National Academies Press: OpenBook

Demographic Change in Sub-Saharan Africa (1993)

Chapter: 3 The Approximate Determinants of Fertility

« Previous: 2 Fertility Levels, Differentials, and Trends
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

3
The Proximate Determinants of Fertility

Carole L.Jolly and James N.Gribble

INTRODUCTION

Fertility levels in sub-Saharan Africa are among the highest in the world. As a result, recent fertility declines in a few countries have gained the attention of researchers and policy makers, and have renewed interest in the factors affecting fertility. As first outlined by Davis and Blake (1956), the factors affecting fertility can be classified into two groups: background variables and intermediate or proximate variables. The former includes cultural, psychological, economic, social, health, and environmental factors. The proximate determinants are those factors that have a direct effect on fertility. The background factors operate through the proximate determinants to influence fertility; they do not influence fertility directly.

Drawing on data from the Demographic and Health Surveys (DHS) and World Fertility Surveys (WFS), this chapter examines the relative effects of four proximate determinants on fertility: marriage patterns, contraceptive use, postpartum infecundability, and primary sterility. Using the Bongaarts model of proximate determinants of fertility, we examine how these four factors influence the levels of fertility and illustrate different effects of each

Carole L.Jolly and James N.Gribble are program officers for the Committee on Population, National Research Council. They thank Kenneth Hill for his assistance in estimating the measure of the degree of childbearing outside marriage and are also grateful for his help in the computations.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

factor with country examples. We also examine differentials across countries for each of the determinants and compare the changes over time by comparing results from the WFS and the DHS for those countries that conducted surveys under the auspices of both programs.

DATA

The data sources for this analysis are the 12 DHS of women conducted in sub-Saharan Africa during the 1980s and the WFS conducted during the 1970s in Ghana, Kenya, Senegal, and northern Sudan. Only four WFS countries are examined—those for which there was a subsequent DHS. The DHS core instrument gathered data on the socioeconomic status and reproductive history of women and the health of their children, as well as their experiences in using health services. The WFS also systematically gathered comparable data on fertility and mortality in nine sub-Saharan African countries, including northern Sudan.1

Although Sudan is included in our analysis, it is important to note that the survey was conducted in northern Sudan, a region that is primarily Arab/Muslim and quite distinct from the black African/Christian or animist south, which is more similar to the rest of sub-Saharan Africa. Table 3–1 provides information on the sample sizes, criteria for being included in the sample, and dates of fieldwork for the surveys.

FRAMEWORK

Bongaarts et al. (1984) enumerate nine major proximate determinants of fertility at the societal level:

  1. marriage or union patterns,

  2. contraception,

  3. lactational amenorrhea,

  4. postpartum abstinence,

  5. pathological sterility,

  6. induced abortion,

  7. frequency of sexual intercourse,

  8. spontaneous intrauterine mortality, and

  9. natural sterility.

1  

Although this analysis generally used data from the core questionnaires of the DHS and the WFS, in some cases a variation of a core question was asked or a core question was eliminated. In these cases, it was necessary to obtain the information from another question or to use an imputation procedure. See Technical Notes at the end of this chapter for details.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–1 Data Sets Used in the Analysis

Country

Abbreviation

Time of Fieldwork

Respondents

Sample Size

Demographic and Health Surveys

 

Botswana

BWA

August-December 1988

All women 15–49

4,368

Burundi

BDI

April-July 1987

All women 15–49

3,970

Ghana

GHA

February-May 1988

All women 15–49

4,488

Kenya

KEN

December-May 1988–1989

All women 15–49

7,150

Liberia

LBR

February-July 1986

All women 15–49

5,239

Mali

MLI

March-August 1987

All women 15–49

3,200

Ondo State, Nigeria

 

September-January 1986–1987

All women 15–49

4,213

Senegal

SEN

April-July 1986

All women 15–49

4,415

Sudana

SDN

November-May 1989–1990

Ever-married women 15–49

5,860

Togo

TGO

June-November 1988

All women 15–49

3,360

Uganda

UGA

September-February 1988–1989

All women 15–49

4,730

Zimbabwe

ZWE

September-January 1988–1989

All women 15–49

4,201

World Fertility Surveys

 

Ghana

GHA

February-March 1979–1980

All women 15–49

6,125

Kenya

KEN

August-May 1987–1988

All women 15–50

8,100

Senegal

SEN

May-October 1978

All women 15–49

3,985

Sudana

SDN

December-April 1978–1979

Ever-married women age 50 or under

3,115

aWFS and DHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Bongaarts and Potter (1983) developed a model to quantify the effects of the six proximate determinants that in their analysis had the most important influences on fertility levels: union patterns, contraception, lactational amenorrhea and postpartum abstinence, pathological sterility, and abortion (Bongaarts, 1982; Bongaarts et al., 1984). The analysis in this chapter is based on the Bongaarts and Potter model, but does not include abortion because reliable and comparable estimates are not available for sub-Saharan Africa.

The model relates total fertility to total potential fertility reduced by a series of indices, each of which reflects the fertility-reducing effect of a proximate determinant. An index, which has a range between 0 and 1 for most of the proximate determinants, is estimated (lactational amenorrhea and postpartum abstinence are joined into one index). An index value of 0 has the strongest effect of reducing fertility (fertility equals zero); a value of 1 has the weakest effect on fertility (the proximate determinant has no fertility-limiting effect). The lower the index, the more influential the proximate determinant is in reducing the total fecundity rate (TF), the level of fertility that would occur in the absence of all of the proximate determinants. Thus, the proximate determinants can be thought of as inhibitors of fertility. For example, delayed entry into marriage, use of family planning methods, and prolonged breastfeeding or postpartum abstinence are factors that reduce fertility to levels lower than those that would occur in the absence of these proximate determinants. Below is a description of the proximate determinants used in this analysis, the way these factors influence fertility through inhibiting TF, and the computational procedure used to estimate the indices. (Equations for deriving these indices are given in the appendix to this chapter.)

Marriage or Union Patterns2

Because entry into marriage is a process and not a single event in many parts of sub-Saharan Africa (see Chapter 4), this analysis looks at the effect of the proportions of women in sexual union, rather than marriage per se, on fertility. The proportion of women in a sexual union in a society indicates the degree to which women of reproductive age are exposed to the risk of becoming pregnant (if one assumes that all sexual intercourse occurs within union). In populations where women marry early and there is little divorce or separation, exposure to pregnancy is very high. In many parts of sub-

2  

In the context of this analysis, the term “marriage” refers to being married or living in a fairly stable union.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Saharan Africa, women marry young; median age at first union among women ages 25–49 at the time of the DHS surveys ranges from 15.7 in Mali to 19.7 in Ondo State, Nigeria. Although union dissolution is relatively common, most women remarry quickly, which results in a large proportion of ever-married women who are actually in a union (Mhloyi, 1988). Such behavior can be expected to result in high levels of fertility.

In the Bongaarts model, the index of the proportion in marriage or union, Cm, is intended to measure the effect on fertility of the proportion of women in a sexual union. The effect of marriage or union patterns on fertility is captured as the ratio of the average number of children a woman bears throughout her life (total fertility rate or TFR) to the number she would bear if she first entered a union at age 15 and stayed in that union until age 50 (total marital fertility rate or TMFR). Cm has the value of 1 when all women of reproductive age are in union and is equal to 0 when none are in union.

This formulation assumes that all fertility occurs within marriage or union. This assumption does not hold in many parts of sub-Saharan Africa, where substantial proportions of births are reported by women who describe themselves as single or never married, which may result in the calculated Cm being greater than 1. Anthropological studies indicate that the Western concept of marriage is not necessarily the appropriate paradigm to be applied to all of sub-Saharan Africa. Union formation may be an extended process, and births do occur outside of union (see Chapter 4; and Working Group on the Social Dynamics of Adolescent Fertility, 1993).

The fact that nonmarital births occur raises a problem for the Bongaarts model (which Bongaarts recognized). If births to unmarried women are excluded from the analysis, the TFR is underestimated, but the TMFR is estimated accurately. If, on the other hand, these births are included in both, the TFR is calculated accurately, but the estimated TMFR is inflated, giving the impression that marriage patterns reduce fertility by a much greater fraction than is actually the case.

To circumvent this problem and to maintain a consistent definition for other variables in the Bongaarts’ model using women currently in union only, we have added a variable to the model. This variable, Mo, captures the effect on total fertility of births outside union. Mo relates total fertility calculated by using all births to total fertility from using births only to women in union. Cm, a modified version of Cm, captures the effect on total fertility of the specific observed union pattern, under the assumption that no births occur outside unions. The product of Mo and Cm is Cm, the usual definition of the effects of marriage patterns on fertility used in the Bongaarts model.

To summarize, in our model, Mo can be thought of as the effect of births outside union on total fertility (thus a value of Mo of 1.43 indicates

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

that the TFR is approximately 43 percent higher than it would have been if all fertility occurred in unions). Cm can be thought of as the effect of reported union patterns on fertility if births occur only in unions, and Cm is the combined result of the fertility-inhibiting effect of union pattern and the fertility-promoting effect of sexual relations outside union. The two new indices are related because, if all women were in unions from age 15 to age 50, there would be no births to women not married, and Mo would be equal to 1 (i.e., there would be no effect on fertility) and Cm would equal Cm. It is important to note that Mo is not a fertility-reducing parameter of the model, but rather a device to maintain comparability across cultures in the interpretation of other parameters of the model. Cm, Cm, and Mo are reported in the tables; only Cm is shown in the figures.

Contraception

The proportion of women using contraception to space or limit births and the effectiveness of the contraception they use directly affect a society’s fertility level. In sub-Saharan Africa, contraceptive prevalence rates are generally low in comparison with other regions of the world (Rutenberg et al., 1991). There is also substantial use of traditional methods, which are not as effective in preventing pregnancy as modern methods.

The index of contraception, Cc, measures the effect of actions intentionally taken to reduce the risk of conception. Cc equals 1 if no form of contraception is used and 0 if all fecund exposed women use modern methods that are 100 percent effective.

Postpartum Infecundability

There are several practices women can follow after the birth of a child that delay a subsequent pregnancy. A woman is unable to conceive after a pregnancy until her normal pattern of ovulation returns. When she is breastfeeding, the length of lactational amenorrhea is determined primarily by the duration, intensity, and pattern of breastfeeding. Moreover, in a number of societies, sexual relations are not permitted while women breastfeed their newborn children, which further reduces the chances of conception.

In much of sub-Saharan Africa, women breastfeed for long periods and refrain from sexual relations after the birth of a child. Both of these practices are seen as necessary to preserve the health of the child and mother (van de Walle and van de Walle, 1988). In most of the sub-Saharan African countries analyzed here, the duration of breastfeeding was much longer than postpartum abstinence (see Table 3–2). However, substantial variation in both practices exists within the region.

The index of postpartum infecundability, Ci, estimates the effect of

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–2 Mean Duration (months) of Postpartum Variables for Women Currently Married

Country

Breastfeeding

Amenorrheic

Abstaining

Nonsusceptiblea

Weighted No. of Births

Botswana

19.2

11.7

8.9

13.3

932

Burundi

23.9

19.4

2.4

19.6

2,306

Ghana

20.9

14.6

12.9

17.7

2,314

Kenya

20.1

11.2

3.9

11.7

3,667

Liberia

17.5

11.7

13.1

15.5

2,554

Mali

21.5

15.7

7.0

17.0

2,101

Ondo State

18.8

14.2

22.7

23.9

1,847

Senegal

19.2

15.8

6.8

17.6

2,433

Sudanb

19.7

14.1

4.6

14.9

3,885

Togo

23.0

14.6

17.2

20.1

1,804

Uganda

19.1

13.1

3.0

13.4

2,654

Zimbabwe

18.2

11.4

4.1

11.9

1,760

NOTE: Data are national-level DHS.

aSee Technical Notes (at end of this chapter) on derivation of indices for a discusssion of nonsusceptible period.

bDHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

postpartum amenorrhea and abstinence on fertility. When there is no lactation or postpartum abstinence, Ci equals 1; when infecundability is permanent, Ci equals 0.

Pathological or Primary Sterility

Several studies of infecundity in sub-Saharan Africa have indicated relatively high levels, particularly in Central Africa (Frank, 1983a; Bongaarts et al., 1984; Farley and Besley, 1988). Bongaarts et al. (1984) found that at least 20 percent of women in much of Central Africa are childless at the end of their reproductive years. In parts of Central and East Africa, between 12 and 20 percent of women ages 45 to 49 are childless. Lower levels generally exist in West Africa. Clearly, such high levels of infecundity inhibit the level of fertility achieved in many African societies.

Although infertility increases naturally as a woman ages (natural sterility), much of the primary sterility (inability to have any children at all) in sub-Saharan Africa is caused by sexually transmitted diseases (STDs) (Caldwell and Caldwell, 1983; Frank, 1983a). It is generally thought that gonorrhea is the most prevalent STD affecting African populations.

Ip, the index of sterility, takes into account only primary sterility and not secondary sterility, which is the inability to bear a second or subsequent child. Calculation of Ip is based on a 3 percent standard rate of childlessness in developing countries (Frank, 1983a; Bongaarts et al., 1984). If the rate of childlessness exceeds 3 percent, Ip will have a value less than 1, indicating that it reduces fertility. However, if less than 3 percent of women aged 40 to 49 are childless, then Ip has a value greater than 1, which indicates that levels of primary sterility are lower than would be expected in a developing country. It is difficult to interpret such a result in the context of a proximate determinants analysis because it suggests that low levels of primary infecundity increase fertility. When calculating Ip with the data used here, most of the indices were greater than 1. As a result, the index was omitted from many of the figures (see further discussion below).

Summary of Model

Each index outlined above (except Mo) acts as an inhibitor to fertility. The observed fertility rate (TFR) is equal to total fecundity rate (TF) multiplied (generally reduced) by each index:

TFR=TF•CmCcCiIp.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

FIGURE 3–1 Relationship between the fertility-inhibiting effects of the proximate determinants and various measures of fertility.

 

Proximate Determinant Indices

Cm:

index of marriage

Mo:

effect of births outside unions on total fertility

Cm:

adjusted index of marriage

Cc:

index of contraception

Ci:

index of postpartum infecundability

Ip:

index of sterility

The model can also be shown graphically, as in Figure 3–1. The column, which represents TF, is divided into five segments. The solid base at the bottom indicates the observed total fertility rate based on the reported number of births occurring in the four years prior to the survey. Moving upward, the height of the next segment indicates the level fertility would be if all women were in a union during the whole of their reproductive years (the TMFR). If no women in union practiced contraception, observed fertility would rise to the top of the next segment. This height represents the total natural marital fertility rate (TNMF). The top two segments of the

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

column indicate the effects of postpartum infecundability and sterility. The height of the column indicates the fertility level one would observe if none of the proximate determinants was exerting a fertility-reducing effect (i.e., if all the indices were equal to 1).

EMPIRICAL RESULTS

National-Level Results for DHS Countries

The national-level results of the proximate determinants analysis for the DHS countries are illustrated in Figure 3–2. The index of primary sterility is not included in the graph because many of the values are greater than 1. (The actual numbers used in the figure are reported in Table 3–3.) The height of the columns estimates the total fecundity rate (TF) of the national population of each country. The columns vary in height, but fall within the range of 12.9 to 16.5, basically within the theoretical range of 13 to 17 suggested by Bongaarts and Potter (1983).

FIGURE 3–2 Relationship between the fertility-inhibiting effects of the proximate determinants and various measures of fertility, by country (for country abbreviations, see Table 3–1). NOTE: PPI: Postpartum infecundability; DHS data for Sudan refer to only northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–3 Proximate Determinants of Fertility

Country

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Botswana

0.87

0.46

1.89

0.70

0.63

1.00

13.0

5.0

Burundi

0.80

0.76

1.06

0.97

0.53

1.03

16.5

6.9

Ghana

0.85

0.77

1.11

0.93

0.55

1.02

14.3

6.4

Kenya

0.86

0.73

1.17

0.80

0.66

1.01

14.4

6.6

Liberia

0.93

0.75

1.24

0.94

0.59

1.00

12.9

6.7

Mali

0.98

0.95

1.02

0.98

0.56

0.99

13.1

7.0

Ondo State

0.83

0.80

1.04

0.96

0.47

1.03

15.8

6.1

Senegal

0.90

0.84

1.07

0.97

0.55

0.98

14.0

6.6

Sudana

0.68

0.66

1.03

0.94

0.60

0.99

12.9

4.9

Togo

0.87

0.82

1.06

0.94

0.52

1.02

15.3

6.6

Uganda

0.92

0.77

1.19

0.97

0.63

0.97

13.6

7.4

Zimbabwe

0.81

0.73

1.12

0.63

0.66

1.01

16.3

5.5

NOTE: Data are national-level DHS.

aDHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

The data from the DHS in the mid-1980s show that the TFRs range from 4.9 in northern Sudan to 7.4 in Uganda. Botswana also exhibits a relatively low level of fertility, 5.0 births. Mali and Burundi show relative high levels of fertility, 7.0 and 6.9, respectively (see Chapter 2 for further discussion of fertility in sub-Saharan Africa).

Marriage

The first proximate determinant estimated is the effect of union or marriage patterns on fertility (see Chapter 4 for a discussion of marriage data). Mo, which describes the effect on total fertility of births outside union, ranges in these populations from 1.02 in Mali to 1.89 in Botswana. This value indicates that in Mali, very little childbearing occurs outside of union (only 2 percent of the fertility in the TFR occurred outside of union). Botswana, on the other hand, shows a high level of fertility occurring to women not currently in union. If the TFR were estimated by using only births in union, it would be 89 percent lower than the level observed when all births are considered.

In populations where Mo is large, Cm, which captures the effect of union patterns on fertility under the assumption that no births occur outside of unions, varies substantially in comparison with Cm. In settings where much childbearing occurs outside of union, the inhibiting effect of union pattern alone on fertility is underestimated by Cm (i.e., Cm is higher than it should be). In Botswana, for example, Cm is 0.46, compared to a Cm value of 0.87. (As explained in the description of the framework, the lower the index, the greater inhibiting effect it has on fertility.) The value of Cm for Botswana suggests that union patterns (relatively late age at marriage and substantial proportions of women not in union) have a large effect in reducing fertility. In Mali, because most childbearing occurs in union, only a small difference is observed between Cm and Cm. Sudan also has a very low Mo value, and consequently, the values of Cm and Cm are very similar. Although part of the low value of the Cm index for Sudan can be explained by rising age at marriage and increasing proportions of women never married, part of the difference may be explained by the way in which the TFR was calculated (see note 2). Kenya, Liberia, and Uganda have substantially lower Cm index values, reflecting the effect of union patterns on fertility. These countries also have correspondingly high Mo values—greater than 1.15—indicating that a relatively large amount of the fertility contributing to the TFR is outside of union. By controlling for childbearing outside of marriage, the reductive effect of union patterns is much stronger than indicated solely by Cm (as evidenced by the Cm values being lower than the Cm values).

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Contraception

The second index Cc evaluates the influence of contraceptive use on fertility. This factor quantifies the difference between the TMFR and the TNMF.

In general, contraceptive use in sub-Saharan Africa is low, as reflected in Cc values close to 1. However, the index values for Botswana, Kenya, and Zimbabwe are lower, relative to other countries, indicating that contraception plays a more important role in limiting fertility. In Zimbabwe, for example, the Cc of 0.63 reflects a relatively high contraceptive prevalence rate of 43 percent and a fairly effective method mix, with 31.0 percent of women using pills, 1.1 percent using the IUD, 2.5 percent using sterilization, and 8.4 percent using other (mostly traditional) methods.

Burundi, Mali, Ondo State, Senegal, and Uganda have relatively high Cc values, all greater than 0.95, reflecting very low contraceptive prevalence rates and ineffective method mixes. For example, Cc in Mali is 0.98, based on a contraceptive prevalence of 3.3 percent, with most women using traditional methods. Because the index of contraception is so close to 1, the TMFR and the TNMF are almost identical, indicating that fertility in union is close to the level that would exist in the absence of contraception.

Postpartum Infecundability

Analysis of the mean durations of breastfeeding, amenorrhea, and abstinence among women in union indicates that these postpartum practices last substantially longer than in many parts of the world, although there is considerable regional variation (Mhloyi, 1988; van de Walle and Omideyi, 1988). Mean duration of breastfeeding ranges from 17.5 months in Liberia to 23.9 months in Burundi. Mean duration of abstinence varies from 2.4 months in Burundi to 22.7 months in Ondo State. These durations translate into relatively long periods of nonsusceptibility after childbirth, particularly in West African countries. (See Table 3–2 for national-level estimates.)

The third index Ci demonstrates the effect of such long postpartum nonsusceptible periods on fertility. In Ondo State, Ci had the largest fertility-reducing effect of all the proximate determinants, with a value of 0.47. The index primarily reflects a very long period of postpartum abstinence, 22.7 months (the longest of all the DHS populations considered here), which contributed greatly to an nonsusceptible period of 23.9 months. The effect of this index on fertility is to reduce the average number of births per woman by 8.2; that is, if postpartum breastfeeding and abstinence ceased, the average observed total fertility rate would increase, ceteris paribus, by 8.2 births.

Many countries also have low values of Ci (.56 and less): Burundi,

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Ghana, Mali, Senegal, and Togo. Four countries (Botswana, Kenya, Uganda, and Zimbabwe) have index values greater than 0.60, indicating a less significant influence of postpartum nonsusceptibility on reducing fertility. Three of these countries—Botswana, Kenya, and Zimbabwe—also exhibit relatively high levels of contraceptive use.

Pathological Sterility

The fourth index Ip is the index of involuntary infecundity or pathological sterility. The proportion of ever-married women ages 40 to 49 who are childless ranges from 0.8 percent in Ondo State to 5.2 percent in Uganda. Altogether, fewer than half of the countries exhibit proportions childless of greater than 3 percent, the average expected for a developing country. Given this result, it may be surmised that an average of 3 percent sterility is too high an estimate, contrary to the conclusions of much of the literature on pathological sterility in sub-Saharan Africa. However, Central Africa, where the highest levels of infecundity are reported to exist (Page and Coale, 1972; Frank 1983a,b), is underrepresented in the DHS thus far, and childlessness may be underreported by women who do not want to reveal that they have borne no children (Larsen, 1989).

The average of 3 percent is based on work by Frank (1983a), who uses data sources published in the 1960s and 1970s to determine the prevalence of childlessness among women ages 45–49. Farley and Besley (1988) report that a large part of the infertility in sub-Saharan Africa is the result of infections (sexually transmitted diseases) that can be treated with antibiotics. The availability of antibiotics for the treatment of other infectious diseases may have reduced the prevalence of STDs in the 1950s and 1960s, thus reducing levels of infertility among women ages 40–49 in the late 1980s.

Overall, the effect of the index is small, with the greatest effect observed in Uganda. There, the index was 0.97, indicating that primary sterility reduced fertility by an average of only 0.40 birth per woman.

Interpreting the Results

Using the proximate determinants and data from the WFS and DHS yields national-level TF estimates ranging from 12.1 to 16.5 (see Tables 3–4 and 3–7). Bongaarts and Potter (1983) indicate that estimates for most populations range from 13 to 17, with an average of 15.3. The TF values for the 16 populations examined in this chapter average about 14.1, more than a birth lower on average per woman. This unexplained difference plus some methodological biases inherent in the proximate determinants model point to the danger of taking these estimates too literally.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–4 Proximate Determinants of Fertility by Age of Women

Country and Age

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Botswana National

0.87

0.46

1.89

0.70

0.63

1.00

13.0

5.0

15–24

0.83

0.24

3.40

0.78

0.61

1.00

4.3

1.7

25–34

0.92

0.57

1.63

0.67

0.64

1:00

5.0

2.0

35–49

0.87

0.61

1.42

0.70

0.63

1.00

3.5

1.3

Burundi National

0.80

0.76

1.06

0.97

0.53

1.03

16.5

6.9

15–24

0.56

0.51

1.10

0.98

0.53

1.03

5.4

1.6

25–34

0.95

0.90

1.06

0.97

0.53

1.03

6.1

3.1

35–49

0.88

0.85

1.04

0.97

0.51

1.03

5.1

2.3

Ghana National

0.85

0.77

1.11

0.93

0.55

1.02

14.3

6.4

15–24

0.71

0.60

1.20

0.95

0.53

1.02

5.3

1.9

25–34

0.95

0.89

1.07

0.93

0.57

1.02

5.1

2.6

35–49

0.90

0.84

1.07

0.91

0.54

1.02

4.0

1.8

Kenya National

0.86

0.73

1.17

0.80

0.66

1.01

14.4

6.6

15–24

0.75

0.56

1.35

0.87

0.66

1.01

5.4

2.3

25–34

0.94

0.85

1.11

0.79

0.68

1.01

5.4

2.7

35–49

0.92

0.86

1.08

0.76

0.64

1.01

3.5

1.6

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Liberia National

0.93

0.75

1.24

0.94

0.59

1.00

12.9

6.7

15–24

0.91

0.61

1.48

0.96

0.58

1.00

4.7

2.4

25–34

0.95

0.83

1.14

0.93

0.60

1.00

4.7

2.5

35–49

0.95

0.82

1.15

0.94

0.59

1.00

3.5

1.8

Mali National

0.98

0.95

1.02

0.98

0.56

0.99

13.1

7.0

15–24

0.98

0.94

1.04

0.98

0.57

0.99

4.6

2.5

25–34

0.99

0.97

1.02

0.98

0.56

0.99

5.2

2.8

35–49

0.98

0.95

1.02

0.99

0.55

0.99

3.4

1.8

Ondo State Regional

0.83

0.80

1.04

0.96

0.47

1.03

15.8

6.1

15–24

0.56

0.52

1.09

0.98

0.48

1.03

5.6

1.5

25–34

0.99

0.98

1.02

0.95

0.51

1.03

6.0

3.0

35–49

0.96

0.93

1.03

0.96

0.42

1.03

4.0

1.6

Senegal National

0.90

0.84

1.07

0.97

0.55

0.98

14.0

6.6

15–24

0.81

0.71

1.14

0.99

0.54

0.98

5.2

2.2

25–34

0.95

0.91

1.04

0.96

0.57

0.98

3.4

2.7

35–49

0.95

0.92

1.03

0.97

0.55

0.98

5.4

1.7

Sudan Northern

0.68

0.66

1.03

0.94

0.60

0.99

12.9

4.9

15–24

0.44

0.43

1.03

0.96

0.61

0.99

4.8

1.2

25–34

0.81

0.79

1.03

0.93

0.60

0.99

5.3

2.4

35–49

0.87

0.85

1.02

0.93

0.59

0.99

2.7

1.3

Togo National

0.87

0.82

1.06

0.94

0.52

1.02

15.3

6.6

15–24

0.74

0.66

1.11

0.96

0.52

1.02

5.4

2.0

25–34

0.95

0.92

1.04

0.94

0.54

1.02

5.3

2.6

35–49

0.91

0.89

1.03

0.94

0.48

1.02

4.7

2.0

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Country and Age

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Uganda National

0.92

0.77

1.19

0.97

0.63

0.97

13.6

7.4

15–24

0.86

0.69

1.25

0.99

0.64

0.97

4.9

2.6

25–34

0.94

0.82

1.16

0.97

0.63

0.97

5.3

3.0

35–49

0.97

0.82

1.18

0.94

0.58

0.97

3.5

1.8

Zimbabwe National

0.81

0.73

1.12

0.63

0.66

1.01

16.3

5.5

15–24

0.69

0.54

1.27

0.62

0.60

1.01

6.8

1.8

25–34

0.90

0.85

1.07

0.56

0.68

1.01

6.9

2.3

35–49

0.86

0.83

1.05

0.71

0.71

1.01

3.2

1.4

NOTE: Data are national-level DHS.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

There are several factors at work that could be affecting the reliability of our estimates. One is errors in the WFS and DHS data. Although both of these survey programs were highly professional operations, the data sets for sub-Saharan Africa do contain reporting errors. Statements regarding age at marriage, duration of postpartum abstinence, use of contraception, and current age are often approximate. For example, in more than half of the DHS surveys for sub-Saharan Africa, less than 50 percent of the women interviewed were able to give their date of birth (Institute for Resource Development, 1990). In all the African surveys (WFS and DHS), there was considerable age heaping (Goldman et al., 1985; Institute for Resource Development, 1990).

The second factor to consider is biases within the proximate determinants model. Menken (1984) and Reinis (1992) both find that the model produces very good estimates under the assumption of random use of contraception. However, Reinis finds that with nonrandom use of contraception, which is more likely given that women tend to use contraception depending on their family-building plans, the estimates produced, except for Ci, are less accurate. In particular, the model performs poorly when women use contraception to stop rather than to space births, when there is delayed marriage, and when contraceptive use is most prevalent at the oldest ages (which happens when large families are sought).

In sub-Saharan Africa, one or more of these conditions exist for many countries. For example, parity progression ratios indicate a stopping pattern in Kenya (DHS; see Chapter 2); in some of the countries, particularly those with low contraceptive prevalence rates, contraceptive use is concentrated in the oldest age group; and in some countries, dates of first union are relatively late.

A third factor not accounted for in our application of the model is the incidence of induced abortion. Although reliable and comparable cross-national estimates do not exist for sub-Saharan Africa, there is limited evidence indicating that abortion is substantial in some regions. The failure to incorporate abortion in the model will affect our assessment of the relative importance of the fertility-inhibiting variables and the estimation of TF (which will be underestimated in areas of high abortion rates). Coeytaux (1988), in a review of the literature on abortion, notes that studies conducted in the late 1970s and early 1980s show rising hospital admissions for complications related to abortions. She notes that ethnographic data suggest that a greater number of abortions are performed than originally thought. Data on adolescent unmarried urban women indicate that abortion may be fairly common (Working Group on the Social Dynamics of Adolescent Fertility, 1993). However, some studies indicate that abortion is also practiced by older married rural women. The survey data that do exist probably

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

underestimate the prevalence of abortion because many women who have had an abortion deny it in interviews (Coeytaux, 1988).

Summary

In all of the DHS sub-Saharan African populations included in this analysis, except Zimbabwe, the proximate determinant having the greatest fertility-inhibiting effect is the postpartum nonsusceptible period. At a minimum, observed fertility would increase by 4.8 births in Botswana and, at a maximum, by 8.2 births in Ondo State in the absence of breastfeeding and postpartum abstinence. The practice of spacing children for the health of the child and mother still continues to exhibit a powerful fertility-reducing effect.

The second most powerful proximate determinant in inhibiting fertility is marriage patterns (Cm). For all populations except Botswana, Kenya, Mali, and Zimbabwe, union patterns are more important than contraceptive use in reducing fertility. However, when controlling for the effect of births among women currently not married, the effect of union patterns, Cm, is even greater except in Zimbabwe. In Mali, the reductive effects of union patterns and contraception are approximately equal and both are very weak. The use of contraception, in the three countries where it has a stronger effect on fertility than union patterns (based on Cm), reduces observed fertility by 2.5 births in Botswana, 1.9 births in Kenya, and 4.0 births in Zimbabwe.

Differentials for DHS Countries

Comparison of the proximate determinants across subpopulations helps illuminate background factors that underlie fertility differentials. For example, observed fertility is generally higher in rural areas than in urban areas, and the proximate determinants provide an understanding of what behavioral or biological factors are associated with this differential. In this section, we examine differentials of the proximate determinants of fertility by age, residence, and education using DHS data.

Age

Age is divided into three groups: 15–24 years (youngest), 25–34 years (middle), and 35–49 years (oldest). Table 3–4 provides estimates across age groups for all the countries studied. In most countries, marriage patterns are the primary factor that distinguishes the youngest group from the middle and oldest groups. Cm and Cm have the strongest fertility-reducing effect during the younger years due to entry into union. The later the age at

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

marriage and the larger the proportion of never married, the greater is the fertility-inhibiting effect of Cm or Cm. In the middle and oldest groups, marriage patterns do not inhibit fertility as much as in the youngest ages because most women are married by age 25, and marriage dissolution through widowhood or divorce has little effect.

Interestingly, in five of the populations, contraceptive patterns have their greatest reductive effect in the middle age group, the years when childbearing is at its peak. Contraception has the second greatest effect in the oldest age group. In the countries where use of contraception is highest—Botswana, Kenya, and Zimbabwe—patterns vary across age group. Botswana follows the general pattern just outlined. In Kenya, contraceptive patterns have the greatest fertility-inhibiting effect in the oldest age group, followed by the middle age group. In Zimbabwe, the strongest effect is in the middle age group, followed by the youngest age group.

In most of the countries studied, the index of postpartum infecundability (Ci) varies less across age groups than do the indices of marriage (Cm or Cm) or contraception (Cc). In seven of the populations, postpartum practices have their greatest effect in reducing fertility in the oldest age group, indicating that younger women are breastfeeding and abstaining for shorter periods than are older women. In five countries, however, postpartum infecundability has its greatest effect in the youngest group.

Two examples, shown in Figure 3–3, illustrate the differentials in fertility across age groups attributable to the proximate determinants.

Ondo State, Nigeria The TFRs by age group in Ondo State are 1.5 in the youngest group, 3.0 in the middle group, and 1.6 in the oldest group (for a total TFR of 6.1 across all age groups). Marriage is almost universal within the middle and oldest groups. Among the youngest group, marriage is far less common and occurs at a relatively late age (the median for those aged 15–24 is 19.7 years).

The index of contraception does little to reduce fertility in Ondo State because contraceptive prevalence in all three age groups is low, ranging from 4.1 percent among the youngest to 7.7 percent among the oldest, and method use-effectiveness is low. In all three age groups, the index of postpartum infecundability has the strongest effect of the indices on reducing TF. Thus, the variation in fertility across age groups in Ondo State is attributable principally to different marriage patterns.

Zimbabwe Fertility levels across age groups in Zimbabwe are similar to those observed in Ondo State, but contraception is the major inhibitor to fertility. Marriage patterns have a smaller fertility-reducing effect in the youngest age group in Zimbabwe than in Ondo State. Even so, marriage

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

FIGURE 3–3 Relationship between the fertility-inhibiting effects of the proximate determinants and various measures of fertility, by age, for Ondo State, Nigeria and Zimbabwe. NOTE: PPI: Postpartum infecundability.

patterns in Zimbabwe reduce fertility more in each of the age groups than they do in most of the other DHS populations.

Contraceptive use is relatively high in Zimbabwe, with prevalences of 41.7 percent in the youngest group, 50.5 percent in the middle group, and 36.1 percent in the oldest group. Consequently, Cc has the greatest effect on inhibiting fertility for each age group of all the proximate determinants.

Finally, Ci has its largest inhibitive effect among the youngest group, corresponding to a nonsusceptible period of 14.7 months. This same pattern was observed in four of our other study populations.

Residence

Fertility tends to be higher in rural areas than in urban areas. Residence may have a strong effect on fertility by influencing a woman’s values, how she spends her time, and her view of the world (Zeidenstein, 1979). Women in rural areas may want larger families to ensure that someone will help with domestic and agricultural activities and provide financial security in old age. In urban areas, women may begin to limit their fertility because of the costs associated with childbearing. Living in an urban area

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

may change women’s values as they are exposed to the modern health sector, family planning, and more Western attitudes (Acsadi and Johnson-Acsadi, 1990).

Table 3–5 presents results of the analysis of proximate determinants by residence, which are generally consistent with this pattern of differential fertility by residence. In all the populations except Ondo State and Botswana, Cm and Cm are lower in urban areas than in rural areas, indicating that a larger proportion of rural women are married. Consistent with this pattern is the tendency of Mo to be higher in urban areas than in rural areas, reflecting greater proportions of nonunion births in urban areas.

Cc has a greater effect in inhibiting fertility among urban women due to higher contraceptive prevalence and use of more effective methods. The index of postpartum infecundability Ci is lower among women in rural areas, indicating that either breastfeeding or postpartum abstinence has a stronger fertility-inhibiting effect there. In all the populations except Botswana and Burundi, Ip is lower in urban areas, indicating higher levels of primary sterility. Larsen (1989) has suggested that despite generally better health care in urban areas, the incidence of STDs is likely to be higher there because prostitution is more common.

Education

Education plays an important role in inhibiting fertility, although substantial variation is observed in how fertility is associated with different levels of education, as illustrated in Table 3–6. Two basic patterns describe the association. In the first pattern, which is common particularly in Latin America and observed in seven of the sub-Saharan African populations analyzed here (Botswana, Ghana, Senegal, Sudan, Togo, Uganda, and Zimbabwe), as women’s education increases, fertility declines monotonically. In the second pattern, which is observed in five of our populations (Burundi, Kenya, Liberia, Mali, and Ondo State), the relationship follows an inverted U-shape pattern, that is, in comparison to women with no education, women with some education have higher fertility, but those with even greater amounts of education have the lowest fertility of all.

These two patterns have been previously documented by Cochrane (1979, 1983) among others. A common explanation of the finding that women with some education have higher fertility is that increased female education and urbanization are generally correlated with decreases in durations of breastfeeding and postpartum abstinence, which lead to shorter intervals between births and thus higher fertility (Adegbola et al., 1977). Women with some education are also likely to seek modern health care for medical problems, such as STDs, leading to reduced subfecundity (Cochrane, 1979; Romaniuk, 1980). Overall, Cc and Cm show the greatest variation by educa-

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–5 Proximate Determinants of Fertility by Urban or Rural Residence

Country and Residence

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Botswana

 

Urban

0.86

0.48

1.80

0.61

0.68

1.02

11.0

4.0

Rural

0.86

0.45

1.91

0.74

0.61

0.99

13.7

5.3

Burundi

 

Urban

0.75

0.60

1.25

0.83

0.70

1.03

11.1

5.0

Rural

0.81

0.76

1.05

0.98

0.52

1.03

16.5

7.0

Ghana

 

Urban

0.78

0.70

1.11

0.89

0.60

1.01

12.6

5.3

Rural

0.89

0.80

1.11

0.95

0.54

1.03

15.0

6.9

Kenya

 

Urban

0.82

0.66

1.24

0.74

0.70

0.96

11.7

4.7

Rural

0.87

0.75

1.17

0.81

0.66

1.01

14.8

7.0

Liberia

 

Urban

0.91

0.68

1.33

0.90

0.62

0.99

12.3

6.1

Rural

0.96

0.80

1.19

0.97

0.58

1.01

13.3

7.1

Mali

 

Urban

0.96

0.93

1.03

0.94

0.62

0.96

11.6

6.2

Rural

0.98

0.97

1.02

1.00

0.55

1.00

13.6

7.3

Ondo State

 

Urban

0.83

0.80

1.04

0.94

0.50

1.01

15.1

6.0

Rural

0.83

0.80

1.04

0.97

0.45

1.04

16.3

6.1

Senegal

 

Urban

0.79

0.71

1.12

0.92

0.63

0.96

12.7

5.6

Rural

0.98

0.93

1.04

0.99

0.53

0.99

14.5

7.3

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Sudana

 

Urban

0.55

0.53

1.03

0.87

0.65

0.98

12.2

3.7

Rural

0.71

0.70

1.01

0.58

0.98

1.00

13.3

5.2

Togo

 

Urban

0.79

0.73

1.08

0.90

0.57

0.99

12.2

4.9

Rural

0.91

0.87

1.05

0.96

0.50

1.03

16.3

7.4

Uganda

 

Urban

0.84

0.62

1.36

0.86

0.70

0.93

11.8

5.6

Rural

0.91

0.77

1.18

0.98

0.62

0.97

14.1

7.5

Zimbabwe

 

Urban

0.70

0.52

1.36

0.52

0.68

1.00

15.6

3.9

Rural

0.87

0.78

1.11

0.67

0.65

1.01

16.5

6.3

NOTE: Data are national-level DHS.

aDHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–6 Proximate Determinants of Fertility by Level of Education

Country and Education Level

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Botswana

 

None

0.85

0.50

1.70

0.83

0.61

0.98

13.8

5.8

1–4 years

0.98

0.53

1.87

0.72

0.59

0.99

13.0

5.5

5–7 years

0.84

0.44

1.93

0.67

0.65

1.02

12.5

4.7

8+years

1.02

0.42

2.44

0.50

0.69

1.05

9.1

3.4

Burundi

 

None

0.81

0.76

1.06

0.98

0.52

1.03

16.4

6.9

1–4 years

0.77

0.74

1.05

0.96

0.54

1.05

16.9

7.1

5–7 years

0.79

0.72

1.09

0.95

0.57

1.05

16.2

7.3

8+years

0.71

0.64

1.13

0.76

0.67

1.05

15.2

5.8

Ghana

 

None

0.88

0.84

1.05

0.95

0.51

1.02

15.7

6.8

1–4 years

0.87

0.78

1.11

0.93

0.54

1.00

15.0

6.6

5–7 years

0.88

0.79

1.11

0.92

0.58

1.05

12.1

6.0

8+years

0.79

0.70

1.13

0.89

0.61

1.03

12.5

5.5

Kenya

 

None

0.91

0.85

1.06

0.88

0.61

1.00

14.8

7.2

1–4 years

0.94

0.80

1.18

0.80

0.65

1.02

15.4

7.7

5–7 years

0.88

0.75

1.18

0.78

0.69

1.04

14.6

7.2

8+years

0.82

0.66

1.24

0.70

0.71

1.00

12.2

5.0

Liberia

 

None

0.96

0.82

1.17

0.97

0.57

1.00

12.9

6.9

1–4 years

1.12

0.71

1.57

0.96

0.59

1.04

11.8

7.7

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

5–7 years

0.79

0.62

1.27

0.90

0.65

1.03

14.9

7.1

8+years

0.82

0.53

1.56

0.75

0.68

0.99

11.0

4.6

Mali

 

None

0.99

0.96

1.03

0.99

0.56

0.99

13.2

7.2

1–4 years

1.05

0.97

1.08

0.96

0.55

1.05

13.4

7.7

5–7 years

0.97

0.96

1.01

0.95

0.59

1.05

11.5

6.5

8+years

0.98

0.94

1.05

0.77

0.68

1.05

10.1

5.4

Ondo State

 

None

0.96

0.92

1.04

0.98

0.43

1.03

16.4

6.8

1–4 years

0.86

0.85

1.02

0.96

0.49

1.05

17.0

7.2

5–7 years

0.89

0.87

1.02

0.96

0.49

1.05

17.4

7.6

8+years

0.77

0.75

1.02

0.90

0.54

0.99

14.3

5.2

Senegal

 

None

0.95

0.91

1.05

0.99

0.54

0.97

14.3

7.0

1–4 years

0.87

0.76

1.14

0.98

0.58

1.05

11.2

5.7

5–7 years

0.78

0.62

1.25

0.88

0.68

1.05

10.4

5.1

8+years

0.60

0.53

1.14

0.74

0.71

1.05

11.3

3.7

Sudana

 

None

0.78

0.76

1.03

0.98

0.57

0.99

13.4

5.8

Primary

0.69

0.67

1.03

0.91

0.62

0.96

13.0

4.9

Secondary+

0.36

0.35

1.03

0.83

0.69

1.05

15.4

3.3

Togo

 

None

0.92

0.89

1.03

0.96

0.50

1.02

16.1

7.2

1–4 years

0.90

0.82

1.09

0.94

0.54

1.05

15.0

7.2

5–7 years

0.78

0.71

1.09

0.91

0.57

0.97

13.2

5.1

8+years

0.75

0.59

1.28

0.81

0.63

1.05

11.0

4.4

Uganda

 

None

0.93

0.82

1.13

0.99

0.60

0.96

15.1

7.9

1–4 years

0.91

0.77

1.18

0.97

0.64

0.98

13.0

7.3

5–7 years

0.87

0.70

1.25

0.96

0.65

1.00

13.0

7.0

8+years

0.84

0.60

1.39

0.87

0.70

0.94

12.0

5.7

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Country and Education Level

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Zimbabwe

 

None

0.93

0.86

1.09

0.73

0.64

1.01

16.2

7.2

1–4 years

0.89

0.80

1.11

0.69

0.65

1.00

16.8

6.7

5–7 years

0.83

0.74

1.13

0.62

0.69

1.01

15.4

5.5

8+years

0.75

0.64

1.17

0.47

0.63

0.98

17.0

3.7

NOTE: Data are national-level DHS.

aDHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

FIGURE 3–4 Relationship between the fertility-inhibiting effects of the proximate determinants and various measures of fertility, by years of education, for Zimbabwe and Liberia. NOTE: PPI: Postpartum infecundability.

tion, with the former having a monotonically negative relation to education in all the populations. Greater contraceptive use and use of more effective methods are generally associated with more education. Cm has a similar relation to education in Senegal, Sudan, Togo, Uganda, and Zimbabwe, but for the other populations, the relation is not uniform. However, when Cm is examined, all but Botswana, Mali, and Ondo State show a monotonically negative effect of union patterns on fertility as education increases.

The inhibitive effect of postpartum infecundability on fertility decreases with education, except in Botswana, Mali, and Zimbabwe. Women with less education tend to breastfeed their children and abstain from sex after birth for longer durations than more-educated women. Even so, there is relatively little variation in Ci across education groups. Ip also varies little by education.

Two examples, shown in Figure 3–4, illustrate how the proximate determinants inhibit fertility when the relationship between fertility and education follows the two above-mentioned patterns: Zimbabwe, with the monotonically inverse relationship, and Liberia, with the inverted U-shape relationship.

Zimbabwe In Zimbabwe, TFR is 7.2 among women with no educa-

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

tion, 6.7 among women with one to four years of schooling, 5.5 among women with five to seven years, and 3.7 among women with eight or more years.

The proximate determinants exhibiting variation across educational groups are Cm (and Cm) and Cc. Most women with no education are married. Cm is lower among women with eight or more years of school. Among more educated women, union patterns—later age at marriage and smaller proportions of women married—result in lower fertility. Cm illustrates the same pattern. Generally, the effect of childbearing among women not in union increases with education level, which may imply more marital instability among better-educated women.

Contraceptive practices vary across the four educational groups, although the national prevalence rate is high when compared with most of sub-Saharan Africa. Prevalence is 32.3 percent among women with no education, and 56.6 percent among the most educated group. Eliminating contraceptive use would increase observed fertility by 5.5 births among women with eight or more years of schooling and by 2.9 births among the least-educated group.

The effect of postpartum nonsusceptibility on inhibiting fertility is large, but the index shows little variation across the four education groups.

Liberia Liberia illustrates the inverted U-shape relationship between fertility and education. The TFR is 6.9 among women with no education. It increases to 7.7 among women with one to four years of schooling, but begins dropping to 7.1 among women with five to seven years, and falls to 4.6 among women with eight or more years of formal schooling.

In Liberia, Cm, Cm, Mo, Cc, and Ci all vary across educational groups and contribute to the observed pattern. Cm decreases monotonically with education, indicating an increasingly stronger effect of union patterns on fertility. Mo is lowest among women with no education and reaches levels greater than 1.5 for two of the education groups, indicating perhaps a relatively higher level of marital instability among educated women or a higher percentage of births occurring outside of union among this group. (Bongaarts’ Cm shows an erratic pattern, reaching a value of 1.12 among women with one to four years of schooling, and contributing to the high level of fertility among this group. In this case, Cm is not interpretable as a fertility-reducing parameter, so our discussion focuses on Cm. Such a finding illustrates the methodological difficulty mentioned in the appendix.)

Contraceptive use increases with education, although its inhibiting effect on fertility is much higher among women with eight or more years of schooling than among other women. The effect of postpartum infecundability decreases monotonically with education, although Ci is the most important proximate determinant in reducing fertility across all education groups.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Change Over Time: Comparison of WFS and DHS Results

A comparison of how the proximate determinants of fertility have changed over time provides an understanding of the biological and behavioral factors underlying fertility change. Four sub-Saharan African countries—Kenya, Senegal, Sudan, and Ghana—have participated in both the WFS and the DHS programs thus far, which allows an examination of changes in these countries from the 1970s to the 1980s. TFRs from these four countries indicate that the total fertility rate declined between the mid-1970s and mid-1980s in Kenya (8.2 to 6.6), Senegal (7.2 to 6.6), and Sudan (6.0 to 4.9), but remained unchanged in Ghana (6.4), as shown in Figure 3–5 and Table 3–7. Comparative estimates for all subgroups are provided in appendix Tables A–1 through A–3 (see Chapter 2 for a more complete discussion of fertility change in sub-Saharan Africa).

Kenya

Kenya illustrates the potential role of contraceptive practices in reducing fertility, as shown in Figure 3–5. The most important determinant of the change in TFR of 1.5 births was a drop in the index of contraception, from

FIGURE 3–5 Comparison of WFS and DHS data of the relationship between the fertility-inhibiting effects of the proximate determinants and various measures of fertility. NOTE: WFS and DHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–7 Proximate Determinants of Fertility

Country and Survey

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Model Estimate of Total Fecundity Rate, TF

Observed TFR

Ghana

 

WFS

0.88

0.81

1.08

0.95

0.56

1.01

13.4

6.4

DHS

0.85

0.77

1.11

0.93

0.55

1.02

14.3

6.4

Kenya

 

WFS

0.91

0.81

1.12

0.96

0.64

1.00

14.7

8.2

DHS

0.86

0.73

1.17

0.80

0.66

1.01

14.4

6.6

Senegal

 

WFS

0.94

0.89

1.05

0.99

0.65

0.99

12.1

7.2

DHS

0.90

0.84

1.07

0.97

0.64

0.98

12.1

6.6

Sudana

 

WFS

0.80

0.78

1.03

0.96

0.63

0.93

13.3

6.0

DHS

0.68

0.66

1.03

0.94

0.60

0.99

12.9

4.9

NOTE: Data are national-level WFS and DHS.

aWFS and DHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

0.94 to 0.76. Underlying this decline was a substantial increase in contraceptive prevalence across all subgroups between the two surveys. At the time of the DHS (1988–1989), contraceptive prevalence in Kenya was estimated at 26.8 percent, with 5.2 percent using pills, 3.7 percent using IUDs, 4.7 percent using sterilization, and 13.2 percent using other, mainly traditional, methods. At the time of the WFS (1977–1978), prevalence was 6.7 percent, with 2.0 percent using pills, 0.7 percent using IUDs, 0.9 percent being sterilized, and 3.1 percent using traditional methods. The effect of the increase in contraceptive use on TF was substantial. At the time of the WFS, Cc contributed to a reduction of 0.4 birth in observed fertility, but at the DHS, it contributed to a reduction of 1.9 births. However, even though contraceptive use rose, the use-effectiveness of the method mix did not change.

Over the same period, the inhibiting effects of marriage on fertility increased only slightly, as Cm decreased from 0.91 to 0.86. Mo increased between the surveys, indicating an increase in nonunion fertility. The Ci index changed little over the intervening period. However, at both times, Ci had the largest fertility-inhibiting effect of all the proximate determinants. The increase in the index reflects the decrease in the nonsusceptible period from 12.7 to 11.7 months.

Senegal3

Between the WFS and the DHS, the observed TFR in Senegal declined by 0.6 birth from 7.2 to 6.6 (see Table 3–7). The change appears to be attributable principally to a change in marriage patterns. For the WFS, the Cm index was 0.94, which contributed a reduction of 0.5 birth relative to TF. For the DHS, Cm had a value of 0.90, which was associated with a reduction of 0.7 birth. This change probably reflects an increase in the proportion of never-married women from 12.9 percent in the WFS to 18.8 percent in the DHS (Ndiaye et al., 1988). Although marriage patterns appear to have changed, the changes are very small and, therefore, do not give solid evidence of a real change in marriage or fertility. The low values of Mo result in very small differences between Cm and Cm, indicating that most childbearing occurs within union.

3  

The WFS for Senegal did not ask women whether they were currently amenorrheic, so the nonsusceptible period was imputed by using a formula based on the duration of breastfeeding that was developed by Bongaarts and Potter (1983) (see Technical Notes at the end of the chapter).

To make the analysis based on DHS and WFS data comparable, the DHS Ci used in Figure 3–5 and Table 3–7 was also imputed by using the same formula.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Comparing Cc and Ip reveals only small changes in the calculated values over time. Cc is 0.99 for the WFS and 0.97 for the DHS, indicating insignificant change in contraceptive use or method mix. Contraceptive prevalence was less than 5 percent at the times of both surveys, and the method mixes were heavily dependent on traditional methods.

Sudan

A comparison of the changes in the proximate determinants of fertility of Sudan indicates that the TFR decreased by 1.1 births from 6.0 to 4.9. The index of marriage appears to be the factor that changed most significantly over the intervening period. At the time of the WFS, Cm was 0.80, indicating marriage patterns contributed a reduction of 1.5 births relative to the TF. By the time the DHS was conducted, the Cm was 0.68, resulting in a fertility reduction of 2.3 births. The values of Mo were very low, indicating that almost all childbearing occurs within union.

The DHS reports that almost all Sudanese women marry during their lifetime; by ages 45–49, 99 percent of all women have married (Sudan, 1991). Although the DHS was based on a sample of ever-married women, the household survey associated with DHS indicated that a substantial increase in the average age at marriage had occurred in recent years. The proportion of never-married women had risen by 12 percentage points in the 10 years since the Sudan Fertility Survey.

Postpartum infecundability had a large effect in reducing fertility at both times, but it did not exhibit as much of a decrease between the two periods as Cm. Contraceptive use was low at the time of both surveys (4.5 percent in the WFS and 8.2 percent in the DHS), resulting in an insignificant decline in Cc.

Ghana

The final sub-Saharan African country for which both the DHS and WFS were available at the time of our analysis is Ghana. TFR remained unchanged between the two surveys (6.4), and the proximate determinants inhibited fertility to roughly the same extent for both surveys. The most important factor at both times was postpartum infecundability, with values of 0.56 and 0.55, respectively. Cm was the second most important proximate determinant, with values falling in the mid-range of those observed in the region. Mo was about 1.1 for both surveys, indicating a consistent level of nonunion fertility. In Ghana, contraceptive use made the smallest contribution to limiting fertility of any of the proximate determinants even though the index of contraception is lower in Ghana than in many other sub-Saharan African populations studied; only Zimbabwe, Kenya, and Botswana are lower.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

CONCLUSION

The proximate determinants framework illustrates how the underlying factors that determine levels of fertility vary across populations and time. The original model developed by Bongaarts assumes that all childbearing occurs in marriage. In the African setting, where entry into union is a process, this assumption is not necessarily correct. In response to this situation, we developed the measures Mo and Cm, which further refine the effect of union patterns on fertility.

The index of postpartum infecundability is generally the most significant inhibitor of fertility. Although prolonged breastfeeding and postpartum abstinence are not universal in sub-Saharan Africa, they generally play an important role in spacing births and reducing total fertility.

Marriage patterns also reduce fertility substantially in many populations. The index of marriage Cm tends to be lowest in the youngest age groups, indicating that marriage is less common among women between ages 15 and 24. The index is generally closer to unity among women living in rural areas because marriage is more common and earlier there. Among women with more education, union patterns are less supportive of high fertility.

Mo illustrates the proportion of fertility occurring outside union. In some populations, such as Mali, Ondo State, and northern Sudan, Mo is very low (close to 1) indicating that most fertility occurs in union. However, in some populations, such as Botswana, Liberia, and Uganda, Mo values are higher, suggesting that a substantial proportion of total fertility occurs outside of union. Cm is substantially lower than Cm in populations where Mo is large, indicating that Cm underestimates the effect of union patterns on fertility in these populations.

Contraceptive use in sub-Saharan Africa is fairly low; notable exceptions are Botswana, Kenya, and Zimbabwe. Generally contraceptive prevalence is higher among women in the middle age group. Younger women may want to bear their children rapidly to demonstrate their value as wives. Once women begin to achieve their desired family size, they may begin to use family planning methods (Mhloyi, 1988). In a few populations, contraception has its most significant effect on inhibiting fertility among the oldest age group, perhaps because these women are trying to end their childbearing. On the other hand, in most of these countries, contraceptive prevalence is still relatively low compared to other parts of the developing world. Cc has a greater fertility-inhibiting effect among women with eight or more years of education than among women with fewer years of schooling, and among women living in urban rather than rural areas.

The index of sterility poses a difficulty in our analysis. Although Frank (1983a) estimates that primary sterility generally affects at least 3 percent

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

of women in developing countries, many of the populations analyzed here show much lower levels. In Botswana, Mali, Senegal, Sudan, and Uganda, the index values were less than 1, indicating that primary sterility was greater than 3 percent. In the remaining countries, less than 3 percent of women between the ages of 40 and 49 had never borne a child, indicating that the calculation for estimating Ip may need to be reconsidered. However, it is also possible that women are underreporting primary sterility by a considerable margin, given the stigma attached to being childless in many parts of sub-Saharan Africa (Larsen, 1989).

The failure to include abortion in our analysis, as a result of a lack of adequate data, is a major shortcoming. It is clear that more data on the incidence of abortion are needed to measure its effect on fertility. This is an important underresearched area, particularly given reports that point to increasing levels of abortion. However, the fact that the total fecundity rates we obtain fall largely within the theoretical range suggested by Bongaarts and Potter may indicate that indeed abortion is not as yet an important fertility-limiting factor in Africa.

Comparing the WFS and DHS for four sub-Saharan African countries illustrates how fertility and its determinants have changed over time. In Kenya, the change in fertility is attributable principally to a change in contraceptive use. In Senegal and Sudan, changing marriage patterns seem to underlie the fertility declines; however, the change in Senegal is too small to provide solid evidence of real changes in marriage patterns or fertility. In Ghana, little change in fertility or its determinants is observed. Although some of the indices changed in importance over time, postpartum infecundability continues to be the greatest fertility-inhibiting proximate determinant for all four populations.

TECHNICAL NOTES

Data Sources

  • DHS Ondo State, Nigeria A question on whether the survey respondent was currently amenorrheic was not asked, so the question (V215) on the time since the woman’s last menstrual period was used to estimate mean months of amenorrhea.

  • DHS Sudan The survey was conducted only for ever-married women of reproductive age, so the denominators for the total fertility rates (TFRs) were estimated by using a set of weights developed by the Institute for Resource Development/DHS to estimate the total number of never-married and ever-married women. Because these weights were developed only for specific subgroups of the population, education categories for Sudan differ from education categories used for other DHS countries in the analysis.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
  • WFS Senegal A question on whether the survey respondent was currently amenorrheic was not asked, so the postpartum nonsusceptible period was estimated as where B equals the mean number of months of breastfeeding for the 36 months prior to the survey. Bongaarts and Potter (1983) estimated this regression equation from data on breastfeeding and amenorrhea from about 25 countries.

  • WFS Sudan This survey, like the DHS, was conducted only for ever-married women of reproductive age. Because the weights developed for the denominators of the TFRs were not included in the WFS Sudan standard recode file, the TFRs used in this analysis are taken from Volume 1 of The Sudan Fertility Survey 1979, Principal Report, Vol. 1 (Sudan, 1982).

Derivation of Indices

Index of Marriage

The index of marriage is calculated as

where

TFR=

the average total number of births a woman would have in her lifetime at current age-specific fertility rates (ASFRs), and

TMFR=

the average total number of births a woman in union from age 15 to 49 would have at current age-specific marital fertility rates.

Both rates are estimated for the 4 years prior to the survey. Four-year fertility rates are estimated instead of 5-year rates because of the underreporting of births in the fifth year preceding some of the DHS (Institute for Resource Development, 1990). The TMFR was estimated for women currently in union.

Measured Births Outside Marriage and Adjusted Index of Marriage

The measured births outside marriage and the adjusted index of marriage are calculated, respectively, as

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

and

where

TUFR is the sum of the age-specific union fertility rates (ASUFRS), and

Because the DHS does not give a complete marriage history, it is impossible to estimate lengths of marital disruption. Therefore, in the estimation of TUFR, it is assumed that there is no marital disruption from the date of first union if the woman is still currently married. If a marital disruption had in fact occurred, unbeknownst to us, we are assuming that it was fairly short, given that most women remarry quickly (Mhloyi, 1988). This assumption results in the classification of births occurring to currently married women during a past period of marital disruption as occurring in union, resulting in an overestimation of the TUFR.

On the other hand, births to women whose status at the time of the survey is divorced or widowed are assumed to be births that occurred outside of union. This assumption leads to an underestimation of TUFR, resulting in an overestimation of Mo (or the effect of childbearing outside of union on fertility), as well as an overestimation of the effect of union patterns on fertility (i.e., an underestimation of Cm). These two errors will to a large extent cancel each other out.

As with the TFRs and TMFRs, the TUFRs are calculated for the four years prior to the survey.

Index of Contraception

The index of contraception is calculated as

Cc=1–1.08ue,

where u is the current contraceptive use prevalence rate among women in sexual union, and e is the average use-effectiveness of contraception.

Abstinence is excluded as a method because most of the women who reported using abstinence as a contraceptive method were practicing postpartum abstinence, which is captured in the Ci index. Periodic abstinence, however, is included as a method.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

The average use-effectiveness of a method is calculated as the weighted average of the method-specific use-effectiveness levels, with the weights equal to the proportion of women using a given method. The levels used, adapted from Bongaarts and Potter (1983) (who employed use-effectiveness levels from a study by Laing (1978) in the Philippines), are shown below. We have divided one of the categories, other methods, into other modern methods and traditional methods in order to give traditional methods a lower use-effectiveness index of 0.30.

Pill

0.90

IUD

0.95

Sterilization

1.00

Other modern methods

0.70

Traditional methods

0.30

Index of Postpartum Infecundability

The index of postpartum infecundability is calculated as

Ci=20/(18.5+i),

where i is the mean number of months of postpartum infecundability (estimated as the mean number of months of postpartum amenorrhea or abstinence, whichever is longer) for women in union.

The mean number of months of postpartum infecundability is estimated by using the prevalence/incidence method. In this analysis, i is the period of nonsusceptibility, calculated as the number of mothers either amenorrheic or abstaining at the time of the survey (prevalence) divided by the average number of births per month over the last 36 months (incidence).

Index of Sterility

The index of sterility is estimated as

Ip=(7.63–.11s)/7.3,

where s is the proportion of ever-married women between ages 40 and 49 who have never had any children.

Bongaarts et al. (1984) used the percentage of women aged 45–49 who are childless. In this analysis, the percentage of childless women aged 40– 49 is used instead to increase the number of women in each subgroup and reduce the standard error in estimating s. It is assumed that most women have had their first birth by 40 years of age in sub-Saharan Africa.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Effects of Indices

To express the effects of each index in births per woman, the following calculations are used (Bongaarts, 1982).

  • The effect of marriage patterns equals TMFR—TFR, where TMFR equals TFR/Cm.

  • The effect of contraception equals TNMF—TMFR, where TNMF equals TFR/(CmCc).

  • The effect of postpartum infecundability and primary sterility equals TF—TNMF, where TF equals TFR/(CmCcCiIp). The effect of postpartum infecundability alone equals TFR/(CmCcCi) —TNMF. The effect of primary sterility alone equals TF—TFR/(CmCcCi). When Ip is greater than 1, the effect of postpartum infecundability is estimated as TF—TNMF; that is, Ip is set equal to 1, because a number greater than 1 is not interpretable in the proximate determinants framework. In these cases, the effect of postpartum infecundability is slightly underestimated.

Care should be taken in interpreting these effects expressed in births per woman, because the number of births estimated depends on the order in which they are calculated. For example, by using the formulas outlined above, the effects of the proximate determinants for Botswana would be as follows (see Table 3–2 and Figure 3–1) in terms of number of births:

Proximate determinant

Number of births

Marriage patterns

0.74

Contraception

2.45

Postpartum infecundability

4.84

Primary sterility

0

If the order in which each variable is calculated is reversed, the results would be as follows (number of births):

Primary sterility

(TFR/Ip—TFR)

0

Postpartum infecundability

(TFR/(IpCi) —TFR/Ip)

2.92

Contraception

(TFR/(IpCiCc) —TFR/(IpCi))

3.38

Marriage patterns

(TFR/(IpCiCcCm) —TFR/(IpCiCc))

1.68

Therefore, the order of estimation matters a great deal. However, because Bongaarts et al. (1984) used the first-outlined approach in their work, we have done the same for consistency.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

The reduction in average number of births per woman can also be expressed in terms of percentages. In Ondo State, Ci=0.47, which indicates that Ci reduces fertility to 47 percent of what it would otherwise have been in the absence of postpartum breastfeeding and abstinence.

APPENDIX

The tables that follow present the proximate determinants of fertility by age (Table 3–A.1), by urban and rural residence (Table 3–A.2), and by level of education (Table 3–A.3). WFS and DHS data are presented.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–A.1 Proximate Determinants of Fertility by Age and by Survey

Country and Survey

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Observed TFR

Ghana WFS

 

National

0.88

0.81

1.08

0.95

0.56

1.01

6.4

15–24

0.77

0.68

1.14

0.96

0.55

1.01

2.0

25–34

0.96

0.91

1.06

0.93

0.58

1.01

2.6

35–49

0.92

0.86

1.06

0.96

0.55

1.01

1.8

DHS

 

National

0.85

0.77

1.11

0.93

0.55

1.02

6.4

15–24

0.71

0.60

1.20

0.95

0.53

1.02

1.9

25–34

0.95

0.89

1.07

0.93

0.57

1.02

2.6

35–49

0.90

0.84

1.07

0.91

0.54

1.02

1.8

Kenya WFS

 

National

0.91

0.81

1.12

0.96

0.64

1.00

8.2

15–24

0.81

0.66

1.23

0.97

0.64

1.00

2.6

25–34

0.97

0.91

1.07

0.95

0.65

1.00

3.2

35–49

0.95

0.88

1.08

0.95

0.62

1.00

2.3

DHS

 

National

0.86

0.73

1.17

0.80

0.66

1.01

6.6

15–24

0.75

0.56

1.35

0.87

0.66

1.01

2.3

25–34

0.94

0.85

1.11

0.79

0.68

1.01

2.7

35–49

0.92

0.86

1.08

0.76

0.64

1.01

1.6

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Senegal WFS

 

National

0.94

0.89

1.05

0.99

0.65

0.99

7.2

15–24

0.87

0.79

1.09

0.99

0.63

0.99

2.5

25–34

0.98

0.95

1.03

0.99

0.66

0.99

3.0

35–49

0.97

0.94

1.03

1.00

0.66

0.99

1.7

DHS

 

National

0.90

0.84

1.07

0.97

0.64

0.98

6.6

15–24

0.81

0.71

1.14

0.99

0.63

0.98

2.2

25–34

0.95

0.91

1.04

0.96

0.65

0.98

2.7

35–49

0.95

0.92

1.03

0.97

0.63

0.98

1.7

Sudan WFS

 

Northern

0.80

0.78

1.03

0.96

0.63

0.93

6.0

15–24

0.61

0.60

1.03

0.97

0.64

0.93

1.9

25–34

0.92

0.89

1.02

0.95

0.62

0.93

2.7

35–49

0.93

0.91

1.02

0.97

0.65

0.93

1.5

DHS

 

Northern

0.68

0.66

1.03

0.94

0.60

0.99

4.9

15–24

0.44

0.43

1.03

0.96

0.61

0.99

1.2

25–34

0.81

0.79

1.03

0.93

0.60

0.99

2.4

35–49

0.87

0.85

1.02

0.93

0.59

0.99

1.3

NOTE: Data are national-level WFS and DHS.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–A.2 Proximate Determinants of Fertility by Urban and Rural Residence and by Survey

Country and Survey

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Observed TFR

Ghana

 

WFS

Urban

0.86

0.78

1.10

0.91

0.62

1.00

5.7

Rural

0.95

0.88

1.08

0.96

0.54

1.01

6.8

DHS

 

Urban

0.78

0.70

1.11

0.89

0.60

1.01

5.3

Rural

0.90

0.83

1.08

0.95

0.54

1.03

6.9

Kenya

 

WFS

Urban

0.84

0.73

1.17

0.90

0.69

0.90

6.1

Rural

0.92

0.82

1.11

0.96

0.64

1.00

8.4

DHS

 

Urban

0.82

0.66

1.24

0.74

0.70

0.96

4.7

Rural

0.87

0.75

1.17

0.81

0.66

1.01

7.0

Senegal

 

WFS

Urban

0.86

0.79

1.08

0.98

0.67

0.99

6.6

Rural

0.98

0.94

1.04

1.00

0.64

0.99

7.5

DHS

 

Urban

0.79

0.71

1.12

0.92

0.69

0.96

5.6

Rural

0.98

0.93

1.04

0.99

0.62

0.99

7.3

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Sudana

 

WFS

Urban

0.72

0.69

1.03

0.90

0.70

0.95

5.1

Rural

0.83

0.81

1.03

0.99

0.61

0.93

6.4

DHS

 

Urban

0.55

0.53

1.03

0.87

0.65

0.98

3.7

Rural

0.71

0.70

1.01

0.98

0.58

1.00

5.2

NOTE: Data are national-level WFS and DHS.

aWFS and DHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

TABLE 3–A.3 Proximate Determinants of Fertility by Level of Education and by Survey

Country and Survey

Index of Marriage, Cm

Adjusted Index of Marriage, Cm

Measure of Births Outside Marriage, Mo

Index of Contraception, Cc

Index of Postpartum Infecundability, Ci

Index of Sterility, Ip

Observed TFR

Ghana

 

WFS

None

0.93

0.88

1.06

0.98

0.54

1.01

6.7

1–4 years

0.90

0.80

1.13

0.95

0.58

1.01

6.9

5–7 years

0.88

0.78

1.13

0.93

0.62

1.05

7.1

8+years

0.83

0.76

1.09

0.88

0.60

0.99

5.3

DHS

 

None

0.88

0.84

1.05

0.95

0.51

1.02

6.8

1–4 years

0.87

0.78

1.11

0.93

0.54

1.00

6.6

5–7 years

0.88

0.79

1.11

0.92

0.58

1.05

6.0

8+years

0.79

0.70

1.13

0.89

0.61

1.03

5.5

Kenya

 

WFS

None

0.96

0.87

1.10

0.98

0.61

0.99

8.2

1–4 years

0.93

0.84

1.11

0.96

0.64

1.03

9.0

5–7 years

0.90

0.79

1.13

0.94

0.69

1.05

7.9

8+years

0.83

0.75

1.11

0.83

0.70

1.03

7.0

DHS

 

None

0.91

0.85

1.07

0.88

0.61

1.00

7.2

1–4 years

0.94

0.80

1.18

0.80

0.65

1.02

7.7

5–7 years

0.88

0.75

1.18

0.78

0.69

1.04

7.2

8+years

0.82

0.66

1.24

0.70

0.71

1.00

5.0

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Senegal

 

WFS

None

0.96

0.92

1.04

1.00

0.64

0.99

7.4

Primary

0.98

0.89

1.10

0.96

0.67

1.05

7.1

Secondary+

0.56

0.53

1.06

0.88

0.71

1.05

3.0

DHS

 

None

0.95

0.91

1.05

0.99

0.63

0.97

7.0

Primary

0.83

0.69

1.20

0.93

0.67

1.05

5.7

Secondary+

0.59

0.52

1.14

0.75

0.74

1.05

3.6

Sudana

 

WFS

None

0.84

NA

NA

0.99

0.61

0.93

6.3

Primary incomplete

0.95

NA

NA

0.90

0.69

0.96

7.6

Primary complete+

0.88

NA

NA

0.73

0.79

0.88

6.0

DHS

 

None

0.78

0.76

1.03

0.98

0.57

0.99

5.8

Primary

0.69

0.67

1.03

0.91

0.62

0.96

4.9

Secondary+

0.36

0.35

1.03

0.83

0.69

1.05

3.3

NOTE: Data are national-level WFS and DHS.

aWFS and DHS data for Sudan refer only to northern Sudan.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

REFERENCES

Acsadi, G.T.F., and G.Johnson-Acsadi 1990 Demand for children and for childspacing. Pp. 155–185 in G.T.F.Acsadi, G. Johnson-Acsadi, and R.Bulatao, eds., Population Growth and Reproduction in Sub-Saharan Africa: Technical Analyses of Fertility and Its Consequences. Washington, D.C.: The World Bank.

Adegbola, O., H.J.Page, and R.Lesthaeghe 1977 Breast-feeding and postpartum abstinence in metropolitan Lagos. Paper presented at the meeting of the Population Association of America, St. Louis, April 21–23.


Bongaarts, J. 1982 The fertility-inhibiting effects of the intermediate fertility variables. Studies in Family Planning 13(6/7):179–189.

Bongaarts, J., and R.G.Potter 1983 Fertility, Biology, and Behavior: An Analysis of the Proximate Determinants. New York: Academic Press.

Bongaarts, J., O.Frank, and R.Lesthaeghe 1984 The proximate determinants of fertility in sub-Saharan Africa. Population and Development Review 10(3):511–537.


Caldwell, J.C., and P.Caldwell 1983 The demographic evidence for the incidence and cause of abnormally low fertility in tropical Africa. World Health Statistics Quarterly 36(1):2–34.

Cochrane, S.H. 1979 Fertility and Education: What Do We Really Know? Baltimore, Md.: Johns Hopkins University Press.

1983 Effects of education and urbanization on fertility. Pp. 587–625 in R.A.Bulatao and R.D.Lee, eds., Determinants of Fertility in Developing Countries, Vol. 2. New York: Academic Press.

Coeytaux, F.M. 1988 Induced abortion in sub-Saharan Africa: What we do and do not know. Studies in Family Planning 19(3):186–190.


Davis, K., and J.Blake 1956 Social structure and fertility: An analytic framework. Economic Development and Cultural Change 4(4):211–235.


Farley, T.M.M., and E.M.Besley 1988 The prevalence and aetiology of infertility. Pp. 15–30 in African Population Conference, Dakar 1988, Vol. 1. Liège: International Union for the Scientific Study of Population.

Frank, O. 1983a Infertility in sub-Saharan Africa: Estimates and implications. Population and Development Review 9(1):137–144.

1983b Infertility in Sub-Saharan Africa. Working Paper of the Center for Policy Studies, No. 97. New York: The Population Council.


Goldman, N., S.O.Rutstein, and S.Singh 1985 Assessment of the Quality of Data in 41 WFS Surveys: A Comparative Approach. WFS Comparative Studies, No. 44. Voorburg, Netherlands: International Statistical Institute.


Institute for Resource Development 1990 An Assessment of DHS-I Data Quality. Demographic and Health Surveys Methodological Reports No. 1. Columbia, Md.: Institute for Resource Development/ Macro Systems, Inc.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

Laing, J. 1978 Estimating the effects of contraceptive use on fertility. Studies in Family Planning 9(6):150–175.

Larsen, U. 1989 A comparative study of the levels and the differentials of sterility in Cameroon, Kenya, and Sudan. Pp. 167–211 in R.Lesthaeghe, ed., Reproduction and Social Organization in Sub-Saharan Africa. Berkeley: University of California Press.


Menken, J. 1984 Estimating proximate determinants: A discussion of three methods proposed by Bongaarts, Hobcraft and Little, and Gaslonde and Carrasco. Paper prepared for the IUSSP Seminar on Integrating Proximate Determinants into Analysis of Fertility Levels and Trends. International Union for the Scientific Study of Population and World Fertility Survey, London.

Mhloyi, M.M. 1988 The determinants of fertility in Africa under modernization. Pp. 2.3.1–2.3.22 in African Population Conference, Dakar 1988, Vol. 1. Liege: International Union for the Scientific Study of Population.


Ndiaye, S., I.Sarr, and M.Ayad 1988 Enquête Demographique et de Santé au Senegal 1986. Dakar: Ministère de l’Economie et des Finances, Direction de la Statistique, Division des Enquêtes et de la Demographic; Columbia, Md.: Institute for Resource Development/Westinghouse.


Page, H.J., and A.J.Coale 1972 Fertility and child mortality south of the Sahara. Pp. 51–67 in S.H.Ominde and C. Ejiogu, eds., Population Growth and Economic Development in Africa. London: Heinemann.


Reinis, K.I. 1992 The impact of the proximate determinants of fertility: Evaluating Bongaarts’s and Hobcraft and Little’s methods of estimation. Population Studies 46:309–326.

Romaniuk, A. 1980 Increases in natural fertility during the early stages of modernization—Evidence from an African case study: Zaire. Population Studies 2(34):293–310.

Rutenberg, N., M.Ayad, L.H.Ochoa, and M.Wilkinson 1991 Knowledge and Use of Contraception. DHS Comparative Studies, No. 6. Columbia, Md.: Institute for Resource Development.


Sudan 1982 The Sudan Fertility Survey 1979, Principal Report, Vol. 1. Khartoum: Department of Statistics of the Sudanese Ministry of National Planning.

1991 Sudan Demographic and Health Survey 1989/1990. Khartoum: Ministry of Economic and National Planning; Columbia, Md.: Institute for Resource Development/Macro International, Inc.


Working Group on the Social Dynamics of Adolescent Fertility 1993 Social Dynamics of Adolescent Fertility in Sub-Saharan Africa. C.H.Bledsoe and B.Cohen, eds. Panel on the Population Dynamics of Sub-Saharan Africa, Committee on Population, National Research Council. Washington, D.C.: National Academy Press.


van de Walle, F., and K.Omideyi 1988 The cultural roots of African fertility regimes. Pp. 2.2.35–2.2.52 in African Population Conference, Dakar 1988, Vol. 1. Liege: International Union for the Scientific Study of Population.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×

van de Walle, E., and F.van de Walle 1988 Postpartum sexual abstinence in tropical Africa. Paper presented at the International Union for the Scientific Study of Population Seminar on the Biomedical and Demographic Determinants of Human Reproduction. Johns Hopkins University, Baltimore, Md.


Zeidenstein, S. 1979 Learning about rural women. Studies in Family Planning 10(11/12):309–312.

Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 68
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 69
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 70
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 71
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 72
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 73
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 74
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 75
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 76
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 77
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 78
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 79
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 80
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 81
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 82
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 83
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 84
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 85
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 86
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 87
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 88
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 89
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 90
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 91
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 92
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 93
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 94
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 95
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 96
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 97
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 98
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 99
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 100
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 101
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 102
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 103
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 104
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 105
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 106
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 107
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 108
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 109
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 110
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 111
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 112
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 113
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 114
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 115
Suggested Citation:"3 The Approximate Determinants of Fertility." National Research Council. 1993. Demographic Change in Sub-Saharan Africa. Washington, DC: The National Academies Press. doi: 10.17226/2207.
×
Page 116
Next: 4 Recent Trends in Marriage Ages »
Demographic Change in Sub-Saharan Africa Get This Book
×
Buy Paperback | $100.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

This overview includes chapters on child mortality, adult mortality, fertility, proximate determinants, marriage, internal migration, international migration, and the demographic impact of AIDS.

  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. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

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

    « Back Next »
  7. ×

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

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    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!