_ U: ~ _ ¢ ~ CC o ._ Ct C) C) ._ 3 C) C) C) Ct Pa o . U) Fl1 m ¢ Em c o C) U. c5 ~ ~ O 0 C,) . I . ~ O 3 ;_4 ~ ~ Q c' ~ ~ ~ in, - o so Ct - Cal o D ~ ¢ o ·- A: Ct .~ ~ 0£ Cal so Ct ._ ~ - m Ct + ~ O Vl o ._ oo cr Do 0 ~ ~ up ~ ~ ~ ~ ~0 CM cr ~ oo ~ ~ oo oo ~ oo oo ~ U~ 00 ~ U~ c~ oO ~[~ cr~ ~ ~ ~ oo ~ ~ r ~ c~ 0 u~ 0 oo ~ O ~ ~oo ~ ~ oo oo ~ ~ ~ 0\ e~ ~ ~ ~ ~ c~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~o ~ o ~ ~ ~ tn ~ ~ ~ 0 ~c~ ~ - ~- ~ ~ - ~ ~ ~4 - ~ ~ ~ ~ ~ ~ oo ~ ~ ~ ~ O ~ ~r. ~ ~ oo ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - ~ c~ ~o ~ ~cr ~ oo 0 u ~ ~ ~ oo O C~ c~ ~ ~oo ~ ~ v~ ~o ~ u~ ~ ~o ~ ~ ~ ~ ~ c~o - ~ ~ s - B o ~e _ ~ ~ B ·~: ~ ~ V ~ ~ ~ Z 3 ~ <0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Z ~V V

189 ~ Us ~ ~ to . . . . . oo cr. oo Do ~ o ~ Cal To o ~ ~ ~ ~ ~ ~ ~ CM Cal o ~ Caloo _ ~ ~ ~ Cal ~ ~ ~ ~or ~ ~ ~ . . . .. . . . So Go oo~ Cal Cal oo oo CM ~ oo ~ ~ ~ ~ _~ ~ of ~ ~ ~ Cal~ ~ en ~ ~ ~ ~ ~ ~_~ ~ _ ~ ~ _4 _(_. ~ ~ ~ ~ ~ ~ ~ ~ oo ~ a~ ~ ~ ~ ~ ~ c~ ~ ~ ~ ~ _ a, 0 ~ oo ~ O~ O ~ O~ ~ ~ ~0\ ~ oo C~ . . . . . . . . . . . . . . .. . . . O ~ ~ ~ ~ ~oo crx ~ ~ u~ ~ ~c~ :^ C) - Ct C Z U y ~ ~ O ~ o O ~ ~ Z Z 5 o ~ C ~ ~ ~ e ~ -o C ~·Z CQ ~ C~ Ct Ct ~: C~ o u) - ~ - ~ ~9 oo i ~ 3 ._ ._ ~o Ct C~ ~.. ;^ U O ~ CQ C) o~ 5: U,

190 TABLE 6A-2 DHS Regional File Part A FACTORS AFFECTING CONTRACEPTIVE USE Married Married Use Contraceptive Who Know Who Want Modern Use- at Least 1 < 4 Method Effectiveness Modern Children Region (%) Index Method (%) (%) Botswana Urban 40.8 40.1 99.7 54.1 Rural 27.5 27.2 91.9 34.1 Burundib Central plateau 0.8 4.3 67.8 31.9 Imbo 6.1 10.4 75.4 43.4 Lowlands 0.0 2.0 52.0 25.6 Mumirwa 2.3 5.4 52.7 33.7 Mugamba 0.9 4.4 63.5 36.0 Ghana Central, Western 4.1 7.0 79.0 32./ Greater Accra, 7.9 14.0 89.7 51.1 Eastern Volta 3.9 10.2 77.5 45.6 Ashanti, Brong Ahafo 6.0 8.8 79.5 44.9 Northern, Upper 0.7 6.7 40.4 6.7 (East, Upper West) Kenya Nairobi 27.9 30.4 94.8 78.8 Central, Eastern 24.5 33.1 94.1 68.2 Rift Valley 18.1 24.4 84.6 50.1 Coast 14.8 16.3 92.3 35.7 Western, Nyanza 10.1 12.0 92.1 49.3 Liberia Grand Gedeh 2.9 2.9 64.1 11.1 Montserrado 9.7 10.8 77.3 29.9 Since 3.9 4.1 87.2 8.8 Rest of country 4.4 4.6 64.4 20. Mali Bamako 6.1 11.4 74.3 33.E Kayes, Koulikoro 0.8 1.6 28.0 12.E Mopti, Gao, 0.9 1.4 15.0 7.S Tombouctou Sikasso, Segou 0.8 1.6 26.0 22.4 Nigeria Ondo State 3.8 5.1 50.0 10.2 Senegal Central 0.5 1.8 70.2 11.2 Northeast 0.6 1.1 38.9 11.1 (Fleuve, Oriental) South (Casamance) 2.4 2.6 53.5 13.0 West (Dakar, Thies ) 5.5 7.6 85.5 25.5

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 19 Average Dummy Duration Variable of Illiterate (1 = major Married in Religion (%) Education or Koranic urban, 0 = Polygamous (years) Only (%) not urban) Union (%) Muslim Christian Traditional 6.72 10.2 1 a 0.O 36.6 63.4 4.94 25.5 0 - a 0.O 29.2 70.8 0.91 62.4 0 8.2 b b b 4.12 46.5 1 24.5 b b b 0.52 73.1 0 20.7 b b b 0.69 70.1 0 12.1 b b b 1.01 65.3 0 6.8 b b b 4.26 58.3 0 27.0 7.3 82.9 9.7 6.25 39.4 1 27.0 8.1 80.5 11.5 4.80 50.6 0 43.8 4.4 59.4 36.2 5.47 46.2 0 30.2 8.4 75.6 16.0 1.27 87.6 0 48.3 28.3 18.7 52.9 7.57 9.3 1 15.5 6.6 88.4 4.9 5.64 21.3 0 14.5 0.3 98.5 1.2 4.60 34.0 0 19.8 0.7 91.4 8.0 3.87 41.5 0 34.1 34.7 46.0 19.3 4.97 31.8 0 33.2 1.1 97.7 1.3 1.49 73.8 0 55.5 6.4 66.5 27.0 4.47 47.5 1 25.8 16.6 61.5 22.0 2.04 68.2 0 35.6 1.2 82.9 15.9 1.88 70.7 0 40.7 14.8 48.7 36.6 2.76 58.2 1 32.7 95.8 4.2 0.0 0.76 88.7 0 50.4 92.0 2.8 5.2 0.56 89.7 0 42.4 94.3 3.9 1.9 0.61 89.6 0 45.4 91.7 0.9 7.5 5.17 40.5 0 46.1 13.4 84.7 1.9 0.54 92.5 0 49.1 98.2 1.8 0.0 0.83 88.3 0 47.1 98.6 1.4 0.0 1.22 82.7 0 51.7 91.1 7.9 1.0 2.79 65.2 1 41.1 93.3 6.6 0.1

Togo 92 FACTORS AFFECTING CONTRACEPTIVE USE TABLE 6A-2 DHS Regional File Part A (continued) Married Married Use Contraceptive Who Know Who Want Modern Use- at Least 1 ~ 4 Method Effectiveness Modern Children Region (%) Index Method (%) (%) Sudan (northern) Khartoum 15.8 18.8 96.3 41.4 North, east 4.4 5.7 74.2 20.8 Central 4.1 5.5 80.2 19.8 Kordofan, Darfur 0.8 1.3 45.3 10.8 Central 1.9 4.1 73.6 25.7 Coastal (including 4.6 11.5 88.8 57.2 Lome) Kara 3.3 9.6 74.4 38.7 Plateau 2.4 8.6 86.8 48.4 Savanna 0.3 0.7 62.1 13.3 Uganda West Nile 0.0 0.5 17.8 18.0 East 2.0 2.8 84.8 24.2 Central 2.4 3.7 78.7 22.4 West 3.4 5.3 61.0 27.1 Southwest 5.9 2.5 83.3 17.4 Kampala 17.9 21.4 96.3 46.0 Zimbabwe Bulawayo 41.2 41.5 99.5 66.3 Harare/Chitungwiza 48.0 48.6 99.0 57.6 Manicaland 25.6 28.7 97.7 29.0 Mashonaland Central 40.1 43.3 95.4 34.6 Mashonaland East . 43.1 44.6 98.2 33.7 (except Harare/Chitungwiza) Mashonaland West 43.2 45.0 98.8 38.1 Masvingo 35.3 41.7 96.8 29.8 Matabeleland North 18.0 23.1 96.9 35.2 (except Bulawayo) Matabeleland South 21.2 24.8 98.7 42.3 Midlands 35.2 39.8 97.2 41.5 aNot asked. bReligion not asked.

REGIONAL ANALYSIS OF CONTRACEPTIVE USE 193 Average Dummy Duration Variable of Illiterate (1 = major Marned in Religion (%) Education or Koranic urban, 0 = Polygamous (years) Only (%) not urban ) Union (%) Muslim Christian Traditional 5.18 33.4 1 13.5 94.5 5.5 0.0 2.36 61.3 0 14.0 97.5 2.5 0.0 2.49 62.7 0 15.1 99.2 0.7 0.1 1.31 78.9 0 32.1 99.7 0.3 0.0 1.78 77.8 0 60.9 46.4 20.5 33.0 2.71 62.3 1 52.5 5.7 40.1 54.1 2.41 66.0 0 52.8 12.0 32.2 55.7 2.69 58.0 0 46.8 6.6 59.8 33.5 0.42 94.9 0 53.9 14.9 4.8 80.4 1.50 77.6 0 33.1 24.2 75.8 0.0 2.99 58.0 0 42.0 17.2 81.8 1.0 3.95 39.0 0 31.3 13.6 86.1 0.3 3.13 45.5 0 39.0 1.2 98.1 0.7 2.62 44.4 0 27.4 3.8 95.3 0.9 7.00 13.1 1 34.3 14.2 85.4 0.4 8.05 6.6 1 7.5 0.0 77.6 22.4 8.00 4.9 1 12.7 0.0 84.6 15.4 5.48 19.0 0 20.2 0.0 66.0 34.0 4.55 32.6 0 22.1 0.0 67.0 33.0 5.81 19.5 0 14.1 0.0 64.1 35.9 5.13 25.9 0 12.2 0.0 64.3 35.7 5.81 15.3 0 18.3 0.0 56.9 43.1 4.76 30.7 0 28.1 0.0 65.1 34.9 6.16 11.0 0 11.5 0.0 56.9 43.1 6.37 14.9 0 20.4 0.0 70.4 29.6

194 m Ct 1 ._ to .= U: ¢ m ¢ _ ~ .= 1_ ~ - o a: . - - 00 ~ ~ ~ ~ ~ ON 00 of ~ox oo x as ~ oo C) ca _ = At, O i_ Cal ct 3 ~ :- ~ O ~ ~ · _ O Z ~ t_ ~ ~4 ~ ~ ~ O . . .. . . . . oo O~ ~ O O oo oO =' ON00 00 C', - 00~ 00 0 ~ ON~ ~ _ ~ ~ . .. . . . .. . . . .. . . . . ~ ~O ~ 0~ crx cr~ ~ ~ ~_ ~ ~ =\ ~ ~ _ ~ ~ _ _ ~ ~_ ~ ~ ~ _ _ ~ ._ C~ C) O ~ ~O ~ ~ O U ~ 10 oo ~ ~ oo _ t_ s~ ~ ~ O ~ ~ ~ ~ ~ ~O O ~ ~ _~ ~ _~ ~ C~ ~_d _ ~C~ ~ ~_ _ _ _ =,:' C O ~ ~ _` U: 5: o Ct ~ =' ·- U: O o Ct . _ ~4 Ct .t Ct e~ ~C C~ ,' - ~ 50 >,_, ·- + ~ Vl o ._ ~ ~O ~ ~ O ~c~ ~cr 0 . .. . . . .. .. . oo ~o ~ ~ ~ oo~ ~ ~ ~ _4 ~ _ _._ ~ _4 ~ oooo - oo ~ O . .. . . . . ~ ~ ~ ~ C~ ~ _ ~ ~ ~ ~ . . . . . cr ~ ~ ~ '_ _ ~4 CM ~ O ~ ~o _U) o ~ ~ oo - ~ =, _ so ~ ~ ~ ~ ooo ~ ~ ~ ~ . . . . .. .. . .. . . . . C~ ~ ~ ~ ~ ~ C~ Ct C: c 3 y 0 D > 3 D ~D 3 _ e D o ' o C o ~E 3 ~ z ~ ~ ~, ~ z ~ z ~ E ~ ~V ~

195 - o. ~ ~ o~ ~_ ~ oo ~) o ~ ~ ~ o_oo~ oo rid ~ ~ ~ ~Do~ ~ ~ oo C-} <~ 4~N C~l~ oc) CM ~__ ~ oo . . . .. . .... .. .. . . .. .. . . ~- C-] at too ~ ~- oo ~To oo -~ ~- ~ ~ ~ ~ o ~ ~ _ _4 ~ _ _ ~CM ~ ~ ~ ~ ~ ~CM ~ ~ ~ ~ ~ ~ ~ ~ ~C ~ ~ ~ . . . . . . . . . . . . . _ ~ ~ oo ~ _ ~_ oo ~ oo o ~ ~ '_ ~ ~ ~_ ~ _ ~ oo ~ ~c~ O cr ~ ~ . . . .. . . . . ~ c~ 0 0~ ~ r~ _ _ ~ CMCM ~ ~ C~ ~ ~ oo ~oo ~ ~cr0o~ ~ ~0 c~oo ~ ~ ~ . . . .. . .... .. .. . . .. .. . . ~ oo ~ ~o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ _~ ~_ ~ ~_ _ . ~_ _ ~_ ~_ _. ~_ o ~ ~ ~ ~ ~ . . . . . . . ~ ~ ~ ~V) ~ ~ ~ ~ ~ _ _ oo _ ~ . . ~ _~ oo ~ ~ ~oo ~ ~ . .. . . .. .. . . ko~ ~ ~ v) M 0 - 0 ~ ~ ~ ~ ~ u~ cr ~ o~ ~ 0 0 oo ~ oo ~ oo ~ v~ ~ O ~ ON ~ 0N ~ ~ ~ 00 a~ ~ c~ ~ ~ ~ ~ c~ ~ c~ ~ O ~ - ~ 5u 0 C ~ G' o ~ ~ ~= ~ U a ~ ~ ~ ~ - ~ ~ ~ ~ ~ ~ ~ _ ~ ~ ~ ~ ~ o ~ ~ z ~ 3 ~ ~ z v ~ ° ~ v ~ ~ ~ ~ ~z ~

196 s o _' m S~ 1 - ._ e~ o U. 1 ¢ m E~ C`: o . Ct . - C) CD C~ S~ _ .= _ U) ~ - U) _ ..= D ~ - s: ~ O C) _ C~0 Ct ~ ~ O . _ O Z ~ 04 .= C) .= ~ O s~ ~ ~ `_ `. Ct ' t ~o. o ~ o _' ~ U: ~S o C5 =. - o "D - o ~ 3 ~ C~ oc C~ ~ Ct i~ . · _ 't ~ ~ Cd + ~ O Vl o ._ o~ ~ 1 ~ ~ Oo ~ ~ ~oo ~ ~ O ~ ~ r~- ', . . . . . .. . . . .. . .. . kD ~ O ~ O ~oo ~ ~ O ~ ~ OM0 ~ V~ ~ ~ ~oo oo ~ ~ oo~ oo ~oo oo O ~ ~ ~ ~ ~ ~ ~ ~ oo~ ~ ~oo . . . . . .. . . . .. . .. . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~cM ~ ~ ~ ~ ~ ~4 ~ - ~4 - -~ ~ ~ ~ ~ cr, oo cr,~ ~ ~ ~ 0~ ~ ~ ~ ~ . . ~ .. . ~ ~ ~. . . ~ oo ~ r~ ~ ~ ~ ~ ~ ~ ~ c~ ~ ~ ~ ~ ~- ~ -~ ~ ~ ~ - t_ ~ ~ ood C~ ~ ~ ~ ~ ~ . . . .. . . . .. . . ~ ~ ~ oooN a~ c~ ~ 00 ~ ~ ~- - ~4 -4 ~- ~ ~ ~ ~o ~ ~ oo ~ ~ ~ ~ c ~'} - . . . . . . ~o In ~ <' d- ') ~t2 ~ ~· In oo ~O ~ O ~ ~ oo ~ .. . . . . . . . N ~ ~ ~- ~- =. ~ ~- t c 3 ~ ~ 3 v, ~ E ·N Cd ._ 3 3 Cd _ 3 ~ Ct Ct _ ~ s~ V ~ Ct ~ _ _ Ct ~ s: o o ~: U, ~ ~s ._ .O C) Ct Ct Ct . ~ tC 3 ~4 c' 3 ~ . ~ C) ~ · > Ce O O .= ;> Ct Ct ~S O ~ O ~ O ~ 3 u: _ ~ C) X ~ Ct - C) C~ C~ Ct ._ C~ C) s:~ Ct C5 ;` :> U, oo ~ 3 ._ ._ k ~. . ~ V O ~ s: - ~ c5 ._ ~s 5- oo

Next: 7 Contribution of Modern Contraceptive Use Relative to Postpartum Practices to Fertility Decline »
Factors Affecting Contraceptive Use in Sub-Saharan Africa Get This Book
×
Buy Paperback | $50.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This book discusses current trends in contraceptive use, socioeconomic and program variables that affect the demand for and supply of children, and the relationship of increased contraceptive use to recent fertility declines.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!