I ;lllll lllll lllll lllll lllll lllll lllll 111111 111111111 1111 
M060067984 
North-West Unh,erslty 
Mafikeng Campus Library 
-
Patterns and determinants of subsequent birth amongst the 
ethnic groups in the black population of South Africa 
GA Mothelesi 
orcid.org/0000-0002-1363-3064 
Mini-dissertation submitted in partial fulfilment of the requirements 
for the degree Master of Social Science in Population and Sustainable 
Development at the North-West University 
Supervisor: Dr K Mhele 
Graduation ceremony: October 2019 
Student number: 24294233 
LIBRARY 
MAFlf<ENG CAMPUS 
CALL NO.: 
2020 -01- 0 8 
ACC.NO.: 
NORTH-WEST UNiVERSITY j 
DECLARATION 
I, Glender Amantle Mothelesi , hereby declare that the research titled " Patterns and determinants 
of subsequent birth amongst ethnic groups in the black population of South Africa" is my own 
work and that I have acknowledged all sources used. No part of this work has been previously 
submitted for a degree or examination at this or any other university. 
ACKNOWLEDGMENTS 
I acknowledge everyone who contributed to the completion and success of my work. I give my sincere 
thanks to my supervisor DR K.E Mhele for his patience, advice along the academic journey we had 
together. I again extend my gratitude to Dr. P.O. Marumo for being my academic mentor and for always 
criticizing and motivating me. Special thanks to my family , and friends (Tsoselesto, Dikobe) for always 
being there for me during the time of my study. Above all , I give thanks to God for being my pillar of 
strength throughout my studies. 
ii 
ABSTRACT 
Background: Progression to the next child (subsequent birth) forms a vital part of fertility as this 
determines fertility behavior for women or couples. This also contributes to the country's population 
size, growth, and structure. Women from different population groups have different fertility behavior 
which affects demographic variables. In South Africa, the black population has over the years been 
having the highest fertility levels. In the contemporary world, fertility levels of the black population 
have remained the highest even though these levels were declining. Therefore, the main objective of 
this study was to compare and contrast fertility patterns specifically focusing on progression to the next 
child (subsequent birth) in different main ethnic groups within the black population in South Africa. 
Methods: The study focused on women aged 15-54 in the selected ethnic groups namely, Zulus, 
Xhosas, Sothos, Pedis and Tswanas with the total population 726 855. Fertility was measured using the 
Parity Progression Ratio (PPR) and the Binary Regression (Odds Ratio) . These methods were used to 
analyze the patterns and to determine progression to the third and more children. 
Results: The results of this study identified that among the African ethnic groups , Zulus have more 
children than Sotho, Pedi, Xhosa, and Batswana. The mean parities of women in these ethnic groups 
rage from I to 3. Two models were used in both adjusted and unadjusted versions. In the adjusted 
model, most variables like the province of residence were statistically significant but on the unadjusted 
odds ratio some variables changed to not being statistically significant like ethnicity 
Conclusion: Zooming on fertility from ethnicity levels rather than population group gives more clarity 
on the fertility behavior of women within the black population The decision taken by a couple and 
women to begin their fertility and progress to the next child is influenced by various factors such as 
marriage type and religion. Therefore, women progressing to the next child during their childbearing 
ages determine how many children they would ultimately have at the end of their childbearing ages (15-
49). Given the above, the study recommended that the government strengthen the policy to reduce the 
fertility of these ethnic groups and keep it below the replacement level. 
iii 
TABLE OF CONTENTS 
DECLARA TION. ........................................................................................................................ I 
ACKNOWLEDGEMENTS ...................................................................................................... 11 
ABSTRACT ................................................................................................................................ 11 
CHAPTER 1 INTRODUCTION AND BACKGROUND TO THE STUDY ......................... 1 
1.1 INTRODUCTION ................................................................................................................ 1 
1.1.1 Parity Progression Ratio method ...................................................................................... 2 
1.1.2 Definition of Parity Progression Ratio ............................................................................. 2 
1.1.3 The evolution of Parity Progression Ratio ......................................................................... 2 
1.1.4 How has PPR been used ..................................................................................................... 3 
1.2 Problem Statement. ............................................................................................................... 4 
1.3 Rationale for study ................................................................................................................ 5 
1.4 Main objective of study ........................................................................................................ 6 
1.4.1 Specific objective ................................................................................................................ 6 
CHAPTER 2 THEORETICAL FRAMEWORK AND LITERATURE REVIEW ............. 7 
2.1 Introduction ........................................................................................................................... 7 
2.2 Factors associated with fertility decline .............................................................................. 7 
2.2.1 Demographic factors .......................................................................................................... 7 
2.2.1.1 Age ............................................................................................................................. 7 
2.2.1.2 Ethnicity ........................................................................................................................... 7 
iv 
2.2.2 Socioeconomic factors ........................................................................................................ 8 
2.2.2.1 Sex Preference ................................................................................................................. 8 
2.2.2.2 Marital Status .................................................................................................................. 8 
2.2.2.3 Religion ............................................................................................................................ 9 
2.2.2.4 Education ......................................................................................................................... 9 
2.2.2.5 Place of residence .......................................................................................................... 10 
2.2.2.6 Province of residence .................................................................................................... 10 
2.3 The evolution offertility decline in South Africa ............................................................. 11 
2.3.1 Trends of Fertility levels within South African population groups ............................. 12 
2.4 Theoretical Framework .................................................................................................. 12 
CHAPTER 3 METHODOLOGY ............................................................................................ 14 
3.1 Introduction ......................................................................................................................... 14 
3.2 Data sources ......................................................................................................................... 14 
3.3 Sampling method ................................................................................................................ 14 
3.4 Study population ................................................................................................................... 15 
3.5 Methods of analysis ............................................................................................................. 15 
3.5.1 Correction of parity-children ever born data ................................................................ 16 
3.5.2 Binary Regression .............................................................................................................. 17 
3.6 Variables ............................................................................................................................. 17 
3.5.1 Dependent variables ......................................................................................................... 17 
3.6.2 Independent variables ...................................................................................................... 17 
3.7 Limitations of study ............................................................................................................ 17 
V 
3.7.1 Language variable ............................................................................................................ 17 
CHAPTER 4 ANALYSIS AND PRESENTATION OF RESULTS ..................................... 19 
4.1 Introduction ......................................................................................................................... 19 
4.2 El-Baldry correction for data on children ever born ....................................................... 19 
4.3 Parity Progression Ratio .................................................................................................... 26 
4.4 The odds ratio ........................................................................................................................ 29 
4.5 Summary of results ............................................................................................................. 33 
CHAPTER 5 DISCUSSION AND RECOMMENDATIONS ............................................... 35 
5.1 Introduction .......................................................................................................................... 35 
5.2 Discussion ............................................................................................................................ 35 
5.3 Recommendations ............................................................................................................... 36 
vi 
LIST OFTA BLES 
Table 1: Total proportions of childless and parity not stated ............................................... 20 
Table 2: Total number of women by age group and parity and mean parity .................... .41 
Table 3: Total births reported in the last year before the survey and ASFR. ..................... .42 
Table 4: Total proportions ever attaining parity ................................................................... 26 
Table 5: Total Parity Progression Ratio of selected ethnic groups ....................................... 27 
Table 6: The odds ratios by different socio-economic variables in study ............................ 31 
Table 7: Number of Zulu women by age group and parity and mean parity. ...................... 43 
Table 8: Number of Xhosa women by age group and parity and mean parity .................... .44 
Table 9: Number of Pedi women by age group and parity and mean parity. ...................... .45 
Table 10: Number of Sotho women by age group and parity and Mean Parity. .................. 46 
Table 11: Number of Tswana women by age group and parity and Mean Parity .............. .47 
Table 12: Zulu Births reported in the last year before the survey and ASFR. ................... .48 
Table 13: Xhosa births reported in the last year before the survey and ASFR. ................. .49 
Table 14: Pedi births reported in the last year before the survey and ASFR. ..................... 50 
Table 15: Sotho births reported in the last year before the survey and ASFR. ................... 51 
Table 16: Tswana Births reported in the last year before the survey and ASFR. ................ 52 
Table 17: Total proportions ever attaining parity-Zulu ......................................................... 53 
Table 18: Total proportions ever attaining parity-Xhosa ...................................................... 54 
Table 19: Total proportions ever attaining parity-Pedi.. ....................................................... 55 
Table 20: Total proportions ever attaining parity-Sotho. ...................................................... 56 
Table 21: Total proportions ever attaining parity-Tswana ................................................... 57 
vii 
Table 22: Parity Progression Ratio of Zulus ...........................................................................5 8 
Table 23: Parity Progression Ratio of Xhosas ......................................................................... 59 
Table 24: Parity Progression Ratio of Pedis ............................................................................. 60 
Table 25: Parity Progression Ratio of Sothos .......................................................................... 61 
Table 26: Parity Progression Ratio of Tswanas ...................................................................... 62 
viii 
LIST OF FIGURES 
Figure 1: Total mean parities of selected ethnic groups ........................................................ 21 
Figure 2: Mean parities of different selected ethnic groups .................................................. 22 
Figure 3: ASFR of selected ethnic groups ............................................................................... 23 
Figure 4: ASFR of different ethnic groups ............................................................................. 24 
Figure 5: TFR of selected ethnic group. ................................................................................... 25 
Figure 6: Parity Progression Ratio of selected ethnic groups .............................................. 28 
Figure 7: Parity Progression Ratio to the third child .............................................................. 29 
ix 
ABBREVIATIONS 
ASFR Age-Specific Fertility Rate 
PPR Parity Progression Ratio 
STATS SA Statistics South Africa 
TFR Total Fertility Rate 
X 
CHAPTER 1: INTRODUCTION AND BACKGROUND OF THE STUDY 
1.1 Introduction 
Fertility constitutes part of the demographic processes which in turn influences the size and composition 
of the population. Sayi (2009:9) contends that establish ing the link between the start and pace offertility 
change and the corresponding traits in birth interval dynamics gives a clear understanding of the 
relationship between fertility and birth spacing. The differences in fertility are influenced by many 
factors such as postponement of birth or birth spacing, which for example shapes the current Sub-
Saharan fertility transition. The current fertility differs fundamentally from earlier transitions (Moultrie 
et al, 2013 :17). 
South Africa has had one remarkable demographic achievement, which is the decline in fertility in the 
past decades (StatsSA, 20 I 5: 4 I ; Palamuleni et al, 2007 : I 13). Moreover, this decline varies distinctly 
by population groups. Amongst these population groups, there has been a sluggish decline within the 
black population from 1970 to 1995 (Zuberi et al, 2005 : 50). The total fertility rate (TFR) of Blacks 
was the highest in I 970 by 5.4 as compared to Coloureds (5.1 ), Indians/ Asians ( 4.1) and Whites by 3. I . 
In I 974, TFR for Blacks declined to 5. I whereas other populations groups were below 4.2 . By 1995 the 
TFR for Blacks was 3.6 whereas for Coloureds it was (2.8), Indians/Asians (2.5) and Whites (2). From 
1996 to 2011 , the trend changed with the Blacks experiencing a decline in TFR, more significantly 
between 200 I and 2011. In 1996, Blacks had a TFR of 3 .49 whereas for Coloureds was 2.64, 
Indian/Asians (2.45), and Whites was 2.02 . In 2011 , Blacks had TFR of2.82 from 3.04 which was 
established for 200 1. For Coloureds, it was 2.41 , Indian/ Asian 1.98 and White 1.82. (StatsSA, 2015 :42). 
Given the fact that fertility has been higher among the Black population as compared to other racial 
groups, there is a need to understand the relative contribution of the different ethnic groups to the 
additional births in the population. Moreover, most demographers do not use the method used in this 
study being PPR but rather use the common ones as elaborated above. Demographers, academics and 
researchers need to know and familiarise themselves with this method so that it can be one of the main 
methods used in the discipline of demography. 
1.1.1 Parity Progression Ratio method 
1.1.2 What is Parity Progression Ratio 
The fertility process can be represented by cohort, but it is not limited to this, it can also be represented 
by the movement of women from one parity to the next (Preston, 200 I: I 04 ). This movement can be 
represented by Parity Progression Ratio (PPR). A woman ' s parity is the number of her live births, 
therefore PPR from parity i to parity i + I is the proportion of a cohort who had at least i live births who 
went on to have at least one more. PPR are sensitive indicator of the family building process because it 
reflects the sequential nature of fertility decisions (Kibet, 2015: 30). 
PPR is an interesting, important and robust measure of fertility (Spoorenberg:2015), which gives insight 
of the pace of the fertility transition from one child to the next. Spoorenberg and Dommaraju (2012:1) 
outline the three-main importance of PPR. Firstly, PPRs outline the liable nature of fertility behavior, 
meaning that the decision of subsequent birth is influenced by previous births. Secondly, PPRs can be 
used to measure family size specifically looking at the total number of children ever born which are 
common. Lastly, the decomposition of PPR can be used to estimate the relative contribution of the 
reduction of births at different orders to overall fertility decline. 
1.1.3 The evolution of Parity Progression Ratio 
Across the world, people make decisions about having children. Some decide on the number of children 
they want to have whereas others decide not to have children at all and others cannot have children due 
to physiological and biological reasons. Based on this , Hinde (2014: I 07) contends that it is desirable to 
study fertility by analyzing it in terms of the number of children a woman has already had. Therefore, 
PPR is one of the ways of analysing fertility by looking at the total number of children women or a 
couple had. 
According to Hinde (2014: 110), the logic of calculating PPR was first coined by Loius6 Henry in the 
1950s who was a French demographer. During the 1980s, PPR' s were independently rediscovered by 
Feeney, Yu and Ni Bhrolchain. PPR have been very useful in the field of demography; therefore, 
demographers explored the possibility of estimating this measure on a period basis. This means working 
out PPRs which apply to particular time periods. Adding to this, McDonald et al (2015: 1584) suggests 
that there are several approaches in which one can use PPR. These approaches include parity 
progression for birth or marriage cohorts, true parity cohorts , and synthetic parity progression. 
2 
1.1.4 How has PPR been used. 
Many demographers and scholars have explored using PPR in different ways. Yadava et al ( 1992) used 
PPR in a population from a truncated distribution of open and closed birth intervals. On the other hand, 
Mutakwa (2012:3) used a method that uses the current distribution of age- order-specific fertility rates 
to estimate PPR. This method estimates the future PPR which are common ly known as projected PPR. 
This method was used for estimating projected PPR of four African counties namely, Malawi, 
Zimbabwe, Cambodia, and Panama using data from two censuses. The results of this study showed a 
good estimation of the PPR. 
Contemporarily, PPR are used for various reason such as tracking fertility decline given the importance 
of this measure. Yadava and Sharma (2009:50) undertook a study which showed that unless PPR values 
at parity 2 to 4 are reduced, fertility decline wi ll not be achieved. Another study undertaken by Moraa 
(2007) showed that about 92% of women in Kenya who had one child progressed to the second while 
the proportion of progressing to the third child increased to 93%. The results further showed that the 
proportion of women who progressed to the fourth child decreased. 
In South Africa, Stats SA uses PPR over time to have a better understanding of factors influencing 
fertility (StatsSA, 2012:30). This is done specifically for the preference in the number of chi ldren 
desired in the country. The 2011 census fertility report outlines the pattern and trend of fertility over 
time in South Africa. The results are shown by 1996 census, 1998 OHS, 2007 community Survey and 
200 1 census. The report showed that cumulative PPR for parity 0-1 for all data sets except the 2011 
census was over 0.9 whereas for 2011 it was below 0.9. PPRs showed a great decline from parity 2-3 
in 2011 as the PPRs fell below 0.8 and in parity 6-7 the PPRs were below 0.6. This pattern shows that 
fertility is declining in South Africa. 
StatsSA also calculates incomplete PPR. These are called incomplete PPR because they are for women 
who are still of childbearing age and who may therefore sti ll have chi ldren (StatsSA, 2012:33). The 
results of this report showed a fluctuating pattern of incomplete PPRs. Age group 45-49 had a PPR of 
approximately 0.9 for parity 0-1 whereas for the age group 15-19 was between 0.1 and 0.2. Age group 
15-19 PPRs where only recoded up to parity 3-4 and this is the indication of incomplete ratios. 
3 
1.2 Problem statement 
Total fertility rate in South Africa has been declining among the different racial groups, however, 
different studies have indicated that fertility of the Black population has remained the highest among 
other racial groups since 1950s (Swartz, 2004 ). As a result, compared to other racial groups in South 
Africa, Black population has had a relatively bigger number of a youthful population and the extent to 
which this differs with other ethnic groups in the context of South Africa is not clear (StatsSA, 2018: 
10). Different studies have highlighted the effect of the youthful structure of a population. The youthful 
structure could result in population momentum which has a positive effect on future population growth; 
provision of services such as schooling; employment and other effects such as a demographic dividend. 
Thus, Wasserman (2016) states that " the higher the percentage of young people (especially those under 
age 15), the more the population will continue to rise as this large cohort of the population enters their 
reproductive years (15-49)". 
While fertility by different racial groups gives insight on the levels and patterns of fertility behaviour 
on these racial groups, it does not show the differences in fertility within each ethnic group. Because of 
the differences in language and other cultural aspects, the fertility behaviour is likely to differ 
significantly between the different groups within each racial groups, especial ly within the Black 
population where the differences are more pronounced. For example, a study by Zuberi et al. , (2005: 
65) shows that the mean age at first birth among the different ethnic groups in the Black population 
differs significantly. Another study quantified these differences by indicating that, in 1996, Xhosas had 
a mean age at first birth of 20.55 which was the highest, followed by Southern Sotho and Tswana 
(20.53). In 2011 , Zulus had the highest mean age at first birth by 23. I %, followed by Xhosas by 16.4 
%. (StatsSA, 2015:52). While the measure for ferti lity outlined above tends to simply indicate the age 
of transition to first birth, it is crucial that there is also a concurrent understanding of the percentages of 
women in different parities who go on to have an additional birth, and how that is influenced by the 
one's ethnicity. Additional births contribute to the changes in the population size and also the structure 
of the population. 
Besides, the other commonly used measurement of fertility , such as age-specific fertility rates and total 
fertility rates, are to some extent, not ab le to detect real changes in fertility in the short term, as they are 
affected by the timing of births (Sloggett,2015). PPR is one particular method that is able to overcome 
the above shortcoming in the measurement of fertility due to the fact that it can measure fertility of 
women even at the end of their childbearing ages (StatsSA,2015:7). 
4 
1.3 Rationale of the study 
There are many measures of fertility that are commonly used to examine patterns, levels, and trends of 
fertility such as age-specific fertility rate, which is expected to be consistent with what is known about 
the fertility behaviors of a country ' s women (StatsSA, 2015:39). Such behaviors include subsequent 
birth which means progressing to the next child, given that you already have a previous live birth. This 
study is influenced by lack of fertility data on ethnic groups and poor usage of PPR as a measure of 
fertility . Spoorenberg (2015) argues that despite the usefulness of common measures such as Age-
Specific Fertility Rate, these constructs are unable to detect real changes in fertility in the short term. 
The researcher argues that this happens as these measures are affected by the timing of births, also 
known as Tempo Effects. The usage of PPRs in this study foregrounds the advantage of PPRs as these 
are insensitive to the Tempo Effects. Furthermore, PPR provides a deeper and more significant 
understanding of the distribution of cohort fertility which is defined as the proportion of women in a 
cohort who end up with exactly no children, exactly one and exactly two children at the end of the 
childbearing years . This study will provide an insight into the differences of fertility transition within 
the black population, different ethnic groups and by so doing the policy makers will be able to identify 
the groups which are most at risk of having higher levels of fertility . Therefore, proper action can be 
taken with consideration to such ethnic groups . 
Again, it is an established fact that the black population is significantly more in the South African 
population. Statistics show that the South African population grew from 44.8 million in 200 I to 51. 77 
- million in 2011 whereby 76.4 % of this population were black (Mikva:2015). The pattern has remained stable and Blacks sti ll constitute the highest in 2018 statistics, showing that the total population of South 
Africa is 57 725 600 whereby 80.9% are black people. This pattern out lines population growth which 
is, amongst others, influenced by high fertility . Therefore, among this black population, the study 
identified which ethnic groups have high levels of fertility by looking at the main ethnic groups. Mikva 
(2015) states that Zulus are 28% of black population, Xhosas (20%), Pedis (11%), Tswanas (9.7%), 
Sothos (9 .5%) , Tsongas ( 5 .5%) , Swazis (3. I%), Vendas (2. 9%) , and Ndebeles (2.6% ). Therefore, this 
study focused on the first five ethnic groups since each accounts for the most significant population 
numbers more than others. 
5 
1.4 Main objective of the study 
The main objective was to compare and contrast fertility patterns and progression to the next child 
(subsequent birth) in different main ethnic groups within the black population in South Africa. 
1.4.1 Specific objectives 
• To examine the Age-Specific Fertility Rate and Total Fertility Rate of main ethnic groups 
within the black population in South Africa. 
• To examine the mean parities of the main ethnic groups within the black population in South 
Africa. 
• To examine how the PPR differs within se lected Black ethnic groups in SA among ever-
attaining parities of the main ethnic groups within the black population in South Africa 
• To estimate, in general , the proportion of women of particular parity who go on to have an 
additional birth 
• To determine how the different socio-economic factors , affect the transition of progressing to 
the third child 
6 
CHAPTER 2: THEORETICAL FRAMEWORK AND LITERATURE REVIEW 
2.1 Introduction 
This chapter consists of a review of the literature that was undertaken within a specific framework. 
Firstly, there is an urgent need to understand fertility from a broader perspective than has hitherto been 
the practice. Therefore, this study outlines the determinants offertility, meaning what the driving factors 
for fertility in South Africa are . Lastly, for a woman or couple to progress, there must be significant 
influential factors . Therefore, the study investigates such factors determining progression to the next 
child which is subsequent birth based on the previous studies in the literature on parity and fertility 
patterns. 
As fertility behavior of women differs, these factors affect them differently and shape their fertility 
differently during their childbearing ages. Therefore, women or couple that has begun their fertility with 
having one child, they are likely to progress to the next child and subsequently have more children. 
There are many factors again influencing subsequent birth . 
2.2. Factors associated with fertility decline 
2.2.1 Demographic factors 
2.2.1.1 Age 
Age as a demographic factor plays a significant role in fertility , this is because there are measures of 
fertility that are calculated based on age such as age at first birth. This refers to the age of a woman 
when she had her first birth. Childbearing ages are from 15 to 49. According to StatsSA (2012:21), 
median age gives a clearer understanding of whether the population is young, old or intermediate. 
Furthermore, a young population is when it has a median of less than 20 and when it has a median of 
30 and above is considered old . The 2011 census shows the median age of black Africans was 24 which 
increased over the years whereby in 1996 census was 21 and census 2001 was 22. 
2.2.1.2 Ethnicity-Language 
South Africa is one of the countries with a diverse population in the world classified according to four 
population groups namely; Black Africans, Whites, Coloureds, and Indians/Asians. These four 
population groups consist of different ethnic groups namely; Zulu, Xhosa, Tswana, Pedi (also known 
as Northern Sotho), Tswana, Sotho (also known as Southern Sotho), Tsonga, Swazi , Venda, Southern 
Ndebele, Coloureds, Afrikaner, English South Africans and Indians (Mikva:2015). Most of these ethnic 
groups constitute the black African population. On the other hand, StatsSA (2012:52) 2011 figures show 
7 
that Zulus begin their childbearing earlier than other ethnic groups and these are followed by the Xhosas. 
Their median age at first birth ranges from 20-21 . 
2.2.2 Socioeconomic factors 
2.2.2.1 Sex Preference 
The family building consists of many factors women or couple consider before or during their 
childbearing age. Desired family size and gender preferences are part of such factors. According to 
StatsSA (2015: v), there has been a decline from 1996 to 2001 in PPR and fertility preference of two 
children in South Africa. Even though this is the case, the preference of two children has been significant 
in South African fertility and this began in 2007 and was most evident in 2011. This preference has 
been experienced across women of all population groups, women with the same educational level, and 
marital status. 
Previous studies that were based on parental gender preferences, found that parents with sex preference 
either divorce of progress to the next child (Saarela and Finnas, 2014:49). The results of the study 
conducted by these two authors show that parents prefer boys more than girls and that parity progression 
is high amongst couples who only have girls than boys. These boy preferences are associated with 
economic productivity and social organizational system of patrilineal families. 
2.2.2.2 Marital Status 
StatsSA (2015 :69) defines marriage as "the act, ceremony or process by which the legal relationship of 
husband and wife is constituted". Therefore, Majumder and Ram (2015:3) see marriage as the legal and 
cultural institution that seals approval or permits childbearing. Thus, marriage is considered an 
important determinant of fertility. Again, marriage is important as it increases the risk of conception for 
those women who are married (Rampagane, 2016:38). Jn South Africa, marriages are categorized as 
civil , religious and traditional or customary. 
In most cases when demographers look at marriage in the context of fertility they look at age at first 
marriage. According to Rampagane (2016:2) studies on age at first marriage are important due to the 
close link between marriage and the beginning of childbearing and fertility . Thus, Kumchulesi (2011) 
states that "marriage signifies the beginning of the reproductive cycle of women, and various studies 
point to the high correlation between age at first marriage and age at first birth" . In South Africa it has 
been established that age at marriage is increasing, the marriage rate is decreasing, and the prevalence 
of cohabitants and marriage dissolution is increasing (Rampagane, 2016 :2). 
8 
Based on the above, Amoateng (2004) argues that the decline of fertility in South Africa is explained 
from patterns and changes occurring in the marital unions. One can, therefore, associate fertility declines 
with decreasing marriage patterns. Another factor that has contributed to the fertility decline in South 
Africa is non-marital fertility. 
Fertility is directly related to marriage in most population groups and this comes from the cultural 
perspectives that suggest marriage initiates child-bearing. The perception is that married women, 
normally have more children than unmarried women. Traditionally and culturally births to unmarried 
women were not accepted in most societies and this led to women beginning their childbearing after 
marriage and continued throughout their reproductive lifetime as long as they were married. Women 
who get married earlier turn to have on average a longer period of being exposed to the risk of falling 
pregnant and having a greater number of lifetime births. This implies that married women have children 
more often regardless of their age at first birth. Pressat (1993) contends that knowledge of PPR allows 
the description of the marital fertility of a group of women. 
StatSA (2012:38) outlined the PPR by marital status in the 2011 census for women aged 45-49 . These 
were women who reached the end of their childbearing ages. The results showed that childlessness is 
higher among women aged 45-49 who have never been married as compared to those women who are 
married. This supports the statement made above that, marriage influence fertility. The PPRs were 
higher amongst those women with 0-2 children and started to decline from parity 2-3 . 
2.2.2.3 Religion 
Religion is also one of the social factors that are part of human socialization and to some extent affect 
fertility patterns. Religion plays a vital role in demography, including fertility, as it forms part of the 
various social institutions, cultural norms, economic and environmental conditions which comprise the 
indirect determinants of fertility. Rampagane (2016 :84) states that religion has been "an important 
factor in influencing the prevention of non-marital childbearing." In South Africa, the pervasive religion 
is Christianity whose outlook is that childbearing can only commence as an aftermath of marriage. The 
other ones include traditionalism and Islam. 
2.2.2.4 Education 
Education is one of the human rights which every human being must attain. It is also considered one of 
the best tools for human empowerment and development thus it is emphasized and prioritized in the 
Millennium Development Goals and Sustainable Development Goals (StatsSA, 2012:30). In 2011 , 35% 
of the black population had some secondary school as their highest level of education attained. On the 
other hand, 10.5% had no schooling, 13.9% had some primary schooling, 4.9% completed primary 
school , 26.9% had Grade 12. 
9 
A number of studies have investigated the relationship between fertility and education. But still , the 
relation between education and PPR remains limited (Wood et al, 2014: 1395). Results of a study done 
by Wood et al shows that women who are highly educated are less likely to enter motherhood as 
compared to those who have low and basic education. Despite this, the results further show that 
progression to the third child has declined among low and middle educated women whereas highly 
educated women have shown more contentious variations. 
2.2.2.5 Place of residence 
In South Africa, there are three main types of a residential dwelling, (referred to as the geographical 
residence) namely, urban, farm and traditional areas. These types ofresidence play a significant role in 
the daily lives of South Africans. In terms of fertility (specifically PPR), the 201 I census statistics show 
that the proportion of women living in urban areas progressing to the third child is far lower than of 
those women in non-urban areas (StatsSA, 2012:36). Even though this is the case, there was a gradual 
decline within the women who are living in farms and traditional areas who are opting for higher birth 
orders. Furthermore, the median age of first birth is higher (65.9%) for those women residing in urban 
areas as compared to those living in non-urban areas (35 .2%). 
2.2.2.6 Province of residence 
South Africa is made up ofnine provinces which are composed of different populations as well as ethnic 
groups. According to Mikva (2015), the majority of Zulu people live in KwaZulu Natal , whereas some 
of them live in Mpumalanga and Gauteng. Xhosas mostly live in Eastern Cape province, Pedis in 
Limpopo, Tswanas in the North West and Northern Cape and Sothos in the Free State. The 2011 census 
report suggests that women from Limpopo tend to have a higher number of children ever born in their 
lifespan as compared to women from the other eight provinces (StatsSA, 2012:58). The 20 I 1 census 
shows that in South Africa there are provincial differences on fertility. For example, Limpopo has the 
lowest proportion of childless women (8%). Even though this is the case, the province has higher PPRs 
across parity 1 through 5. On the other hand, the proportion of women progressing from parity 5 to 6 is 
higher in Mpumalanga (60%), KwaZulu Natal (62%) and Eastern Cape (63%) . 
One can, therefore, conclude that fertility is influenced by all the above-mentioned determinants . These 
are normally the common and broad mechanisms that define fertility patterns. Fertility is one of the 
main components in demography that contributes to an understanding of probabilities in population 
growth, aging and the structure of the population in various ways. This study unpacked fertility by 
looking at PPR within the major ethnic groups of the black population in South Africa. Therefore, to 
narrow the study, the factors that determine or influence women or a couple to progress to the next child 
are examined in significant detail. 
10 
2.3 The evolution of fertility decline in South Africa 
South Africa displays demographic management that characterizes both developed and developing 
countries and this management is directly linked with socio-economic indicators along racial lines. This 
is because standards of living are patently associated with race in South Africa. Furthermore, South 
Africa ' s fertility transition experience is one of the most advanced in Sub-Saharan Africa (Swartz, 
2004:539). The fertility decline in South Africa is explicated as a demographic achievement because 
South Africa has one of the lowest fertility levels on the continent (Palamuleni et al, 2007: 113). This 
fertility decline varies by population groups. Among all the population groups in South Africa, a decline 
in fertility has been observed from as early as the I 960s. Prior to that, fertility for South Africa as a 
whole was high and stable between 1950 and 1970 with an estimated average of 6 to 7 children per 
women. 
These variations were influenced by the apartheid era. Palamuleni et al (2007 : 113) contend that the 
observed fertility decline has not been the same in the South African nation, which before 1994 was 
dominated by a smaller political administration that was committed to racial ization, classification, and 
ethnification of society. Therefore, these differences in fertility are a reflection of socio-economic and 
general imbalances of the past in terms of standard of living and were influenced by apartheid policies. 
Palamuleni et al (2007) further contend that the corre lation between levels, trends, and differentials in 
South African fertility and apartheid ideology and policies is complex and therefore cannot be easily 
concluded. Although this is the case, one clear thing is that demographic concerns were central to many 
apartheid policies. 
Camlin et al (2004) argue that during the apartheid era, official population policies were intended to 
reduce African fertility and this was influenced by White South Africa' s fear of being swamped by the 
black population. Other policies ensured that African people were systematically denied access to 
education, health care, and urban residence. These factors were important in determining the pace of 
fertility decline in a variety of settings for both developed and developing countries. 
Adding to these characteristics, delays in childbearing contributed to the decline in fertility as African 
women in white areas first gave birth later in life and had longer intervals between births. Again, it is 
essential to understand that the proportion of women who want to have another child soon decreases 
sharply with the number of living children (StatsSA, 2017: 13). Swartz (2004:540) submits that the 
results of the provision of family planning were impressive as by 1983 over half the eligible women in 
South Africa were practicing contraception. 
In South Africa, the choice of contraception method follows racial stratification whereby whites, who 
make the least use of public family planning services, choose from a wider range of contraceptive 
methods. On the other hand, Africans and Coloureds, who comprise the bulk of contraceptive client of 
11 
organized public family planning services, tend to predominantly use injection. This goes back to the 
issue of information sharing and widening of reproductive choices as well as the issue of women ' s 
control over their own bodies and sexuality. The use of contraceptives thus has an influence on the 
country's levels of fertility. 
2.3.1 Trends of fertility levels within South African population groups 
The great decline in South African fertility is explained by the trends of Total Fertility Rates (TFR). As 
outlined earlier, between 1950 and 1970 South Africa had high and stable fertility. From the end of the 
19th century, Whites experienced a long-sustained fertility decline until achieving below-replacement 
fertility by 1989 with a TFR of 1.9. In tandem, Asian fertility also declined gradually from a TFR of 
about 6 in the 1950s to 2.7 in the late 1980s. As for Coloureds, fertility declined amazingly from 6.5 in 
the late I 960s to approximately 3 by the late 1980s. Lastly, African fertility was estimated to have 
decreased from a high TFR of 6.8 to a low TFR of about 3.9 between the mid- l 950s and the early 1990s. 
Although African fertility continues to decline it is still substantially higher than other population 
groups (Swartz, 2004: 540). The TFR of South Africa Blacks stood at 2.9 during the 1990s (SAD HS : 
1999). 
Stats SA (2015 :41) explains the ferti Ii ty trends from 1985 to 2011 as documented by various authors . 
The trend shows that South Africa TFR was between 4.5 and 2.5 from 1985 to 20 I I. Moreover, South 
African TFR declined from 2.84 children per woman in 200 I to almost 2.67 children per woman in 
2011. Even though this is the case, the level of decline in TFR from 1996 to 2001 is striking. 
The census indicates a decline from 3.23 in 1996 to 2.67 in 20 I I. Looking at the 2011 census, the 
variations in fertility amongst population groups are as follows: black African women have a TFR of 
2.82, Coloured women is 2.57 and has remained the highest (it increased from 2.41 in 200 I to 2.57 in 
2001 ), and Whites and Indian/A sians women have a below replacement with a rate of I. 70 and 1.85 
respectively. The accelerated decline in TFR from 2001 to 2011 seems the highest amongst black 
African women (StatsSA, 2015:41 ) . 
These variations also overlap to total fertility rates by provinces and therefore support variations of 
population groups. TFRs for 2011 suggest that fertility is high in Limpopo (3.25), Mpumalanga (2.90), 
Eastern Cape (2.87) and KwaZulu Natal (2.73) and the lowest has been recorded in Western Cape 
(2.28) and Gauteng (2.27). Northern Cape showed an increase in TFR from 2.43 in 2010 to 2. 75 in 2011 
and this may be influenced by an increase in the TFR of Coloured women. North West also showed a 
slight increase from 2.77 in 2001 to 2.83 in 2011 . One can, therefore, argue that the provincial TFR is 
also influenced by the population group living there, particularly women. 
12 
2.4 Theoretical Framework 
Demographic factors 
• Age 
• Ethnicity-Language 
Socio-economic factors 
• Sex preference SUBSEQUENT BIRTH 
• Marital Status 
• Religion 
• Education 
• Place of residence 
• Province of residence 
There are many factors that influence a couple and women to begin their fertility and among those are 
the well-known direct and indirect determinants of fertility such as contraceptive use. The Bongaarts 
came up with a well -known and commonly used theory that underpins the determinants of fertility . 
According to Bongaarts and Potter ( 1983), when one makes a detailed and comprehensive analysis of 
factors influencing fertility , they must make a distinction between two classes of determinants being: I. 
Proximate variables and 2. Socioeconomic and environmental "background" variables. Fertili ty 
revolves, is influenced and dependent on these factors . Proximate determinants cons ist of all biological 
and behavioural factors and socioeconomic consist of social, cultural , economic, institutional , 
psychological , health and environmental factors. The main relationship between these factors is that 
socioeconomic factors operate through proximate factors to affect fertility (Bongaarts et al, 1984:515). 
From the theory of Bongaarts, socioeconomic factors namely; age and language (referred to as 
demographic factors in this study), sex preference, marital status, religion, education, place ofres idence 
and province of residence came out in thi s study as factors associated with fertility decline and factors 
influencing progression to the next child (subsequent birth) . 
13 
CHAPTER3:METHODOLOGY 
3.1 Introduction 
This chapter details the methodology used for this study, including the sources of secondary data used, 
a sample of the study, methods of analysis, software used and the variables that were deemed critical 
use for this study. The study examined two factors which are patterns and determinants of fertility 
amongst different ethnic groups within the black population in South Africa. Therefore, the PPR were 
used to analyze the patterns and the odds ratio were used for determining progression to the third and 
more children. 
3.2 Data sources 
This study used the data from the 2016 Community Survey in South Africa. This data was collected by 
Statistics South Africa and it was national representative and was the most recent data that contained 
the most relevant variables for the study especially fertility levels among the different ethnic groups in 
the black population. According to StatsSA (2016:8), the 2016 Community Survey was the second such 
survey conducted by Statistics South Africa and it bridges the data gap between census 2011 and the 
upcoming census 2021. 
3.3 Sampling Method 
The sample design for 20 I 6 Community Survey was stratified single stage sample design. The geo-
reference dwelling frame was used as a sampling frame for Community Survey 2016 and only the points 
classified as Dwelling Units (DU) were considered for sampling as these included households that are 
part of the target population. The number of DUs at a point is used for the selection of DUs within an 
EA and the geo-reference point is used to locate the sampled DUs within an EA. Community Survey 
2016 was based on a single-stage sample design whereby all eligible Census 2011 EAs were included 
in the initial frame and a selection ofDUs within the eligible EAs was taken based on the sample design . 
Furthermore, the Community Survey 2016 DU sample was drawn using the systematic sampling 
technique (SYS) to identify the household from which individuals were numerated. 
14 
3.4 Study population 
The total population enumerated in the survey was 1 763 060 for all ages. For the purpose of the study 
a total of 726 855 women in age group 15-54 was selected from the, Zulus, Xhosas, Sothos, Ped is and 
Tswanas, ethnic groups. 
3.5 Methods of Analysis 
The study employed different measures of fertility which are chronicled as steps of calculating PPR: 
• Mean parities- Children ever born 
Number of women 
• Age-Specific Fertility Rate- Bx * 1000 
Pfx 
• Total Fertility rate = Sum of ASFR * 5 
• The proportion of women ever-attaining parities- swx(i) = L}-i SNxU) 
• P an.t y p rogress1. on R at1. 0 (PPR) - sa x (i' ) = SMx(ci+·)l ) 
SMx t 
• The odds ratios 
• Observed PPR 
Moultrie et al (2013) argue that the examination of fertility data by parity using PPR and their 
equivalents gives additional information on childbearing patterns and can be used to assess changes 
in the parity distribution of fertility. Hinde (2014: 107) argues that it seems desirable to analyze 
fertility in a way which takes account of the number of children a woman has already had . 
Therefore, PPR fills the gap in probability trends and estimating the population patterns. 
Data required for calculating PPR are Parity by age group of women aged 45-49 or more and the 
number of children born during the year preceding the census classified by mother's age and 
children ever born. 
15 
To obtain the observed PPR, there are several steps and formulas that were followed which are as 
follows: 
The application of this method starts with the extraction of data for children ever born by parity, i, and 
age group. From this, mean parities were calculated using the following formula 
Mean parity= Children ever born 
Number of women 
Thereafter, the number of births by parity in the 12 months prior to the census date is also extracted. 
These tabulations are then used to calculate the proportions of women who have ever attained each 
specific parity, i. To achieve this, the number of women in each five-year age group who have given 
birth to I or more children is initially calculated. Then ASFR and TFR were calculated 
ASFR= Bx * 1000 
Pfx 
TFR= Sum of ASFR * 5 
The proportion of women who have attained parity i or more, 5Mx, is then given by dividing the total 
number of women with i 5x children or greater by the total number of women in that age group. 
From this, we can now derive the proportions of women who have progressed from parity i to i+ I , 
which are the PPR. 
These are obtained by dividing the proportion of women with i+ 1 children by that with i children 
(Moultrie et al, 2013 :70) 
3.5.1 Correction to parity (Children ever born) data 
According to the United Nations (1983:230) during the collection and processing of data on children 
ever born, there were errors such as misclassification of women with zero parities as women whose 
parities are not stated. This error is normally made by the interviewer during the time of the survey. The 
2016 Community Survey data on children ever born have the following responses: 1-14 children, never 
had children born alive which is women with zero parities, not applicable, women, who did not specify 
on ever given birth, and women who did not know if they had children born alive. According to United 
Nations (1983:230), there might be a mistake of having a greater proportion of childless women under 
the category of parity not stated, therefore excluding these women in the denominator when calculating 
mean parities would result in overestimated parities. On the other hand, excluding women who did not 
state their parity would make the denominator too large and result in an underestimation of true average 
parities. To correct these errors, the EI-Badry correction method was used . This method outlines the 
16 
procedure to estimate the proportion of women belonging to the category "parity not stated" who should 
logically have been classified as childless. 
3.5.2 Binary Regression (The odds ratios) 
Binary regression method was employed in this study to model the probability of a women progressing 
to the third child given that's has already had 2 children. The motivation of selecting the third birth was 
based on fertility replacement level which is 2.1. The dummy variable was coded I if a women 
progressed from second to the third child and O if she did not. 
In this case, the independent variables used in the model were ethnicity, education, marital status, 
religion, provinces and place ofresidence. The analysis was carried out in two different methods namely 
Method 1: Unadjusted odds ratios and Method 2: Adjusted odds ratios with the aim of outlining the 
probability differences in the selected variables. 
3.6 Variables 
3.6.1 Dependent variable 
Children Ever Born- This will be used as it outlines the incremental number of children that a woman 
of particular parity goes on to have. 
3.6.2 Independent Variables 
5-year age groups- This is the age of the respondents grouped by 5-year age group. The study uses 
age groups from 15 to 45 as they are childbearing ages and 50 to 85 plus for completed fertility in order 
to measure parity progression from x-n parity. 
Language most spoken in the household- This variable is used as a proxy for ethnicity. The 
respondents were asked to state the language they mostly speak within their household. For this study, 
the main spoken languages within the black African population namely; lsiZulu, lsiXhosa, Sepedi, 
Sesotho, and Setswana will be used. These are referred to as the main ethnic groups as outlined in the 
rational of the study. These groups comprise the greater population in the black population. 
Births in the last 12 months- Respondents were asked whether they have had any birth in the last 12 
months and the expected answers were yes and no. 
17 
3. 7 Limitations of the study 
3. 7.1 Language Variable 
In the 2011 census nor 20 16 Community Survey, there is no specific question asked respondents on 
ethnicity. Therefore, for this study language was used as a proxy for ethnicity as supported by StatsSA 
(20 15:53). The report states that " language is a proxy indicator of the different ethnic groups that reside 
in South Africa". Furthermore, one must take note of the limitation present in this variable. The first 
language spoken in the household may not necessarily reflect their culture or ethnicity. The question 
was asked as the most spoken language in the household, therefore this also covers the marriage context. 
18 
CHAPTER 4: ANALYSIS AND PRESENTATION OF RESULTS 
4.1 Introduction 
The chapter presents the results showing fertility differentials among the Zulu, Xhosa, Pedi , Sotho, and 
Tswana ethnic groups in South Africa. Firstly, the EI-Badry method was used to correct data on children 
ever born. Secondly, study calculated the total mean parities for the total population followed by mean 
parities for different ethnic groups . Thirdly, age-specific fertility rates for different ethnic group were 
calculated from which total fertility rate. The study went further to calculate the proportion of women i 
number of parity and went on to have parity i+ I and calculate. 
4.2 EI-Baldry correction for data on children ever born 
The El-Baldry method was proposed in 1961 to rectify the errors that may appear on the data of children 
ever born as discussed in chapter three. The method was based on the high observed correlation that 
exists between the proportion childless and proportion of parity not stated. The method may be 
calculated using the following formulas : 
Data requirement 
a) The number of Zulu, Xhosa, Pedi , Sotho and Tswana women in the zero-parity category 
meaning those who are childless classifies by five-year age group. 
b) The number of Zulu, Xhosa, Pedi , Sotho and Tswana women in the category parity no stated 
classified by five-year age group. 
c) The total number of Zulu, Xhosa, Pedi , Sotho and Tswana women classified five-year age 
group. 
Step 1:  Calculation of proportion with parity not stated 
NS(i)=FNS(i)/FP(i) 
Step 2: Calculation ofp roportion childless 
Z(i)= FZ(i)/FP(i) 
Table I shows the total number of women, women who were childless and those who did not state their 
parity of the selected ethnic groups . The results show that within all the ethnic groups that were selected 
88% of women aged 15-19 were childless, followed by age group 20-24. The proportions of childless 
women decrease as age increases. Again, across all ages , the proportion of women who did not state 
their parity was below 400. However, the proportions of women who did not state their parity were all 
below 2%. According to Moultrie et al (2013 :36), if the proportion of parity not stated is less than 2%, 
19 
the correction of data on chi ldren ever born will not be necessary needed but if it was more the data 
needs to be corrected with further steps of the correction method . 
Table 1: Total Proportions of childless and parity not stated 
Age group Total FP(i) Parity not stated Chi ldless z Parity not stated 
Childless FZ (i) FNS (i) (i) NS(i) 
15-19 115240 100934 174 
0.140 0,002 
20-24 114921 59365 230 
0.932 0,004 
25-29 111395 34617 323 
0,453 0,004 
30-34 98900 22194 280 
0,290 0,004 
35-39 82018 15850 249 
0,240 0,004 
40-44 74098 12425 226 
0,202 0,004 
45-49 66700 10798 30 1 
0,194 0,005 
4.3 Pa rity Prog ressio n Rat io 
As stated in chapter three, calculating PPR consists of several steps. From these steps, there are other 
measures of fertility namely mean parities and age-specific fertility rates. Therefore, before reaching 
the final and main method of analysis of the study which is the PPR mean parities, ages specific and 
women ever-attaining parity will be calculated . 
4.3.1 Mean parity for all women 
Figure 1 below presents the mean parities for all ethnic groups combined. The resu lts show that the 
average parity increased with the age group. Consequently, the mean parities increased gradually and 
reached a peak at age group 45-49 at a parity of2.78. However, the average number of children dec lines 
slightly thereafter to 2,73 in the age group 50-54 . This simply means women who reached the end of 
childbearing ages have lower parities than those who are at the exact end of childbearing age (49). The 
lower reported fertility among the older age cohort is not consistent with fertility patterns in South 
Africa where fertility has been decreasing in the recent period . This can simply mean that women in 
this age group have not reported some of the children they had in the past. 
20 
Figure I: Total mean parities of selected ethnic groups 
Total mean parities of selected ethin ic groups 
3,000 
2,739 
2,500 
.,, 2,000 
QJ 
·.::; 
·;:: 
~ 1,500 
C 
ro 
QJ 
2 
1,000 
0,500 
0,000 
15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
Age groups 
The results in Figure 2 are slightly different from figure I as these ones show the mean parity of different 
age cohorts disaggregated by ethnic groups. The pattern of reported mean parity shows that the mean 
parity is gradually increasing with age in all ethnic groups until it reaches age group 45-49. But there 
are differences in thi s case. For age groups 15-19, 20-24, and 25-29 women of all se lected ethnic groups 
had a simi lar mean parity. Fertility started to differ from the age group 30-34 up to 50-54 . For example, 
Tswana women had higher mean parity at age group 30-34 compared to other ethnic groups and 
however it started to decline to the point where it is the second lowest at the end of age group 50-54. 
The results further showed that, from age 35 the highest mean parity is observed among Pedi women 
followed by Xhosas. On the other hand, the Sotho women had the lowest mean parity. As noted in 
figure I, the same pattern is observed where women in age group 50-54 had lower reported fertility than 
younger age cohorts. This implies that misreporting of children by older age cohort was common in all 
the ethnic groups 
21 
Figure 2: Mean parities by different ethnic groups 
Mean pari ties of different selected eth nic groups 
3,500 
3,000 
2,500 
V) 
QJ 
:§ 2,000 
"Q' . 
~ 1,500 
QJ 
~ 
1,000 
0,500 
0,000 
15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
Age geoups 
- Zu lu - Xhosa ..._Pedi - Sotho - Tswana 
4.3.2 Age-Specific Fertility Rate of all ethnic groups 
The figure shows that the Age Specific Fertility Rate (ASFR) of all women selected in the study had a 
concave pattern whereby women in the age group 15-19 had the least number of live births and peaks 
at the age group 20-24 and decreases thereafter. This is normally expected as younger women do not 
have more children but mostly start their childbearing at age 20 and above. Women aged 50 and above 
are normally expected to have completed their ferti li ty, thus the least rates of live births. 
22 
Figure 3: ASFR of selected ethnic groups 
ASFR of all women in study 
0,120 
0,106 
0,100 0,100 
I 0,088 0,080 er: 
~ 0,060 0,064 
<! 0,051 
0,040 
0,029 
0,020 
0,000 0,005 0,000 
15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
Age groups 
Figure 4 presents the ASFRs disaggregated by different ethnic groups. The Figure shows that the 
Xhosas and Tswanas had an early transition to childbearing in the year the survey was conducted 
compared to other ethnic groups. For example, while those under age of20 ( 15-19) among the Pedi and 
Sotho had a rate of0.040 and 0.044 respectively, the Xhosas and Tswanas had the highest rates of0.056 
and 0.049 respectively. With the exception of Sothos, the fertility rates in the year reached a peak in 
age group 20-24 where the Tswanas had the highest rate. While the rates for all age groups declines 
thereafter, the rates for Pedis remains the highest in subsequent age groups. 
23 
Figure 4: ASFR of different ethnic groups 
ASFR of different ethnic groups 
0,140 
0,120 
0,100 
er: 0,080 
LL 
V, 
<l'. 0,060 
0,040 
0,020 
0,000 
15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
Age grou 
- Zu lu - Xhosa - Pedi - Sotho - Tswana 
The ASFR above was used to calculate the expected number of children that a woman wi ll have at the 
end of the reproductive life by different ethnic groups. Figure 5 shows that fertility rate and pattern of 
Pedi and Tswana are farmi liar as they TFR are both 2.4 and are the highest among all the selected ethnic 
groups. Zulus had the lowest TFR of2.0 which is below th replacement. 
24 
Figure 5: Total Fertility Rates of selected ethnic groups 
TFR of diffrent ethnic groups 
2,500 
2,413 2,424 
2,400 2,347 
2,300 
2,200 
a:: 
u.. 2,125 
I-
2,100 
2,040 
2,000 
1,900 
1,800 
Ethnic Groups 
■ Zulu ■ Xhosa ■ Pedi ■ Sotho ■ Tswana 
25 
4.3.3 Proportions ever atta ining parity i 
Table 4 below shows women of selected ethnic groups who progress to having the next child. The table 
shows that 78% of women aged 50-54 had at least I child as they are assumed to have completed their 
fertility . On the other hand, those women who are aged 15-19 which are women starting their 
childbearing 12% of them had at least one child and only 1% of them had the second child . Women in 
the early ages of childbearing 15-29 did not have more children thereafter but those who were aged 30-
49 had the eighth child . In the age group, 30-34 86% of women had four children while 2.65 had the 
fifth child. 
Table 4: Total proportions ever attaining parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 1 1 I 1 1 1 1 1 
1 0,123 0,481 0,686 0,773 0,804 0,829 0,834 0,777 
2 0,010 0, 127 0,348 0,538 0,635 0,696 0,724 0,685 
3 0,001 0,021 0,110 0,246 0,366 0,462 0,526 0,515 
4 0,000 0,003 0,026 0,086 0,164 0,251 0,322 0,337 
5 0,000 0,000 0,005 0,026 0,065 0,125 0,178 0,200 
6 0,000 0,000 0,001 0,008 0,025 0,059 0,094 0,110 
7 0,000 0,000 0,000 0,002 0,010 0,027 0,047 0,056 
8 0,000 0,000 0,000 0,001 0,004 0,012 0,024 0,031 
9 0,000 0,000 0,000 0,000 0,001 0,005 0,012 0,016 
10 0,000 0,000 0,000 0,000 0,000 0,002 0,005 0,008 
11 0,000 0,000 0,000 0,000 0,000 0,001 0,002 0,002 
12 0,000 0,000 0,000 0,000 0,000 0,000 0,001 0,001 
13 0,000 0,000 0,000 0,000 0,000 0,000 0,001 0,001 
14 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
15 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
16 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
17 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
26 
4.3.4 Parity Progression Ratio 
The table below displays the overall PPR of selected ethnic groups (Zulu, Xhosa, Pedi , Sotho, and 
Tswana). Women aged 15-19 progress up to the fourth child and only 15% did progress. The table 
shows that as the ages increase for younger age groups, the parity decrease. In contrast, when the age 
increase for older age groups the parity increases. The table shows only 12% of women in the age group 
15-19 progressed to have the first child and this was the least across all age groups. The highest 
proportion were women aged 45-49 whereby 83% of them progressed to their first child. In the older 
age groups, greater proportions of women progressed to the second child. Age group 40-44 (83%), 45-
49 (86%) and 50-54 (88%). For younger age groups, the proportions were lower whereby the age group 
15-19 it was only 8%, 20-24 (26%) and 25-29 (50%). The pattern shows that as parity increases over 
parity 4 the progression proportions decreases. 
Table 5: Total Parity Progression Ratio of selected ethnic groups 
Parity 
15-19 20-24 25-29 
(i) 30-34 35-39 40-44 45-49 50-54 
0 0,123 0,481 0,686 0,773 0,804 0,829 0,834 0,777 
1 0,084 0,264 0,507 0,696 0,789 0,839 0,868 0,882 
2 0,09 0,164 0,317 0,457 0,577 0,664 0,727 0,751 
3 0,15 0,134 0,235 0,349 0,448 0,543 0,613 0,654 
4 0 0, 122 0,209 0,307 0,399 0,499 0,552 0,594 
5 0,23 1 0,214 0,299 0,390 0,470 0,526 0,551 
6 0 0,202 0,269 0,381 0,454 0,507 0,511 
7 0,23 1 0,301 0,376 0,456 0,508 0,549 
8 0 0,333 0,363 0,441 0,49 0,53 
9 0,286 0,376 0,451 0,455 0,459 
10 0 0,268 0,356 0,45 0,32 
11 0,545 0,359 0,491 0,5 
12 0 0,348 0,468 0,438 
13 0,25 0,541 0,429 
14 0 0,3 0,667 
15 0,333 1 
16 0,5 0,5 
17 0 0 
The figure shows that from approximately 80% to approximately 1%  of women progress from childless 
to the eighth child. The pattern of this progression shows a decline from the age group of women who 
completed their fertility 50-54 to those who are starting their fertility 15-19. Those younger ages are 
not likely to progress tho the ninth child and above. The graph shows that is it women in the older age 
27 
groups who progress to those parities. The pattern of that progression fluctuates from 0% to less than 
60%. Below is figure 5 of the plotted graph of tab le 25 above. The graph displays the PPR of the se lected 
ethnic groups within the black population. 
Figure 6: Parity Progression Ratio of selected ethnic groups 
Parity Progression Ratio of selected ethnic groups 
1,000 
0,900 
0,800 
0,700 
0,600 
lf 0,500 
Q. 
0,400 
0,300 
0,200 
0,100 
0,000 
50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 
Age group 
- o - 1 - 2 - 3 - 4 - 5 - 6 - 7 -8 - 9 - 10 
Figure 7 below shows the PPR to the third child by women in different ethnic groups. The fi gure shows 
that Sotho women in age group 15-19 are less likely to progress to the third ch ild. However, the 
progression ratios increase by age for all the ethnic groups. Ped is are more likely to progress to the third 
child as compared to the other ethnic groups. For example, 63% of Pedi women progressed to the third 
child while for Tswanas it was only 55% who progressed. 
28 
Figure 7: Parity Progression Ratio to the third child 
Parity progression to the third child 
0,9 
0,8 
0,7 
0,6 
er: 0,5 
c.. 
c.. 0,4 
0,3 
0,2 
0,1 
0 
15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
Age group 
4.4 Odds Ratio-Factors associated with the transition to the third parity 
This section presents the results of the logistic regression method showing the likelihood that a woman 
who had a parity of two, wi ll proceed to third and higher-level parity. The results are presented both an 
unadjusted odds ratio whi le model 2 present the resu lts adjusted for by other socio-economic factors 
Table 5 below shows the odds ratios of those women who have the second child and progressed to three 
and more chi ldren . Model 1 in the table shows that women with no schooling and those who had 
primary school were 3 times more likely to progress to the third and more than those with tertiary 
education. Those with the secondary school were 1.3 times more likely to progress to the third child 
and more. These resu lts were statistically s ignificant. However, the same results persisted when the 
variables were contro ll ed with other socio-economic variab les in model 2. 
The table further shows that within the selected ethnic groups, women who are widowed are 3.2 times 
more likely to progress to the third child and more fo llowed by those who are legally married than those 
who are single and never li ved with a partner. Furthermore, women who were living together, divorced, 
and single but been living with a partner were 1.7 times more li kely to progress to the third chi ld and 
more. All these results were statistically significant. When this variable was controlled with other 
variables in model 2 there were changed. 
29 
Looking at religion, the table shows that women who are Christians were 12 times less likely to progress 
to the third child than those who were Traditional African, and this was statistically significant. 
However, this changed when it was controlled by other variables. Model 2 shows that Christian were 
1.0 times more likely to progress as compared to the traditional African religion. This was not 
statistically significant. Islam women were 1.0 times more likely to progress to the third child and more 
and this was not statistically significant. Furthermore, only Sotho women were 12 times less likely to 
progress to the third child and more than Tswana's. All other ethnic groups were I time more likely to 
progress. These were statistically significant. However, in model 2 Sotho women were 9 times less 
likely to progress and this was not significant. 
Women who lived in urban areas were 38 less likely to progress to the third child and more as compared 
to those who resided at farm areas . On the other hand, who resided traditional areas were 4 times less 
likely to progress to the third child and these was not statistically significant. In model 2, women who 
lived in traditional areas were 1.0 times more likely to progress to the third and more children as 
compared to those who resided in farm areas. 
When looking at the provinces, women of all provinces were less likely to progress to the third and 
more children as compared to women from Limpopo and these were statistically significant. Women 
who stayed at KZN were 20 times less likely, from 23 times less likely and those from Eastern Cape 
were 8 times less likely. Lastly, women who progress to the second child and preferred boys were not 
likely and this was not statistically significant. 
30 
Table 6: The odds ratios by different socio-economic variables in the study. 
Model 1: Unadjusted Model 2: Adjusted odd 
odd ratios ratios 
Variable Significance Exp (b) Significance Exp (b) 
Education 
Tertiary Education (R) 1.00 
No schooling 0.000 3.396 0.000 3.284 
Primary school 0.000 3.768 0.000 3.793 
Secondary school 0.000 1.307 0.000 1.493 
Marital status 
Single and never lived with a partner(R) 1.00 1.00 
Legally married 0.000 2.545 0.000 2.658 
Living together 0.000 1.759 0.000 1.679 
Divorced 0.000 1.883 0.000 2.280 
Separated 0.000 2.347 0.000 2.429 
Widowed 0.000 3.218 0.000 3.012 
Single but been living with a partner 0.000 1.601 0.000 1.48 I 
Religion 
Traditional Africa religion(R) 1.00 1.00 
Christianity 0.000 0.886 0.576 1.020 
Islam 0.957 1.006 0.848 1.044 
Ethnicity 
Setswana (R) 1.00 1.00 
lsiXhosa 0.000 1.147 0.000 1.197 
lsiZulu 0.000 1.064 0.030 1.089 
Sepedi 0.000 1.223 0.062 1.085 
Sesotho 0.000 0.885 0.025 0.915 
Place of residence 
Farm Areas(R) 1.00 1.00 
Urban Areas 0.000 0.625 0.000 0.843 
Traditional Areas 0.122 0.962 0.160 1.071 
31 
Province of residence 
Limpopo(R) 1.00 1.00 
Western Cape 0.000 0.617 0.000 0.676 
Eastern Cape 0.000 0.926 0.000 0.801 
Northern Cape 0.000 0.765 0.022 0.842 
Free State 0.000 0.639 0.000 0.756 
KwaZulu Natal 0.000 0.804 0.109 0.926 
North West 0.000 0.770 0.014 0.879 
Gauteng 0.000 0.632 0.000 0.838 
Mpumalanga 0.000 0.781 0.011 0.876 
Boys Sex preference 
3 and above boys(R) 1.00 1.00 
2 boys 0.922 0.000 0.920 0.000 
32 
4.5 Summary of results 
This chapter presented the fertility behavior of Zulu, Xhosa, Pedi , Sotho and Tswana women who form 
part of the black population in South Africa. The analysis was based on PPR that calculates the 
subsequent possibility of women in the black population progressing to the second and other children. 
The analysis started with the correction of data on children ever born by El-Bal dry. The results reveal 
that in all women of selected ethnic parity not stated were less than 2%. Therefore, the data did not need 
the corrections and it was therefore used in calculating the PPR. 
The first step in calculating PPR was to derive data on children ever born by five-year age group and 
calculating the mean parities. The results showed that age group 15-19 has the highest total number 
of children ever born as compared to other age groups. Zulu women have the highest (24 883) children 
ever born followed by Xhosa ( I 02 912). The least was Sotho women by 85 623. The mean parities 
increase as the age groups increase. Again, parity differs from 14 to 17 as per ethnic group 
The second step was to derive births in the previous year prior to the survey as well as the age-specific 
fertility rate. For the selected ethnic groups within the black population, there were I 00 366 births a 
year before the 2016 community survey. Age group 30-34 having the greater percentage whereby Zulu 
women were the ones who had many births. In almost all the selected ethnic group's women aged 20-
24 had the highest births a year before the survey. One can argue that these are age groups where most 
women have completed their tertiary education, are married and some starting employment and 
therefore see it good to increase their parities. All the ethnic groups follow the same pattern of PPR in 
the first child whereby its mostly approximately 80% of women who have at least one child. The age-
specific fertility ratios of women from different ethnic groups differ and range from 0.001 to 0.117. 
Across all ethnic groups, age group 50-54 had the least ASFR. This is probably because these women 
are completing their fertility. 
The step was to establish women who are ever attaining parity. The results show that as the parity and 
age groups increase to up to 17 children, the proportions decrease. Mostly women who are aged 44-49 
are the ones who normally attain parity up to 17 children. The last and the main step of the study was 
calculating PPR of different ethnic groups within the black population. The results show that 12 % of 
women aged 15-19 were childless while for the age group 45-49 was 83%. The results show a pattern 
of declining ratios as parity increases. For example, from 77% of women aged 30-34, only 69% 
progressed to the first child, 45% to the second child and 34% to the third child. In conclusion, the 
results show that Zulu followed by Xhosa women account for the higher fertility within the black 
population. Again, women who are aged 15-19 and 30-34 are the one who has higher fertility levels 
like children ever born. 
33 
The odds ratio shows variations in the results. Two models were used in both adjusted and unadjusted 
versions. In the adjusted model , most variables like the province of residence were statistically 
significant but on the unadjusted odds ratio, some variables changed to not being statistically significant 
like ethnicity. Even though this is the case, some variables like education showed consistency in both 
models. The results further showed that women who obtained higher levels of education were less likely 
to progress having more than two children as compared to those who only had primary or no schooling 
at al I. 
34 
CHAPTER 5: DISCUSSIONS AND RECOMMENDATIONS 
5.1 Introduction 
In estimations and analysing fertility, many studies have been using common well-known measures of 
fertility such as Age-Specific Fertility Rates and Total Fertility Rate which are inclusive of the tempo 
effect. PPR have been neglected more specifically when analysing subsequent birth. As a result, there 
is no information on how different groups progress to higher levels of parity as observed by recent 
studies including Palamuleni (2014; Kolk et al (2014) and StatsSA (2015). 
5.2 Discussion 
While different studies in South Africa have indicated that fertility levels differ with different racial 
groups, no other study known to the author has looked into the differences between the ethnic groups 
among the Black population. The purpose of the study was to estimate the levels of fertility among the 
main ethnic groups in South Africa which are Tswana, Sotho, Pedi Zulu, and Xhosa groups. The study 
used mean parity, age-specific fertility rate, and PPR to estimate the differences in fertility patterns. The 
following section presents the discussion based on the findings in the study. 
Firstly, the results outlined that there is a consistent pattern of declining mean parities for age group 50-
54 which is not consistent with fertility patterns in South Africa. This is unusual , thus StatsSA (2012:54) 
contends that women who give birth at an early age experience a highest number of births in their 
childbearing lifespan. This means women who completed their fertility are out to have higher mean 
parities. The declining mean parities of women aged 5-54 may be due to the misreporting of date by 
respondents. On the other hand, a study done by StatsSA (2012:52) in 2011 showed that Zulus begin 
their childbearing earlier than other ethnic groups followed by the Xhosas, the findings in this study 
indicated that higher portion of women among Tswana in the age group 15-19 had childbearing earlier 
than other age groups. 
The analysis showed that the fertility between the ethnic group tend to differ, and are more pronounced 
in the advanced age groups where the Pedi and Xhosa had the highest mean parity while the Sothos and 
Tswana had the lowest parities. The possible reason for this could be emanating from the cultural 
differences within these groups. The subsequent analyses showed that Pedi , Xhosa, and Zulu were more 
likely to progress to the third parity compared to other ethnic groups. A study by Zuberi et al, (2005) 
has shown similar results indicating that the number of children differs between groups. 
Secondly, the study indicated that the pattern of Age-Specific Fertility rates differed according to ethnic 
groups. It is evident from the results that there are fertility differences among different ethnic groups 
within the black population. This is observed in the ASFR of different ethnic groups as Xhosas and 
35 
Tswanas has higher ASFR at age group 15-19 as compared to other age groups. 2011 Census outlined 
that ASFR of black population ranged from 0.007 to 0.159 across all ages of childbearing. Moreover, 
it was indicted in the literature that fertility has been declining on South Africa. In 2011 , the South 
African TFR was 2.67 and the study results shows that TFR within selected ethnic groups range from 
2.040 to 2.424. 
Lastly, women of the selected ethnic groups progress up to the eighth child whereby Xhosas are the 
highest and this is observed among women aged 50-54 which are women assumed to have completed 
their fertility. As stated in the literature that inception of the first birth is mostly likely to result in more 
births, it is also evident in this study. Fertility gradually increases after the first birth, whereby Xhosas 
and Pedis are still the highest. Sothos are the least to progress to the third child . 
In conclusion, zooming on fertility from ethnicity levels rather than population group give more clarity 
on the fertility behavior of women within the black population. Knowing how women progress from 
one parity to the next is critically vital as it outlines the fertility of women at the end of their childbearing 
ages. Fertility levels of the black population remain slightly above the replacement level. 
5.3 Recommendations 
Given the findings in the study, the following are recommendations: 
1. Given the transition to higher parity levels differed between different ethnic groups, the study 
recommends that cultural issues influencing differentials in the transition to higher be 
investigated. 
ii. Furthermore, the study recommends that sexual and reproductive health issues be raised in rural 
areas where most of these ethnic resides . 
111. Furthermore, given that education plays a significant role in the transition to higher parities, the 
study recommends that education be improved among the different ethnic groups. 
iv. It is recommended that institutions deal with data collection should start to collect, analyze and 
release fertility data on ethnic groups. This is because the study has outlined which ethnic 
groups have the highest levels of fertility. This would enable comprehensive research and in-
depth understanding of fertility behavior. 
v. This could assist policymakers regarding areas of focus when dealing with fertility. It is not 
helpful for the government and policymakers to focus on the population group and confirming 
the high levels of fertility without knowing which ethnic group contributes more 
There should be an interest in explaining fertility from the ethnicity level rather than only focusing on 
the population group. It is recommended that StatsSA should start to collect, analyze and release fertility 
36 
data on ethnic groups. This is because the study has outlined which ethnic groups have the highest levels 
of fertility. This would enable comprehensive research and in-depth understanding offertility behavior. 
Furthermore, this could assist policymakers regarding areas of focus when dealing with fertility. It is 
not helpful for the government and policymakers to focus on the population group and confirming the 
high levels of fertility without knowing which ethnic group contributes more. It would be more 
significant identifying which ethnic group contributes more to high levels of fertility and consequently 
crafting specific policy to be directed to that ethnic group rather than focusing on all groups which 
might not even need those policies. 
37 
References 
Amoateng A.Y. 2004. T he South African Family: Continuity or change. South Africa: MUN Publishers 
Bongaarts, J. , Frank, 0 ., and Lesthaeghe, R. 1984. The proximate determinants of fertility in Sub-
Saharan Africa. Population and deve lopment review, I 0(3): 515 
Bongaarts, J. , and Potter. , R. G . 1983. Fertility, Biology, and Behavior: An Analysis of the Proximate 
Determinants. New York: Academic Press. 
Camlin, C. , Garenne, M & Moultrie T .A. 2004, Fertility Trends and Pattern in a Rural Area of South 
Africa in the Context of HIV/AIDS, African Journal of Reproductive Health, (82): 38-54 
Hinde. A. 2014. Demographic Methods. London: Hodder Headline Group. 
Kibet, M. 2015 . Population processes: Fertility, mortality and Migration-A review: Fertility. 
Potchefstroom: NWU, Potchefstroom Campus. (Study guide HPOP812). 
Kolk, M., Daniel Cownden, D. & Magnus Enquist, M. 2014. Correlations in fertility across 
generations: can low fertility persist? Biological Sciences (281) 1779 
Kumchulesi , G. 2011 . 'Factors affecting age first marriage in Malawi', UA PS (Princeton). 
Majumder, N. & Ram, F. 2015 . Explaining the role of proximate determinants on fertility decline 
among poor and non-poor in Asian countries. PLOS One, 10(2):3 
McDonald, P., Hosseini-Chavoshi , M., Abbasi-Shavazi . M . J., & Rashidian, A . 2015. An assessment 
of recent Iranian fertility trends using Parity Progression Ratios. Demographic Research. 32 (58): 
1584 
Mikva, K. 2015 . Most populous ethnic groups in South Africa. https://moguldom.com/87296/most-
populous-ethnic-groups-in-south-africa/. Date of access : 09 . Nov 2018 
Moultrie, T. , Dorrington, R. , Hill , A. , Hill , K., Timreus, I & Basia Zaba, B. 2013 . Tools for 
Demographic Estimation. France: International Union for the Scientific Study of Population 
Moraa, M .N. 2007. Indirect estimation of Parity Progress ion Ratios from birth interval data. Nairobi -
University of Nairobi (MA Thesis) 
38 
Mutakwa, D. 2012. Analysis of the projected parity progression ratio method using two successive 
censuses/surveys. Cape Town- University of Cape Town (MA Thesis) 
Palamuleni , M. E. 2014. Social and Economic factors affecting Ethnic Fertility differentials in Malawi . 
International Journal of Development and Sustainability, 3( 1) : J 
Palamuleni, M.E. , Kalule-Sabiti , I & Makiwane, M. 2007. Fertility and childbearing in South Africa. 
Pretoria: HSRC 
Pressat. R. 1993. Parity Progression Ratios . Readings in population research methodology, 3( I) 
Preston, S., H., Heuveline, P., & Guillot, M. 2001. Demography: Measuring and modeling population 
processes. USA: Blackwell Publishers Ltd 
Rampagane, K.V. Marriage dynamics and fertility in the era of HIV/AIDS in Mafikeng local 
Municipality of the North West Province South Africa. Mafikeng: NWU (Thesis-PhD) 
Saarela, J. , & Finnas, F. 2014. Sex composition of children, parental separation, and parity progression 
is Finland a Nordic outlier? Demographic Research, 30(2):49 
Sayi, T. S. 2009. Fertility, birth intervals, and their proximate determinants in Zimbabwe. Cape Town: 
University of Cape Town (Thesis-Masters) 
Sloggett. A.2015 . Measuring fertility. In Population Analys is for Policy and Programmes Paris: 
International Union for the Scientific Study of Population . 
http://papp.iussp.org/sessions/papplOI s04/PAPP101 s04 010 010.html. Date of access: 04 Sep 
2019 
South African Demographic Health Survey (SADHS). 1999. South Africa in transition: selected 
findings from the South African Demographic Health Survey. Pretoria: Government Printers. 
Spoorenberg, T., & Dommaraju, P. 2012. Regional fertility transition in India: An analysis using 
synthetic Parity Progression Rations. International Journal ofP opulation Research, I ( 1) : 1 
Spoorenberg, T. 2015. Evaluation and analysis of fertility data. Regional workshop on the production 
of population estimates and demographic indicators . 
http://www.un.org/en/development/desa/population/events/pdf/other/ I I /ppt_Ferti Ii ty .pdf. Date of 
access: 18 Sep 2018 [PowerPoint Presentation] 
39 
Statistics South Africa (StatsSA). 2012. Census 2011: Stati stical Release. Pretoria: Statistics South 
Africa 
Statistics South Africa (StatsSA).2015. Census 2011: Fertility in South Africa. Pretoria: Statistics South 
Africa 
Statistics South Africa (StatsSA). 2017. Mid-year population estimates. Pretoria: Statistics South Africa 
Statistics South Africa (StatsSA). 2018. Mid-year population estimates . Pretoria: Statistics South Africa 
Swartz, L. 2004. Fertility transition in South Africa and its implications on the four major population 
groups. http://www.un.org/esa/population/publications/completingfertility/RevisedSwartzpaper.PDF 
Date of access: 23 Sep 2017 
The Lund Research Ltd . 2018. Binomial Logistic Regression using SPSS Statistics. 
https ://stati stics. Ia erd .com/spss-tutori als /bi nom ial-1 ogisti c-regress ion-us i ng-spss-stati sti cs . ph p. Date 
of access: 12 Aug 2019 
United Nations. 1983. Manual X: Indirect techniques for demographic estimations . 
http://www.un.org/esa/population/publications/Manual X/Manual X.htm Date of access:23 Jun 20 I 8 
Yadava, R.C. , & Sharma, S.S . 2007. Estimation of Parity Progression Ratios using birth interval. The 
journal off amily welfare, 55(1) :50 
Yadava, R.C. , Pandey, A., & Saxena, N.C. 1992. Estimation of Parity Progression Ratios from the 
truncated distribution of closed and open birth intervals. Mathematical Biosciences, 110 (2) 
Zuberi , T. , Sibanda, A. & Udjo, E. 2005 . The demography of South Africa. USA: M.E Sharpe Inc 
Wasserman, P. 2016. Population momentum explained. https ://populationeducation .org/population-
momentum-explained/. Date of access: 18 Oct 2018 
Woods, J. , Neels, K. , & Kil , T. 2014. The educational gradient of childlessness and cohort parity 
progression in 14 low fertility countries. Demography Research. 31 ( 46): 1395 
40 
Addendum 
1. Mean Parities 
Table 2: Total Number of women by age group and parity and Mean Parity 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 101108 59595 34940 22474 16099 12651 11099 2 968 
1 12942 40696 37690 23250 13876 9868 7341 1212 
2 1083 12234 26477 28875 21990 17331 13175 2265 
3 91 2076 9396 15813 16588 15643 13592 2366 
4 16 281 2289 5885 8097 9329 9637 1814 
5 0 30 474 1826 3277 4912 5616 1194 
6 0 9 103 568 1294 2382 3079 715 
7 0 0 20 146 497 1078 1556 337 
8 0 0 6 42 191 505 818 193 
9 0 0 0 15 68 219 429 118 
10 0 0 0 6 30 116 197 68 
11 0 0 0 0 5 41 82 16 
12 0 0 0 0 6 15 42 9 
13 0 0 0 0 0 6 17 4 
14 0 0 0 0 0 2 14 1 
15 0 0 0 0 0 0 4 0 
16 0 0 0 0 0 0 1 I 
17 0 0 0 0 0 0 1 I 
Total 115240 114921 111395 98900 82018 74098 66700 13282 
CEB 15445 72720 1131164 166070 170203 183081 184752 38384 
Mean 
Parity 0,134 0,633 1,177 1,679 2,075 2,471 2,770 2,739 
41 
Table 3: Total births reported in the last year before the survey and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
1 5337 7600 3068 1027 352 98 19 2 
2 1078 7240 9388 6050 2446 686 96 4 
3 123 2400 7260 7947 4869 1578 189 3 
4 24 472 2924 5128 4176 1840 228 4 
5 0 60 925 2435 2640 1570 275 15 
6 0 30 234 1098 1590 1188 174 24 
7 0 0 42 371 903 721 231 7 
8 0 0 16 144 368 400 136 0 
9 0 0 0 18 189 252 63 0 
10 0 0 0 10 50 220 30 0 
11 0 0 0 0 11 88 44 0 
12 0 0 0 0 36 12 36 12 
13 0 0 0 0 0 0 13 0 
14 0 0 0 0 0 0 14 0 
15 
16 0 0 0 0 0 0 16 0 
Total 5923 12155 11145 8727 5240 2151 341 15 
ASFR 0.051 0.106 0.100 0.088 0.064 0.029 0.005 0.000 
42 
Table 7: Number of Zulu women by age group and parity and Mean Parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 36665 22124 14225 9449 6568 5090 4327 1107 
1 4956 14710 13009 8063 4986 3538 2545 395 
2 437 4610 9343 9508 6957 5538 3979 722 
3 35 811 3582 5319 5193 4793 4106 731 
4 7 136 926 2070 2604 2936 2946 538 
5 0 13 178 699 1138 1596 1824 405 
6 0 1 55 222 445 823 1075 268 
7 0 0 11 69 175 393 557 122 
8 0 0 5 16 64 194 312 66 
9 0 0 0 9 24 83 154 46 
10 0 0 0 4 15 40 69 28 
11 0 0 0 0 3 13 31 7 
12 0 0 0 0 4 7 15 4 
13 0 0 0 0 0 2 5 2 
14 0 0 0 0 0 2 2 1 
15 0 0 0 0 0 0 2 0 
16 0 0 0 0 0 0 0 1 
Total 42100 42405 41334 35428 28176 25048 21949 4443 
Total 2348 4276 3642 2702 1583 627 103 5 
Births 
Mean 
parity 0,142 0,636 1,149 1,605 1,968 2,371 2,701 2,718 
43 
Table 8: Number of Xhosa women by age group and parity and Mean Parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 24454 14174 8075 5241 3790 2907 2677 734 
1 3611 9880 8957 5701 3475 2516 1834 313 
2 297 3123 6158 6662 5374 4185 3233 542 
3 26 555 2294 3756 3975 3842 3282 576 
4 6 68 573 1474 2023 2322 2482 489 
5 0 4 113 448 881 1355 1539 327 
6 0 2 17 138 370 713 903 189 
7 0 0 4 28 155 340 497 115 
8 0 0 0 11 64 175 266 65 
9 0 0 0 3 20 74 179 41 
10 0 0 0 1 8 43 88 27 
11 0 0 0 0 1 21 30 5 
12 0 0 0 0 I 5 16 5 
13 0 0 0 0 0 4 8 I 
14 0 0 0 0 0 0 8 0 
TOTAL 3940 13632 18116 18222 16347 15595 14365 2695 
Total 1601 3094 2801 2180 1300 563 96 4 
Births 
44 
Table 9: Number of Pedi women by age group and parity and Mean Parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 14878 8880 4838 2914 2181 1747 1588 411 
1 1543 5818 5601 2965 1493 964 730 140 
2 110 1576 3859 4455 2954 2124 1544 254 
3 9 232 1206 2424 2724 2448 1932 315 
4 I 29 242 865 1430 1715 1617 271 
5 0 4 53 230 533 898 985 182 
6 0 I 9 64 189 396 537 143 
7 0 0 I 10 75 172 258 58 
8 0 0 I 6 21 69 134 31 
9 0 0 0 I 6 30 61 14 
10 0 0 0 0 3 19 20 10 
11 0 0 0 0 I 3 11 3 
12 0 0 0 0 0 I 6 0 
13 0 0 0 0 0 0 3 I 
14 0 0 0 0 0 0 I 0 
15 0 0 0 0 0 0 I 0 
16 0 0 0 0 0 0 0 I 
TOTAL 16541 16540 15810 13934 11610 10586 9428 1834 
Total 659 1760 1697 1428 891 413 75 5 
Births 
Mean 
parity 0,108 0,593 1,154 1,742 2,229 2,696 2,982 2,937 
45 
Table 10: Number of Sotho women by age group and parity and Mean Parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 11992 7192 3995 2554 1752 1414 1179 349 
1 1284 4944 5129 3441 2083 1479 1092 194 
2 97 1337 3566 4153 3379 2786 2272 338 
3 5 196 1056 2118 2312 2301 2086 357 
4 I 19 214 641 953 1062 1203 246 
5 0 6 61 200 320 486 575 129 
6 0 2 11 51 129 197 250 52 
7 0 0 1 19 33 55 103 23 
8 0 0 0 3 16 21 44 10 
9 0 0 0 1 6 8 12 5 
10 0 0 0 1 0 2 8 2 
11 0 0 0 0 0 3 3 1 
12 0 0 0 0 0 0 1 0 
13 0 0 0 0 0 0 2 0 
TOTAL 13379 13696 14033 13182 10983 9814 8830 1706 
Total Births 588 1370 1418 1138 691 265 28 1 
Mean 
Parity 0,112 0,608 1,187 1,680 2,037 2,292 2,538 2,461 
46 
Table 11: Number of Tswana women by age group and parity and Mean Parity 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 13099 7204 3784 2287 1783 1466 1304 363 
1 1548 5344 4994 3080 1839 1371 1140 170 
2 142 1588 355 1 4097 3326 2698 2147 409 
3 16 282 1258 2196 2384 2259 2186 387 
4 1 29 334 835 1087 1294 1389 270 
5 0 3 69 249 405 577 693 151 
6 0 3 11 93 161 253 314 63 
7 0 0 3 20 59 118 141 19 
8 0 0 0 6 26 46 62 21 
9 0 0 0 1 12 24 23 12 
10 0 0 0 0 4 12 12 1 
11 0 0 0 0 0 I 7 0 
12 0 0 0 0 1 2 4 0 
13 0 0 0 0 0 0 I 0 
14 0 0 0 0 0 0 I 0 
15 0 0 0 0 0 0 I 0 
16 0 0 0 0 0 0 I 0 
17 0 0 0 0 0 0 1 0 
18+ 727 1655 1587 1279 775 283 39 12 
TOTAL 14806 14453 14004 12864 11087 10121 9427 1866 
Total Births 
Mean 
Parity 0,127 0,658 1,259 1,804 2,143 2,439 2,642 2,562 
47 
2. Births reported a year before survey and ASFR 
Table 12: Zulu Births reported in the last year before the survey and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
I 2118 2655 984 338 138 27 10 1 
2 420 2516 2950 1804 706 234 36 2 
3 54 903 2490 2406 1368 441 57 0 
4 8 220 1080 1548 1196 516 64 4 
5 0 30 295 890 880 415 80 0 
6 0 6 120 396 594 366 60 6 
7 0 0 14 154 280 231 42 0 
8 0 0 16 56 88 80 24 0 
9 0 0 0 0 63 81 18 0 
10 0 0 0 0 20 60 0 0 
11 0 0 0 0 0 55 11 0 
12 0 0 0 0 24 0 12 12 
13 0 0 0 0 0 0 13 0 
14 
15 
TOTAL 2348 4276 3642 2702 1583 627 103 5 
ASFR 0.056 0.101 0.088 0.076 0.056 0.025 0.005 0.001 
48 
Table 13: Xhosa Births reported in the last year before the survey and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
I 1432 1882 819 287 92 27 2 0 
2 302 1954 2214 1464 656 186 26 2 
3 45 621 1887 637 407 136 18 0 
4 12 104 756 1332 936 400 56 0 
5 0 10 250 645 595 395 80 10 
6 0 0 30 276 396 330 48 0 
7 0 0 14 70 224 245 77 7 
8 0 0 0 40 120 152 24 0 
9 0 0 0 9 45 81 36 0 
10 0 0 0 0 10 70 20 0 
11 0 0 0 0 0 22 22 0 
12 0 0 0 0 12 12 24 0 
13 
14 0 0 0 0 0 0 14 0 
Total 1601 3094 2801 2180 1300 563 96 4 
ASFR 0.056 0.111 0.107 0.093 0.065 0.030 0.006 0.001 
49 
Table 14: Pedi Births reported in the last year before the and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
1 602 1173 490 122 37 19 3 1 
2 112 966 1546 1060 326 90 18 0 
3 3 276 1008 1425 834 288 31 2 3 
4 0 40 284 884 948 420 40 0 
5 0 10 105 275 480 72 70 0 
6 0 0 36 108 282 228 42 18 
7 0 0 0 35 154 119 56 0 
8 0 0 0 16 48 96 72 0 
9 0 0 0 0 666 36 9 0 
10 0 0 0 0 0 50 10 0 
11 0 0 0 0 11 0 11 0 
12 
13 
14 
15 
16 
659 1760 1697 1428 891 413 75 5 
Total 
ASFR 0.040 0.106 0.107 0.102 0.077 0.039 0.008 0.003 
50 
Table 15: Sotho Births reported in the last year before the survey and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
I 535 875 373 157 45 15 2 0 
2 104 810 1304 832 416 82 14 0 
3 0 243 855 987 678 219 21 0 
4 4 24 316 580 460 240 36 0 
5 0 5 125 305 270 190 10 5 
6 0 12 24 120 132 138 6 0 
7 0 0 0 49 84 49 0 0 
8 0 0 0 8 48 32 0 0 
9 0 0 0 9 27 18 0 0 
JO 0 0 0 10 0 JO 0 0 
11 0 0 0 0 0 11 0 0 
12 
14 
588 1370 1418 1138 691 265 28 5 
Total 
ASFR 0.044 0.100 0.101 0.086 0.063 0.027 0.003 0.001 
51 
Table 16: Tswana Births reported in the last year before the survey and ASFR 
Parity 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
1 650 1015 402 123 40 10 2 5 
2 140 994 1374 890 342 94 2 4 
3 21 357 1020 1218 768 222 21 24 
4 0 84 488 784 636 264 32 23 
5 0 5 150 320 415 210 35 9 
6 0 12 24 198 631 126 18 5 
7 0 0 14 63 161 77 56 34 
8 0 0 0 24 64 40 16 0 
9 0 0 0 0 18 36 0 5 
10 0 0 0 0 20 30 0 5 
11 
12 
13 
14 
15 
16 0 0 0 0 0 0 16 10 
17 
Total 727 1655 1587 1279 775 283 39 12 
ASFR 0.049 0.115 0.113 0.099 0.070 0.028 0.004 0.006 
52 
3. Proportions ever-attaining 
Table 17: Total proportions ever attaining Parity-Zulu 
Parity 
15-19 20-24 25-29 30-34 35-39 40-44 
(i) 45-49 50-54 
0 1 1 1 1 1 1 I I 
I 0,129 0,478 0,656 0,733 0,767 0,797 0,803 0,751 
2 0,011 0,131 0,341 0,506 0,59 0,656 0,687 0,662 
3 0,001 0,023 0, 115 0,237 0,343 0,434 0,506 0,499 
4 0 0,004 0,028 0,087 0,159 0,243 0,319 0,335 
5 0 0 0,006 0,029 0,066 0,126 0,184 0,214 
6 0 0 0,002 0,009 0,026 0,062 0,101 0, 123 
7 0 0 0 0,003 0,01 0,029 0,052 0,062 
8 0 0 0 0,001 0,004 0,014 0,027 0,035 
9 0 0 0 0 0,002 0,006 0,013 0,02 
10 0 0 0 0 0,001 0,003 0,006 0,01 
11 0 0 0 0 0 0,001 0,003 0,003 
12 0 0 0 0 0 0 0,001 0,002 
13 0 0 0 0 0 0 0 0,001 
53 
Table 18: Total proportions ever attaining Parity-Xhosa 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 
1 0,139 0,490 0,692 0,777 0,812 0,843 0,843 0,786 
2 0,012 0, 135 0,350 0,534 0,639 0,707 0,735 0,695 
3 0,001 0,023 0, 115 0,250 0,372 0,481 0,546 0,537 
4 0,000 0,003 0,027 0,090 0, 175 0,273 0,353 0,369 
5 0,000 0,000 0,005 0,027 0,074 0,148 0,207 0,226 
6 0,000 0,000 0,001 0,008 0,031 0,074 0,117 0, 131 
7 0,000 0,000 0,000 0,002 0,012 0,036 0,064 0,076 
8 0,000 0,000 0,000 0,001 0,005 0,017 0,035 0,042 
9 0,000 0,000 0,000 0,000 0,001 0,008 0,019 0,023 
10 0,000 0,000 0,000 0,000 0,000 0,004 0,009 0,011 
11 0,000 0,000 0,000 0,000 0,000 0,002 0,004 0,003 
12 0,000 0,000 0,000 0,000 0,000 0,000 0,002 0,002 
13 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
54 
Table 19: Total proportions ever attaining Parity-Pedi 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 1,00 1,00 1,00 1,00 1,00 1,00 1,00 1,00 
1 0,10 0,46 0,69 0,79 0,81 0,83 0,83 0,78 
2 0,01 0, 11 0,34 0,58 0,68 0,74 0,75 0,70 
3 0,00 0,02 0,10 0,26 0,43 0,54 0,59 0,56 
4 0,00 0,00 0,02 0,08 0, 19 0,31 0,39 0,39 
5 0,00 0,00 0,00 0,02 0,07 0, 15 0,21 0,24 
6 0,00 0,00 0,00 0,01 0,03 0,07 0, 11 0,14 
7 0,00 0,00 0,00 0,00 0,01 0,03 0,05 0,06 
8 0,00 0,00 0,00 0,00 0,00 0,01 0,03 0,03 
9 0,00 0,00 0,00 0,00 0,00 0,01 0,01 0,02 
10 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,01 
55 
Table 20: Total proportions ever atta ining Parity-Sotho 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 
1 0, 104 0,475 0,715 0,806 0,840 0,856 0,866 0,795 
2 0,008 0,114 0,350 0,545 0,651 0,705 0,743 0,682 
3 0,000 0,016 0,096 0,230 0,343 0,421 0,486 0,484 
4 0,00 0,002 0,020 0,069 0, 133 0,187 0,249 0,274 
5 0,000 0,001 0,005 0,02 1 0,046 0,079 0, 113 0, 130 
6 0,000 0,000 0,001 0,006 0,017 0,029 0,048 0,055 
7 0,000 0,000 0,000 0,002 0,005 0,009 0,020 0,024 
8 0,000 0,000 0,000 0,000 0,002 0,003 0,008 0,01 1 
9 0,000 0,000 0,000 0,000 0,001 0,001 0,003 0,005 
10 0,000 0,000 0,000 0,000 0,000 0,001 0,002 0,002 
11 0,000 0,000 0,000 0,000 0,000 0,000 0,001 0,00 1 
12 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
13 0,000 0,000 0,000 0,000 0,000 0,000 0,000 0,000 
56 
Table 21: Total proportions ever attaining Parity-Tswana 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 
1 0, 115 0,502 0,730 0,822 0,839 0,855 0,862 0,805 
2 0,011 0,132 0,373 0,583 0,673 0,720 0,741 0,714 
3 0,001 0,022 0,120 0,264 0,373 0,453 0,513 0,495 
4 0,000 0,002 0,030 0,094 0,158 0,230 0,281 0,288 
5 0,000 0,000 0,006 0,029 0,060 0,102 0, 134 0,143 
6 0,000 0,000 0,001 0,009 0,024 0,045 0,060 0,062 
7 0,000 0,000 0,000 0,002 0,009 0,020 0,027 0,028 
8 0,000 0,000 0,000 0,001 0,004 0,008 0,012 0,018 
9 0,000 0,000 0,000 0,000 0,002 0,004 0,005 0,007 
10 0,000 0,000 0,000 0,000 0,000 0,001 0,003 0,001 
11 0,000 0,000 0,000 0,000 0,000 0,000 0,002 0,000 
12 0,000 0,000 0,000 0,000 0,000 0,000 0,001 0,000 
13 0,000 0,000 0,000 0,000 0,000 0,000 0,001 0,000 
57 
4. Parity Progression Ratios 
Table 22: Parity Progression Ratio of Zulus 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 0,129 0,478 0,656 0,733 0,767 0,797 0,803 0,751 
1 0,088 0,275 0,520 0,690 0,769 0,823 0,856 0,882 
2 0,088 0,173 0,337 0,469 0,581 0,663 0,736 0,755 
3 0,167 0,156 0,247 0,367 0,463 0,560 0,630 0,671 
4 0,000 0,093 0,212 0,3 30 0,418 0,518 0,579 0,638 
5 0,071 0,285 0,314 0,391 0,494 0,549 0,574 
6 0,000 0,225 0,306 0,390 0,471 0,516 0,508 
7 0,313 0,296 0,386 0,465 0,514 0,560 
8 0,000 0,448 0,418 0,431 0,471 0,574 
9 0,308 0,478 0,435 0,446 0,483 
10 0,000 0,318 0,375 0,444 0,349 
11 0,57 1 0,458 0,436 0,533 
12 0,000 0,364 0,375 0,500 
13 0,500 0,444 0,500 
14 0,000 0,500 0,500 
15 0,000 0,900 
16 0,000 
17 
58 
Table 23: Parity Progression Ratio ofXhosas 
Parity (i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 0,139 0,490 0,692 0,777 0,812 0,843 0,843 0,786 
l 0,084 0,275 0,506 0,687 0,787 0,839 0,872 0,884 
2 0,097 0,168 0,328 0,468 0,583 0,680 0,742 0,772 
3 0,188 0, 118 0,236 0,359 0,470 0,568 0,647 0,687 
4 0,000 0,08 1 0,190 0,299 0,426 0,540 0,587 0,613 
5 0,333 0,157 0,288 0,413 0,504 0,565 0,578 
6 0,000 0,190 0,238 0,402 0,481 0,547 0,578 
7 0,000 0,349 0,378 0,486 0,545 0,556 
8 0,267 0,319 0,457 0,553 0,549 
9 0,250 0,333 0,497 0,456 0,481 
10 0,000 0,200 0,411 0,413 0,289 
11 0,500 0,300 0,516 0,545 
12 0,000 0,000 0,250 0,000 
13 
59 
Table 24: Parity Progression Ratio of Pedi 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 0,101 0,463 0,694 0,791 0,812 0,835 0,832 0,776 
1 0,072 0,240 0,490 0,731 0,842 0,891 0,907 0,902 
2 0,083 0,144 0,282 0,447 0,628 0,730 0,783 0,802 
3 0,100 0,128 0,202 0,327 0,453 0,574 0,653 0,694 
4 0,000 0,147 0,209 0,264 0,367 0,481 0,555 0,620 
5 0,200 0,172 0,260 0,356 0,435 0,512 0,589 
6 0,000 0,182 0,210 0,359 0,426 0,480 0,452 
7 0,500 0,412 0,292 0,415 0,479 0,508 
8 0,000 0,143 0,323 0,434 0,435 0,483 
9 0,000 0,400 0,434 0,408 0,517 
10 0,250 0,174 0,524 0,333 
11 0,000 0,250 0,500 0,400 
12 0,000 0,455 0.900 
13 0,400 0,500 
14 0,500 0.900 
15 0,000 0.900 
16 0,000 
60 
Table 25: Parity Progression Ratio of Sotho's 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 0, 104 0,475 0,715 0,806 0,840 0,856 0,866 0,795 
1 0,074 0,240 0,489 0,676 0,774 0,824 0,857 0,857 
2 0,058 0,143 0,274 0,422 0,527 0,597 0,654 0,709 
3 0,167 0,121 0,214 0,302 0,387 0,444 0,513 0,567 
4 0,000 0,296 0,254 0,300 0,346 0,421 0,453 0,474 
5 0,250 0,164 0,273 0,365 0,370 0,424 0,419 
6 0,000 0,083 0,320 0,299 0,311 0,409 0,441 
7 0,000 0,208 0,400 0,382 0,405 0,439 
8 0,400 0,273 0,3 82 0,371 0,444 
9 0,500 0,000 0,385 0,538 0,375 
10 0,600 0,429 0,333 
11 0,000 0,500 0,000 
12 0,667 
13 
61 
Table 26: Parity Progression Ratio of Tswana's 
Parity 
(i) 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 
0 0, 115 0,502 0,730 0,822 0,839 0,855 0,862 0,805 
I 0,093 0,263 0,5 I I 0,709 0,802 0,842 0,860 0,887 
2 0,107 0,166 0,321 0,454 0,554 0,630 0,693 0,693 
3 0,059 0, 110 0,249 0,354 0,424 0,507 0,548 0,581 
4 0,000 0, 171 0,199 0,306 0,38 1 0,444 0,476 0,497 
5 0,500 0,169 0,325 0,394 0,441 0,450 0,434 
6 0,000 0,214 0,225 0,388 0,445 0,447 0,457 
7 0,000 0,259 0,422 0,419 0,445 0,642 
8 0,143 0,395 0,459 0,451 0,382 
9 0,000 0,294 0,385 0,549 0,077 
10 0,200 0,200 0,571 0,000 
11 1,000 0,667 0,563 0.000 
12 0,000 0,000 0,556 0.000 
13 0,800 0.000 
14 0,750 0.000 
15 0,667 0.000 
16 0,500 0.000 
17 0,000 0.000 
18+ 
62