Swaziland - Demographic and Health Survey - 2008

Publication date: 2008

Swaziland Demographic and Health Survey 2006-07 Swaziland Demographic and Health Survey 2006-07 Central Statistical Office Mbabane, Swaziland Macro International Inc. Calverton, Maryland USA May 2008 This report summarises the findings of the 2006-07 Swaziland Demographic and Health Survey (SDHS) carried out by the Swaziland Central Statistical Office (CSO). The SDHS is part of the worldwide MEASURE Demographic and Health Surveys (DHS) pogram, funded by the United States Agency for International Development (USAID). Most of the funds for the local costs of the survey were provided by the Government of Swaziland and multiple donors, namely the National Emergency Response Council on HIV/AIDS (NERCHA), UNAIDS/Swaziland, UNFPA/Swaziland, UNICEF/Swaziland, Italian Corporation the World Health Organization, and Population Services International (PSI). The United States Agency for International Development (USAID) provided additional funds through the MEASURE DHS project to support the local costs and the technical assistance provided by Macro International Inc. Through an arrangment with Macro, the Global Clinical Virology Laboratories (GCVL) of South Africa also provided assistance with the training and laboratory processing for the HIV testing component of the survey. The Human Sciences Research Council (HSRC) of South Africa assisted during the design phase of the survey. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID or other funding partners. Additional information about the survey may be obtained from Central Statistical Office (CSO), Ministry of Economic Planning and Development, P.O. Box 456 Mbabane, Swaziland H100 (Telephone 268-404-2151; Fax 268-404-3300; e-mail: statistics@africaonline.co.sz). Additional information about the DHS program may be obtained by writing to: MEASURE DHS, Macro International Inc., 11785 Beltsville Drive, Suite 300, Calverton, MD 20705, USA (Telephone 301-572-0200; Fax 301-572-0999; e-mail: reports@orcmacro.com). Suggested citation: Central Statistical Office (CSO) [Swaziland], and Macro International Inc. 2008. Swaziland Demographic and Health Survey 2006-07. Mbabane, Swaziland: Central Statistical Office and Macro International Inc. Contents | iii CONTENTS Page TABLES AND FIGURES .xi PREFACE. xxi SUMMARY OF FINDINGS . xxiii MAP OF SWAZILAND .xxviii CHAPTER 1 INTRODUCTION Henry Ginindza and Rachel Masuku 1.1 History, Geography, and Economy. 1 1.2 Population . 2 1.3 Population, Family Planning, and HIV Policies and Programmes. 2 1.4 Objectives. 3 1.5 Orgnisation of the Survey. 3 1.6 Sample Design . 4 1.7 Questionnaires. 4 1.8 Anaemia and HIV Testing. 5 1.9 Pretest, Training, and Fieldwork . 5 1.10 HIV Testing . 6 1.11 Data Processing. 7 1.12 Response Rates . 7 CHAPTER 2 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS Petronella Mamba 2.1 Population by Age And Sex .9 2.2 Household Composition .10 2.3 Education of the Household.11 2.3.1 Educational Attainment.11 2.3.2 School Attendance Rates .13 2.3.3 Grade Repetition and Dropout Rates .15 2.4 Household Environment .17 2.4.1 Drinking Water.17 2.4.2 Household Sanitation Facilities .19 2.4.3 Housing Characteristics.19 2.5 Household Possessions.22 2.6 Wealth Index .23 2.7 Birth Registration.24 iv │ Contents CHAPTER 3 CHARACTERISTICS OF RESPONDENTS Henry Ginindza and Petronella Mamba 3.1 Characteristics of Survey Respondents.27 3.2 Educational Attainment by Background Characteristics.30 3.3 Literacy .31 3.4 Access to Mass Media .34 3.5 Employment .39 3.6 Occupation.44 3.7 Earnings and Type of Employment .47 CHAPTER 4 FERTILITY LEVELS, TRENDS, AND DIFFERENTIALS Dudu Dlamini 4.1 Introduction.49 4.2 Current Fertility.49 4.3 Fertility Differentials .50 4.4 Fertility Trends .51 4.5 Children Ever Born and Living.53 4.6 Birth Intervals.54 4.7 Age at First Birth.55 4.8 Teenage Pregnancy and Motherhood.57 CHAPTER 5 FAMILY PLANNING Marjorie Mavuso 5.1 Knowledge of Contraceptive Methods.59 5.2 Ever Use of Contraception .61 5.3 Current Use of Contraceptive Methods .64 5.4 Number of Children at First Use of Contraception.69 5.5 Use of Social Marketing Brands .70 5.6 Disposal of Condoms .71 5.7 Knowledge of the Fertile Period .71 5.8 Timing of Sterilisation.71 5.9 Source of Contraception .71 5.10 Cost of Contraceptives .72 5.11 Informed Choice.73 5.12 Future Use of Contraception .75 5.13 Reasons for Non-use of Contraception .75 5.14 Preferred Method of Contraception for Future Use .76 5.15 Exposure to Family Planning Messages .76 5.16 Contact of Nonusers with Family Planning Providers .78 5.17 Husband’s Knowledge of His Wife’s Use of Contraception.79 5.18 Male Attitudes about Contraceptive Use .80 Contents | v CHAPTER 6 OTHER DETERMINANTS OF FERTILITY Sri Poedjastoeti 6.1 Current Marital Status .83 6.2 Polygyny .84 6.3 Age at First Marriage .86 6.4 Age at First Sexual Intercourse.88 6.5 Recent Sexual Activity .91 6.6 Postpartum Amenorrhoea, Abstinence, and Insusceptibility.94 6.7 Termination of Exposure to Pregnancy .96 CHAPTER 7 FERTILITY PREFERENCES Dudu Dlamini 7.1 Desire for More Children .97 7.2 Desire to Limit Childbearing by Background Characteristics .98 7.3 Need for Family Planning Services.99 7.4 Ideal Number of Children . 101 7.5 Mean Ideal Number of Children by Background Characteristics . 103 7.6 Fertility Planning Status . 103 7.7 Wanted Fertility Rates . 104 CHAPTER 8 INFANT AND CHILD MORTALITY Sri Poedjastoeti 8.1 Data Quality . 107 8.2 Levels and Trends in Infant and Child Mortality. 108 8.3 Socio-economic and Demographic Differentials in Infant and Child Mortality . 110 8.4 Perinatal Mortality. 112 8.5 High-Risk Fertility Behaviour . 113 CHAPTER 9 MATERNAL HEALTH Nhlanhla M. Nhlabatsi 9.1 Antenatal Care . 115 9.2 Number of ANC Visits, Timing of First Visit, and Source Where ANC Received. 116 9.3 Components of Antenatal Care . 117 9.4 Tetanus Toxoid Injections . 119 9.5 Place of Delivery. 120 9.6 Assistance during Delivery. 121 9.7 Postnatal Care. 123 9.8 Problems in Accessing Health Care . 125 vi │ Contents CHAPTER 10 CHILD HEALTH Zodwa Dlamini-Mthethwa 10.1 Child’s Size at Birth . 127 10.2 Vaccination Coverage . 127 10.2.1 Collection of Data . 128 10.2.2 Level and Trend in Vaccination Coverage . 129 10.2.3 Vaccination Coverage by Background Characteristics. 130 10.3 Acute Respiratory Infection and Fever . 131 10.4 Diarrhoeal Disease. 133 10.4.1 Prevalence of Diarrhoea . 133 10.4.2 Treatment of Diarrhoea . 134 10.4.3 Feeding Practices. 136 10.5 Knowledge of ORS Packets . 138 10.6 Stool Disposal . 138 CHAPTER 11 NUTRITION OF CHILDREN AND ADULTS Nelisiwe Sikhosana 11.1 Nutritional Status of Children . 141 11.1.1 Measurement of Nutritional Status among Young Children . 141 11.1.2 Results of Data Collection . 141 11.1.3 Nutritional Status Indices . 143 11.2 Initiation of Breastfeeding. 145 11.3 Breastfeeding Status by Age. 146 11.4 Duration and Frequency of Breastfeeding . 148 11.5 Dietary Diversity among Young Children. 149 11.5.1 Foods and Liquids Consumed by Infants and Young Children . 149 11.5.2 Infant and Young Child Feeding (IYCF) Practices . 150 11.6 Use of Iodized Salt . 152 11.7 Prevalence of Anaemia in Children . 153 11.7.1 Children Age 6-69 Months. 154 11.7.2 Children Age 5-11 and 12-14 Years . 155 11.8 Micronutrient Intake among Children. 157 11.8.1 Consumption of Vitamin A-rich and Iron-rich Foods . 157 11.9 Nutritional Status of Women and Men . 160 11.10 Foods Consumed by Mothers. 162 11.11 Prevalence of Anaemia in Women and Men . 163 11.12 Micronutrient Intake among Mothers . 166 CHAPTER 12 MALARIA AND OTHER HEALTH ISSUES Africa Magongo 12.1 Malaria . 169 12.1.1 Ownership and Use of Mosquito Nets . 169 12.1.2 Intermittent Preventive Treatment (IPT) by Women during Pregnancy. 172 Contents | vii 12.1.3 Prompt Treatment of Fever in Children. 174 12.1.4 Indoor Residual Spraying . 175 12.2 Male Circumcision . 175 12.3 Health Insurance Coverage . 178 12.4 Knowledge and Attitudes Towards Tuberculosis . 179 12.5 Use of Tobacco. 181 CHAPTER 13 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOUR Rejoice Nkambule 13.1 Introduction. 183 13.2 HIV/AIDS Knowledge, Transmission, and Prevention. 183 13.2.1 Awareness of HIV and AIDS. 183 13.2.2 Knowledge of HIV Prevention . 184 13.2.3 Rejection of Misconceptions about HIV/AIDS . 186 13.3 Knowledge of Prevention of Mother-to-Child Transmission of HIV . 188 13.4 Attitudes Towards People Living With AIDS . 190 13.5 Attitudes towards Negotiating Safer Sex. 192 13.6 Attitudes Towards Condom Education for Youth . 193 13.7 Higher-Risk Sex. 193 13.7.1 Multiple Partners and Condom Use . 193 13.7.2 Condom Use and Knowledge of Source. 196 13.8 Coverage of HIV Testing and Counselling. 197 13.9 Self-Reporting of Sexually Transmitted Infections. 201 13.10 Prevalence of Medical Injections . 204 13.11 HIV/AIDS Knowledge and Sexual Behaviour among Youth. 205 13.11.1 HIV/AIDS-Related Knowledge among Young Adults . 205 13.11.2 Knowledge of Condom Sources among Young Adults. 207 13.11.3 Age at First Sex. 207 13.11.4 Condom Use at First Sex. 208 13.11.5 Abstinence and Premarital Sex . 209 13.11.6 Higher-Risk Sexual Intercourse and Condom Use among Young People. 210 13.11.7 Cross-Generational Sexual Partners . 213 13.11.8 Drunkenness during Sexual Intercourse among Young People. 213 13.11.9 HIV Testing and Counselling among Young People . 215 CHAPTER 14 HIV PREVALENCE AND ASSOCIATED FACTORS Rachel Masuku 14.1 Coverage of HIV Testing. 217 14.2 HIV Prevalence . 221 14.2.1 HIV Prevalence by Age . 221 14.2.2 HIV Prevalence by Residence, Region, and Wealth . 222 14.2.3 HIV Prevalence by Education and Employment. 224 14.2.4 HIV Prevalence by Sociodemographic Characteristics . 225 14.2.5 HIV Prevalence by Sexual Behaviour Indicators. 228 viii │ Contents 14.3 HIV Prevalence by Other Characteristics Related to HIV Risk . 233 14.4 HIV Prevalence and Male Circumcision . 235 14.5 HIV Prevalence among Youth. 235 14.6 HIV Prevalence among Couples . 238 CHAPTER 15 ADULT AND MATERNAL MORTALITY Ann A. Way 15.1 Data. 239 15.2 Direct Estimates of Adult Mortality . 240 15.3 Direct Estimates of Maternal Mortality. 241 CHAPTER 16 WOMEN’S EMPOWERMENT AND DEMOGRAPHIC AND HEALTH OUTCOMES Rachel Masuku 16.1 Employment and Form of Earnings . 243 16.1.1 Women’s Control Over Their Own Earnings and Relative Magnitude of Women’s Earnings. 245 16.1.2 Control over Husband’s Earnings. 246 16.2 Woman’s Participation in Decisionmaking . 248 16.3 Attitudes Towards Wife Beating . 250 16.4 Attitudes Towards Refusing Sex with Husband . 253 16.5 Women’s Empowerment Indicators . 256 16.6 Current Use of Contraception by Women’s Status. 257 16.7 Ideal Family Size and Unmet Need By Women’s Status . 258 16.8 Women’s Status and Reproductive Health Care . 259 16.9 Early Childhood Mortality Rates by Women’s Status. 260 CHAPTER 17 ORPHANS AND VULNERABLE CHILDREN Amos Zwane 17.1 Children’s Living Arrangements and Orphanhood . 263 17.2 Orphans and Vulnerable Children. 264 17.3 Social and Economic Situation of Orphaned and Vulnerable Children. 266 17.3.1 School Attendance by Survivorship of Parents and OVC Status. 266 17.3.2 Basic Material Needs. 267 17.4 Orphans Not Living with Siblings. 268 17.5 Underweight Orphans and Vulnerable Children. 268 17.6 Early Sexual Intercourse . 269 17.7 Succession Planning . 270 17.8 Widows Dispossessed of Property . 271 17.9 External Support for Very Sick Persons . 271 17.10 External Support for Orphans and Vulnerable Children . 273 Contents | ix CHAPTER 18 YOUTH Africa Magongo and Nelisiwe Sikhosana 18.1 Caregiver . 276 18.2 Supervision Going to and from School and at School . 277 18.3 Knowledge and Attitudes about Sex . 279 18.4 Knowledge and Attitudes about HIV/AIDS. 281 18.5 Exposure to Media Messages on HIV/AIDS. 283 18.6 Opinion on Information about HIV/AIDS . 286 18.7 Knowledge of HIV/AIDS Help Line. 288 18.8 Discussion about HIV/AIDS. 289 18.9 Knowledge of Places to be Tested for the AIDS Virus. 290 REFERENCES .293 APPENDIX A SAMPLE IMPLEMENTATION . 295 APPENDIX B ESTIMATES OF SAMPLING ERRORS. 301 APPENDIX C DATA QUALITY TABLES . 321 APPENDIX D EFFECT OF NONRESPONSE ON THE SDHS HIV PREVALENCE RESULTS. 325 APPENDIX E PERSONS INVOLVED IN THE 2006-07 SWAZILAND DEMOGRAPHIC AND HEALTH SURVEY. 327 APPENDIX F QUESTIONNAIRES . 331 Tables and Figures | xi TABLES AND FIGURES Page CHAPTER 1 INTRODUCTION Table 1.1 Basic demographic indicators. 2 Table 1.2 Results of the household and individual interviews. 7 Table 1.3 Results of the household and individual interviews in households selected for youth and older adults survey. 8 CHAPTER 2 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS Table 2.1 Household population by age, sex, and residence. 9 Table 2.2 Household composition. 11 Table 2.3.1 Educational attainment of the female household population . 12 Table 2.3.2 Educational attainment of the male household population . 13 Table 2.4 School attendance ratios . 14 Table 2.5 Grade repetition and dropout rates. 16 Table 2.6 Household drinking water. 18 Table 2.7 Household sanitation facilities. 19 Table 2.8 Housing characteristics. 20 Table 2.9 Type of windows . 22 Table 2.10 Household possessions . 23 Table 2.11 Wealth quintiles. 24 Table 2.12 Birth registration of children under age five . 25 Figure 2.1 Population Pyramid . 10 Figure 2.2 Age-Specific Attendance Rates of the De-Facto Population 5 to 24 Years . 15 CHAPTER 3 CHARACTERISTICS OF RESPONDENTS Table 3.1.1 Background characteristics of respondents: Women and men 15-49 . 28 Table 3.1.2 Background characteristics of respondents: Youth age 12-14 . 29 Table 3.1.3 Background characteristics of respondents: Older adults age 50+. 29 Table 3.2.1 Educational attainment by background characteristics: Women 15-49. 30 Table 3.2.2 Educational attainment by background characteristics: Men 15-49 . 31 Table 3.3.1 Literacy: Women 15-49. 32 Table 3.3.2 Literacy: Men 15-49 . 32 Table 3.3.3 Literacy: Youth age 12-14. 33 Table 3.3.4 Literacy: Older adults age 50+ . 34 Table 3.4.1 Exposure to mass media: Women 15-49. 35 Table 3.4.2 Exposure to mass media: Men 15-49 . 36 Table 3.4.3 Exposure to mass media: Youth age 12-14. 37 Table 3.4.4 Exposure to mass media: Older adults age 50+ . 38 xii | Tables and Figures Table 3.5.1 Employment status: Women 15-49. 40 Table 3.5.2 Employment status: Men 15-49 . 42 Table 3.5.3 Employment status: Older adults age 50+ . 43 Table 3.6.1 Occupation: Women 15-49. 44 Table 3.6.2 Occupation: Men 15-49 . 45 Table 3.6.3 Occupation: Older adults age 50+. 46 Table 3.7 Type of employment: Women and men age 15-49. 47 Figure 3.1 Women’s Employment Status (past 12 months) . 41 Figure 3.2 Type of Earnings of Women Employed in the Past 12 Months . 48 CHAPTER 4 FERTILITY LEVELS, TRENDS, AND DIFFERENTIALS Table 4.1 Current fertility . 49 Table 4.2 Fertility by background characteristics . 51 Table 4.3 Trends in age-specific fertility rates. 51 Table 4.4 Trends in fertility. 52 Table 4.5 Children ever born and living. 53 Table 4.6 Birth intervals. 55 Table 4.7 Age at first birth . 56 Table 4.8 Median age at first birth . 57 Table 4.9 Teenage pregnancy and motherhood. 58 Figure 4.1 Total Fertility Rates for Selected Countries in Southeast Africa. 50 Figure 4.2 Trends in Fertility . 52 CHAPTER 5 FAMILY PLANNING Table 5.1 Knowledge of contraceptive methods . 60 Table 5.2 Knowledge of contraceptive methods by background characteristics . 61 Table 5.3.1 Ever use of contraception: Women . 62 Table 5.3.2 Ever use of contraception: Men . 64 Table 5.4 Current use of contraception by age . 65 Table 5.5 Current use of contraception by background characteristics . 67 Table 5.6 Trends in current use of contraception. 68 Table 5.7 Number of children at first use of contraception . 70 Table 5.8.1 Brand of pills used . 70 Table 5.8.2 Brand of condoms used . 70 Table 5.9 Mode of disposal of condoms . 71 Table 5.10 Knowledge of fertile period. 71 Table 5.11 Source of modern contraception methods . 72 Table 5.12 Cost of modern contraceptive methods. 73 Table 5.13 Informed choice . 74 Table 5.14 Future use of contraception . 75 Table 5.15 Reason for not intending to use contraception in the future . 75 Table 5.16 Preferred method of contraception for future use. 76 Table 5.17.1 Exposure to family planning messages: Women 15-49 . 77 Tables and Figures | xiii Table 5.17.2 Exposure to family planning messages: Men 15-49 . 78 Table 5.18 Contact of nonusers with family planning providers . 79 Table 5.19 Husband/partner's knowledge of women's use of contraception . 80 Table 5.20 Male attitudes about the use of contraception. 81 Figure 5.1 Ever Use of Contraception Among Women. 63 Figure 5.2 Contraceptive Prevalence of Modern Methods Among All Women in Selected Countries in Southeast Africa . 66 Figure 5.3 Contraceptive Prevalence Rates Among Currently Married Women Age 15-49. 68 Figure 5.4 Trends in Current Contraceptive Use Among Currently Married Women . 69 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY Table 6.1.1 Current marital status: Women and men 15-49 . 83 Table 6.1.2 Current marital status: Women and men age 50+ . 84 Table 6.2 Polygyny: Women and men age 15-49 . 85 Table 6.3 Number of co-wives: Women and men age 50+. 86 Table 6.4 Age at first marriage: Women and men age 15-49 . 87 Table 6.5 Age at first marriage: Women and men age 50+ . 87 Table 6.6 Median age at first marriage. 88 Table 6.7 Age at first sexual intercourse: Women and men age 15-49 . 89 Table 6.8 Age at first sexual intercourse: Women and men age 50+ . 90 Table 6.9 Median age at first intercourse . 90 Table 6.10.1 Recent sexual activity: Women age 15-49 . 92 Table 6.10.2 Recent sexual activity: Men age 15-49. 93 Table 6.10.3 Recent sexual activity: Women and men age 50+. 94 Table 6.11 Postpartum amenorrhoea, abstinence, and insusceptibility. 95 Table 6.12 Median duration of amenorrhoea, postpartum abstinence, and postpartum insusceptibility. 96 Table 6.13 Menopause. 96 CHAPTER 7 FERTILITY PREFERENCES Table 7.1 Fertility preferences by number of living children . 98 Table 7.2.1 Desire to limit childbearing: Women . 99 Table 7.2.2 Desire to limit childbearing: Men. 99 Table 7.3 Need and demand for family planning among currently married women . 100 Table 7.4 Ideal number of children . 102 Table 7.5 Mean ideal number of children. 103 Table 7.6 Fertility planning status. 104 Table 7.7 Wanted fertility rates. 105 Figure 7.1 Planning Status of Births, 1988 and 2006-07. 104 xiv | Tables and Figures CHAPTER 8 INFANT AND CHILD MORTALITY Table 8.1 Early childhood mortality rates . 109 Table 8.2 Early childhood mortality rates by socio-economic characteristics . 111 Table 8.3 Early childhood mortality rates by demographic characteristics. 112 Table 8.4 Perinatal mortality. 113 Table 8.5 High-risk fertility behaviour . 114 Figure 8.1 Infant Mortality Rates in Selected Sub-Saharan African Countries for 0-4 Years Preceding the Survey. 109 Figure 8.2 Neonatal, Postneonatal, Infant, Child, and Under-Five Mortality Rates for Five-Year Periods Preceding the Survey . 110 CHAPTER 9 MATERNAL HEALTH Table 9.1 Antenatal care provider. 116 Table 9.2 Number of antenatal care visits, timing of first visit, and source where ANC received . 117 Table 9.3 Components of antenatal care . 118 Table 9.4 Tetanus toxoid injections . 119 Table 9.5 Place of delivery . 121 Table 9.6 Assistance during delivery . 122 Table 9.7 Timing of first postnatal checkup. 124 Table 9.8 Type of provider of first postnatal checkup. 125 Table 9.9 Problems in accessing health care . 126 CHAPTER 10 CHILD HEALTH Table 10.1 Child's weight and size at birth. 128 Table 10.2 Vaccinations by source of information. 129 Table 10.3 Vaccinations by background characteristics . 131 Table 10.4 Prevalence and treatment of fever and symptoms of ARI. 132 Table 10.5 Prevalence of diarrhoea . 134 Table 10.6 Diarrhoea treatment . 135 Table 10.7 Feeding practices during diarrhoea . 137 Table 10.8 Knowledge of ORS packets or pre-packaged liquids. 138 Table 10.9 Disposal of children's stools. 139 Figure 10.1 Trend in Vaccination Coverage . 130 Figure 10.2 Treatment Practises for Children Ill with the Symptoms of an Acute Respiratory Infection or a Fever. 133 CHAPTER 11 NUTRITION OF CHILDREN AND ADULTS Table 11.1 Nutritional status of children . 142 Table 11.2 Initial breastfeeding. 146 Table 11.3 Breastfeeding status by age . 147 Tables and Figures | xv Table 11.4 Median duration and frequency of breastfeeding . 148 Table 11.5 Foods and liquids consumed by children in the day and night preceding the interview . 150 Table 11.6 Infant and young child feeding (IYCF) practices . 151 Table 11.7 Presence of iodized salt in household . 153 Table 11.8.1 Prevalence of anaemia in children 6-59 months. 154 Table 11.8.2 Prevalence of anaemia in children 5-11 years . 155 Table 11.8.3 Prevalence of anaemia in children 12-14 years . 156 Table 11.9 Micronutrient intake among children . 158 Table 11.10.1 Nutritional status of women . 160 Table 11.10.2 Nutritional status of men. 161 Table 11.11 Foods consumed by mothers in the day and night preceding the interview. 162 Table 11.12.1 Prevalence of anaemia in women age 15-49. 163 Table 11.12.2 Prevalence of anaemia in men age 15-49 . 164 Table 11.12.3 Prevalence of anaemia in women age 50 and older . 165 Table 11.12.4 Prevalence of anaemia in men age 50 and older . 165 Table 11.13 Micronutrient intake among mothers . 167 CHAPTER 12 MALARIA AND OTHR HEALTH ISSUES Table 12.1 Ownership of mosquito nets . 170 Table 12.2 Use of mosquito nets by children. 171 Table 12.3 Use of mosquito nets by pregnant women . 172 Table 12.4 Prophylactic use of antimalarial drugs and use of Intermittent Preventive Treatment (IPT) by women during pregnancy. 173 Table 12.5 Prevalence and prompt treatment of fever . 174 Table 12.6 Interior wall of dwelling sprayed against mosquitoes . 175 Table 12.7 Male circumcision. 176 Table 12.8 Desire to be circumcised . 177 Table 12.9.1 Health insurance coverage: Women . 178 Table 12.9.2 Health insurance coverage: Men. 179 Table 12.10.1 Knowledge and attitudes concerning tuberculosis: Women. 180 Table 12.10.2 Knowledge and attitudes concerning tuberculosis: Men . 181 Table 12.11 Use of tobacco. 182 CHAPTER 13 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOUR Table 13.1 Knowledge of AIDS. 184 Table 13.2 Knowledge of HIV prevention methods. 185 Table 13.3.1 Comprehensive knowledge about AIDS: Women . 187 Table 13.3.2 Comprehensive knowledge about AIDS: Men. 188 Table 13.4 Knowledge of prevention of mother-to-child transmission of HIV. 189 Table 13.5.1 Accepting attitudes towards those living with HIV/AIDS: Women. 190 Table 13.5.2 Accepting attitudes towards those living with HIV/AIDS: Men . 191 Table 13.6 Attitudes towards negotiating safer sexual relations with husband. 192 Table 13.7 Adult support of education about condom use to prevent AIDS . 193 xvi | Tables and Figures Table 13.8.1 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: Women . 194 Table 13.8.2 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: Men. 195 Table 13.9 Condom use characteristics among men . 196 Table 13.10 Knowledge of a source for male and female condoms. 197 Table 13.11.1 Coverage of prior HIV testing: Women . 198 Table 13.11.2 Coverage of prior HIV testing: Men. 199 Table 13.12 Time to get HIV test result. 200 Table 13.13 Pregnant women counselled and tested for HIV. 201 Table 13.14 Self-reported prevalence of sexually transmitted infections (STIs) a nd STI symptoms . 202 Table 13.15 Reason for not seeking treatment for STIs. 203 Table 13.16 Actions taken when had STI/STI symptoms . 203 Table 13.17 Prevalence of medical injections . 204 Table 13.18 Comprehensive knowledge about AIDS and knowledge of a source for condoms among youth . 206 Table 13.19 Age at first sexual intercourse among youth. 207 Table 13.20 Condom use at first sexual intercourse among youth. 209 Table 13.21 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth. 210 Table 13.22.1 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months: Women . 211 Table 13.22.2 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months: Men. 212 Table 13.23 Age mixing in sexual relationships among women age 15-19 . 213 Table 13.24 Drunkenness during sexual intercourse among youth. 214 Table 13.25 Recent HIV tests among youth . 215 Figure 13.1 Women and Men Seeking Advice or Treatment for STIs. 203 Figure 13.2 Type of Facility Where Last Medical Injection Was Received . 205 Figure 13.3 Abstinence, Being Faithful, and Condom Use (ABC) among Young Women and Men. 208 CHAPTER 14 HIV PREVALENCE AND ASSOCIATED FACTORS Table 14.1 Coverage of HIV testing among the population age 2 years and older by age. 218 Table 14.2 Coverage of HIV testing among population age 2 years and older by selected background characteristics. 220 Table 14.3 HIV prevalence among population age 2 years and older by age. 222 Table 14.4 HIV prevalence among the population age 2 years and older by residence, region, and wealth quintile. 223 Table 14.5 HIV prevalence by education and employment status: Women and men age 15 and older. 224 Tables and Figures | xvii Table 14.6.1 HIV prevalence by demographic characteristics: Women age 15 and older . 226 Table 14.6.2 HIV prevalence by demographic characteristics: Men age 15 and older . 227 Table 14.6.3 HIV prevalence by demographic characteristics: Women and men age 15 and older. 228 Table 14.7.1 HIV prevalence by sexual behaviour: Women age 15 and older . 229 Table 14.7.2 HIV prevalence by sexual behaviour: Men age 15 and older. 230 Table 14.7.3 HIV prevalence by sexual behaviour: Women and men age 15 and older . 232 Table 14.8 HIV prevalence by other characteristics: related to HIV risk: Women and men age 15 and older. 233 Table 14.9 Prior HIV testing by current HIV status . 234 Table 14.10 HIV prevalence by male circumcision . 235 Table 14.11 HIV prevalence by background characteristics: Young people age 15-24 . 236 Table 14.12 HIV prevalence by sexual behaviour: Young people age 15-24 . 237 Table 14.13 HIV prevalence among couples. 238 CHAPTER 15 ADULT AND MATERNAL MORTALITY Table 15.1 Completeness of information on siblings . 239 Table 15.2 Adult mortality rates. 240 Table 15.3 Maternal mortality . 242 Figure 15.1 Age-Specific Mortality Rates by Sex. 241 CHAPTER 16 WOMEN’S EMPOWERMENT AND DEMOGRAPHIC AND HEALTH OUTCOMES Table 16.1 Employment and cash earnings . 244 Table 16.2.1 Control over women's cash earnings and relative magnitude of women's earnings: Women. 245 Table 16.2.2 Control over men's cash earnings . 247 Table 16.3 Women's control over their own earnings and over those of their husbands . 248 Table 16.4 Women's participation in decisionmaking . 248 Table 16.5 Women's participation in decisionmaking by background characteristics. 249 Table 16.6.1 Attitude toward wife beating: Women . 251 Table 16.6.2 Attitude toward wife beating: Men. 252 Table 16.7.1 Attitude toward refusing sexual intercourse with husband: Women . 254 Table 16.7.2 Attitude toward refusing sexual intercourse with husband: Men. 255 Table 16.8 Indicators of women's empowerment. 257 Table 16.9 Current use of contraception by women's status. 258 Table 16.10 Women's empowerment and ideal number of children and unmet need for family planning . 259 Table 16.11 Reproductive health care by women's empowerment . 260 Table 16.12 Early childhood mortality rates by women's status . 261 xviii | Tables and Figures Figure 16.1 Number of Decisions in Which Married Women Participate . 250 CHAPTER 17 ORPHANS AND VULNERABLE CHILDREN Table 17.1 Children's living arrangements and orphanhood. 264 Table 17.2 Orphans and vulnerable children (OVC) . 265 Table 17.3 School attendance by survivorship of parents and by OVC status . 266 Table 17.4 Possession of basic material needs by orphans and vulnerable children. 267 Table 17.5 Orphan not living with siblings . 268 Table 17.6 Underweight orphans and vulnerable children. 269 Table 17.7 Sexual intercourse before age 15 by OVC status . 269 Table 17.8 Succession planning. 270 Table 17.9 Widows dispossessed of property. 271 Table 17.10 External support for very sick persons. 272 Table 17.11 External support for orphans and vulnerable children. 273 CHAPTER 18 YOUTH Table 18.1 Relationship of caregiver(s). 276 Table 18.2 Number of caregivers. 277 Table 18.3 Adult supervision going to and at school . 278 Table 18.4 Knowledge and attitudes about sex . 279 Table 18.5 Knowledge about HIV/AIDS. 281 Table 18.6 Knowledge about HIV/AIDS-related issues . 283 Table 18.7 Exposure to information on HIV/AIDS through mass media. 284 Table 18.8 Exposure to HIV/AIDS messages through other media . 285 Table 18.9 Places for information about HIV/AIDS . 286 Table 18.10 Opinion on information about HIV/AIDS . 287 Table 18.11 Knowledge of HIV/AIDS help line . 288 Table 18.12 Discussion about HIV/AIDS. 289 Table 18.13 Persons with whom youth would like to discuss HIV/AIDS . 290 Table 18.14 Knowledge of place to be tested for AIDS virus . 291 Figure 18.1 Attitudes about Dating and Decisions to Have Sex . 280 Figure 18.2 Knowledge of Various Modes to Avoid AIDS. 282 APPENDIX A SAMPL.E IMPLEMENTATION Table A.1 Sample implementation: Women 15-49 . 295 Table A.2 Sample implementation: Men 15-49 . 296 Table A.3 Sample implementation: Girls 12-14. 297 Table A.4 Sample implementation: Boys 12-14 . 298 Table A.5 Sample implementation: Women age 50+. 299 Table A.6 Sample implementation: Men age 50+ . 300 Tables and Figures | xix APPENDIX B ESTIMATES OF SAMPLING ERRORS Table B.1 List of selected variables for sampling errors . 304 Table B.2 Sampling errors for national sample . 306 Table B.3 Sampling errors for urban sample. 308 Table B.4 Sampling errors for rural sample. 310 Table B.5 Sampling errors for Hhohho sample . 312 Table B.6 Sampling errors for Manzini sample . 314 Table B.7 Sampling errors for Shiselweni sample. 316 Table B.8 Sampling errors for Lubombo sample . 318 Table B. 9 Sampling errors for persons age 12-14 (youth) and age 50 + (older adults), national sample . 320 APPENDIX C DATA QUALITY TABLES Table C.1 Household age distribution . 321 Table C.2.1 Age distribution of eligible and interviewed women . 322 Table C.2.2 Age distribution of eligible and interviewed men. 322 Table C.3 Completeness of reporting . 323 Table C.4 Births by calendar years . 323 Table C.5 Reporting of age at death in days . 324 Table C.6 Reporting of age at death in months. 324 APPENDIX D EFFECT OF NONRESPONSE ON THE SDHS HIV PREVALENCE RESULTS Table D.1 Observed and adjusted HIV prevalence among women and men . 325 Preface | xxi PREFACE This detailed report presents the major findings of the 2006-07 Swaziland Demographic and Health Survey (2006-07 SDHS). The 2006-07 SDHS is the first survey of its kind to be undertaken in Swaziland. It was a nationwide survey aimed at generating estimates at the country level, regional level, and for urban and rural areas. The survey was commissioned by the Ministry of Health and Social Wel- fare and implemented by the Central Statistical Office. Fieldwork was carried out between July 2006 and March 2007. The primary objective of the 2006-07 SDHS was to collect up-to-date information for policymak- ers, planners, researchers, and programme managers that would provide guidance in the planning, imple- mentation, monitoring and evaluation of population and health programmes in Swaziland. Specifically, the 2006-07 SDHS collected information on fertility levels, marriage, sexual activity, fertility preferences, awareness and use of family planning methods, breastfeeding practices, nutritional status of women and young children, childhood and maternal mortality, care and protection of youth, and awareness and be- haviour regarding HIV/AIDS and other sexually transmitted infections (STIs). In addition, it collected information on malaria, the use of mosquito nets, and the prevalence of HIV in the population age two years and above. I would like to acknowledge the efforts of a number of organizations that contributed immensely to the success of the survey. First, I would like to acknowledge the financial assistance from the Govern- ment of Swaziland, the United Nations Joint Programme on HIV/AIDS (UNAIDS), the National Emer- gency Response Council on HIV/AIDS (NERCHA), the United Nations Children’s Fund (UNICEF), the United Nations Population Fund (UNFPA), the U.S. Centers for Disease Control (CDC) and Prevention, which channeled its support through Population Services International (PSI), the Swaziland National AIDS Programme (SNAP), HIV/AIDS Prevention and Care (HAPAC), the World Health Organisation (WHO), and Italian Co-operation. Secondly, in the area of technical backstopping, I would like to ac- knowledge Macro International Inc. and the Global Clinical Viral laboratory. We owe an immense gratitude to the Survey Director, Mrs Rachel Masuku from the Central Sta- tistical Office, the Technical Survey Director, Mrs Nelisiwe Dlamini from the Ministry of Health and So- cial Welfare, the Field Coordinator, Mr Henry Ginindza from the Central Statistical Office, Mrs Thoko Nhlabatsi also from the Central Statistical Office, and the Data Processing staff, Interviewers, Supervi- sors, Field Editors, Laboratory Technicians and Drivers for their hard work and dedication. We are also grateful to all of the respondents for their patience and generosity with their time. Isabella Hlophe Director of Statistics Summary of Findings | xxiii SUMMARY OF FINDINGS The 2006-07 Swaziland Demographic and Health Survey (SDHS) is a nationally represe- ntative survey of 4,843 households, 4,987 women age 15-49, and 4,156 men age 15-49. The SDHS also included individual interviews with boys and girls age 12-14 and older adults age 50 and over. The survey of persons age 12-14 and age 50 and over was carried out in every other household selected in the SDHS. Interviews were completed for 459 girls and 411 boys age 12-14, and 661 women and 456 men age 50 and over. The 2006-07 SDHS is the first national sur- vey conducted in Swaziland as part of the De- mographic and Health Surveys (DHS) pro- gramme. The data are intended to furnish pro- gramme managers and policymakers with de- tailed information on levels and trends in fertil- ity; nuptiality; sexual activity; fertility prefer- ences; awareness and use of family planning methods; breastfeeding practices; nutritional status of mothers and young children; early childhood mortality and maternal mortality; ma- ternal and child health; and awareness and be- haviour regarding HIV/AIDS and other sexually transmitted infections. The survey also collected information on malaria prevention and treat- ment. The 2006-07 SDHS is the first nationwide survey in Swaziland to provide population-based prevalence estimates for anaemia and HIV. Children age 6 months and older as well as adults were tested for anaemia. Children age 2 years and older as well as adults were tested for HIV. FERTILITY Fertility in Swaziland has been declining ra- pidly, with the TFR falling from 6.4 births per woman in 1986 to 3.8 births at the time of the SDHS. As expected, fertility is higher in rural areas (4.2 births per woman) than in urban areas (3.0 births per woman). Fertility differentials by education and wealth are substantial. Women with no education have on average 4.9 children compared with 2.4 children for women with tertiary education. Fertility varies widely according to household wealth. Women in the highest wealth quintile have 2.9 chil- dren fewer than women in the lowest quintile (2.6 and 5.5 births per woman, respectively). Unplanned pregnancies remain common in Swa- ziland, despite the falling fertility. Overall, 37 percent of births are unwanted, while 27 percent are mistimed (wanted later). If all unwanted births were prevented, women would have an average of 2.1 children com- pared with the actual average of 3.8 children. While marriage and cohabitation are generally considered to be primary indicators of regular expo- sure to the risk of pregnancy, many women in Swazi- land bear children before entering a stable union. Marriage occurs comparatively late in Swaziland. Only 23 percent of women and 7 percent of men mar- ry before age 20. Around one-quarter of women and one-third of men age 30-34 have not yet married. Initiating sexual activity before marriage is common in Swaziland. Half of women age 20-49 had first sexual intercourse by age 18, and more than 70 percent were sexually active by age 20. While men generally initiate sexual activity at a later age than women, 31 percent of men age 20-49 had first sexual intercourse before age 18, and 57 percent were sexu- ally active by age 20. The 2006-07 SDHS shows that 18 percent of cur- rently married women are in a polygynous union, i.e., their husband has more than one wife. Older women are more likely to be in a polygynous union than younger women. Polygyny is about twice as preva- lent in rural areas as in urban areas. Regional varia- tion is substantial, with Lubombo having the highest proportion of women in polygynous marriages (23 percent) and Manzini having the lowest proportion (15 percent). xxiv | Summary of Findings FAMILY PLANNING Knowledge of family planning is universal in Swaziland. The most widely known method is the male condom (99 percent for both males and females). Among women, other widely known methods include injectables (96 percent), the pill (95 percent), and the female condom (91 per- cent). For men, the best known methods besides the male condom are the female condom (94 percent) and the pill and injectables (84 percent each). More than half (51 percent) of currently married women in Swaziland are using a method of contraception ; most of them use a modern method (48 percent). Contraceptive use among sexually active unmarried women (65 percent) is higher than that among married women, primar- ily because of the greater use of the male con- dom. Government-sponsored facilities remain the chief providers of contraceptive methods in Swaziland (45 percent), while 14 percent are supplied through private medical sources, 9 per- cent through missions, 9 percent through non- government organisations (NGOs), and 18 per- cent through other private sources (e.g., shops). The most common single source of contracep- tive methods in Swaziland is PHU/clinics, which supply about one-quarter of users of modern methods (25 percent). Shops supply 15 percent of users, followed by government hospitals (9 percent). Unmet need for family planning among cur- rently married women is 24 percent. If all mar- ried women with an unmet need for family plan- ning were to use a contraceptive method, the contraceptive prevalence rate in Swaziland would increase from the current level of 51 per- cent to 75 percent. CHILD HEALTH Children are considered fully vaccinated when they receive one dose of BCG vaccine, three doses each of DPT and polio vaccines, and one dose of measles vaccine. BCG coverage among children age 12-23 months is nearly uni- versal (97 percent); coverage is also high for the first doses of DPT (96 percent) and polio (97 percent). The proportion of children receiving subsequent do- ses of DPT and polio vaccines drops slightly, with 92 percent of children receiving the third dose of DPT and 87 percent receiving the third dose of polio. Ninety-two percent of children had received a mea- sles vaccination by the time of the SDHS. Overall, 82 percent of children age 12-23 months are fully im- munised. In the two weeks prior to SDHS, 8 percent of children under age five experienced symptoms of ARI, and 28 percent had a fever. Diarrhoea was a more prevalent problem among young children than fever; 13 percent of children under age five had diar- rhoea in the two weeks preceding the survey. More than 70 percent of children with diarrhoea were taken to a health provider. Nine in ten children with diar- rhoea were treated with some type of oral rehydration therapy (ORT), and 26 percent were given increased fluids. Only 6 percent of children with diarrhoea did not receive any treatment at all. Data from the 2006-07 SDHS indicate that for the most recent five-year period preceding the sur- vey, the under-five mortality rate was 120 deaths per 1,000 live births. This means that one in every seven children born in Swaziland dies before reaching the fifth birthday. The infant mortality rate is 85 per 1,000. One-quarter of all infant deaths take place in the neonatal period, that is, during the first month of life. MATERNAL HEALTH In Swaziland, almost all women who had a live birth in the five years preceding the survey received antenatal care from health professionals (97 percent); 9 percent received care from a doctor, and 88 percent received care from a trained nurse or midwife. Only 3 percent of mothers did not receive any antenatal care. Tetanus toxoid injections are given during preg- nancy to prevent neonatal tetanus. Seventy-five per- cent of last-born children born during the five years preceding the SDHS were fully protected against neonatal tetanus, either because the mother had had at least two tetanus toxoid injections during that preg- nancy or because she had had the number of doses required for lifetime protection. Summary of Findings | xxv The majority of births in the five years be- fore the survey were delivered in a health facil- ity (74 percent). The births took place more of- ten in public health facilities (43 percent) than in Mission or other private health facilities. Twelve percent of births were assisted by a doctor, 62 percent by a nurse or nursing assistant, and 5 percent by a traditional birth attendant. Eight percent of births were delivered by caesarean section. Twenty-five percent of mothers received a postnatal checkup for the most recent birth in the five years preceding the survey, with 22 percent having the checkup within the critical 48 hours after delivery. BREASTFEEDING AND NUTRITION Overall, 87 percent of children in Swaziland are breastfed for some period of time (ever breastfed). The median duration of any breast- feeding in Swaziland is almost 17 months. How- ever, the median duration of exclusive breast- feeding is much shorter (0.7 months). Appropriate infant and young child feeding (IYCF) practices include increasing the amount and variety of foods a child consumes as it gets older, while maintaining frequent breastfeeding. Seven in ten children age 6-23 months in Swazi- land were fed according to the recommended minimum standards with respect to food diver- sity. Among breastfed children age 6-23 months, about three-quarters were fed according to the minimum standards (consumed foods from 3 or more food groups), while, among non-breastfed children age 6-23 months, only 60 percent were fed according to the minimum standards (con- sumed foods from 4 or more food groups). Overall, 42 percent of children age 6-59 months have some degree of anaemia. About one in five children are mildly anaemic, 19 per- cent have moderate anaemia and less than 1 per- cent have severe anaemia. The proportion of children age 5-11 years with some degree of anaemia is 18 percent; less than 1 percent of these children are severely anaemic. The na- tional prevalence of anaemia among children age 12-14 years is virtually identical to that among children age 5-11 years, and a majority of these children are mildly anaemic (17 percent). Thirty percent of women age 15-49 have some degree of anaemia, with the majority classified as mildly anaemic (23 percent). Less than 1 percent of these women are considered severely anaemic. Preg- nant women are more likely to be anaemic (40 per- cent) than breastfeeding women (29 percent) or wo- men who are neither pregnant nor breastfeeding (30 percent). This could be a result of the high demand of iron and folate during pregnancy. Men age 15-49 are substantially less likely to be anaemic than women the same age (13 percent and 30 percent, respec- tively), and less than 1 percent of these men are con- sidered severely anaemic. Women and men age 50 and over show a pattern that is the reverse of that seen for women and men age 15-49, with men age 50 and over substantially more likely to be anaemic than their female counterparts (31 percent and 21 percent, respectively). At the time of the survey, 29 percent of children under age five were stunted (short for their age), 3 percent were wasted (thin for their height), and 5 per- cent were underweight (thin for their age). Nation- ally, only 2 percent of children are overweight for their age. Malnutrition rates are generally highest during the period when children are being weaned. Nearly half of children age 18-23 months are stunted, and 19 percent are severely stunted. Overall, 46 percent of women and 72 percent of men have a body mass index (BMI) in the normal range. Comparatively few women are malnourished; only 3 percent of women are thin, and 1 percent are severely thin. Malnutrition is higher among men, with 10 percent of men assessed as too thin, and 3 percent considered moderately or severely thin. At the other end of the BMI range, 14 percent of men are assessed as overweight (BMI 25-29.9) and 4 per- cent are obese (BMI >30). Among women, 28 per- cent are classified as overweight, and 23 percent are considered obese. MALARIA In interpreting the malaria programme indicators in Swaziland, it is important to recognise that the dis- ease affects an estimated 30 percent of the population where malaria is most prevalent (the Lubombo Pla- teau, the lowveld, and parts of the middleveld). Ma- xxvi | Summary of Findings laria is also seasonal, occurring mainly during or after the rainy season (from November to March). A substantial part of the SDHS field- work took place outside of this period. Overall, 6 percent of households in Swazi- land have at least one mosquito net (treated or untreated), and 4 percent have at least one insec- ticide-treated net. Usage of bednets is relatively low among young children and pregnant women, groups that are considered particularly vulner- able to the effects of the disease. On the night before the survey, less than 1 percent of children under age five and less than 1 percent of pregnant women slept under an ever-treated net. Prophylactic use of antimalarial drugs is not common in Swaziland. Only 7 percent of women with a live birth in the two years preced- ing the survey reported taking antimalarial drugs for prevention. Less than 1 percent of children under age five with fever were given an antima- larial drug. Indoor residual spraying is another compo- nent of efforts to control malaria transmission in Swaziland. Twelve percent of households re- ported that the interior walls of their dwelling had been sprayed, principally as part of a gov- ernment programme. The prevalence of indoor spraying was highest in Lubombo (46 percent), where malaria is most prevalent. HIV/AIDS-RELATED KNOWLEDGE AND BEHAVIOUR Knowledge of HIV and AIDS is universal in Swaziland. All women and 99 percent of men age 15-49 have heard of AIDS. Among those age 50 and over, 96 percent of women and 97 percent of men have heard about AIDS. At the same time, however, only half of women (52 percent) and men (51 percent) have what can be considered comprehensive knowl- edge about the modes of HIV transmission and prevention. Comprehensive knowledge means knowing that consistent use of condoms and having just one uninfected, faithful partner can reduce the chances of getting HIV, knowing that a healthy-looking person can have HIV, and re- jecting the two most common local misconceptions about HIV transmission and prevention. Comprehen- sive knowledge is lower among those age 50 and over (21 percent for women and 25 percent for men). A high proportion of women and men age 15-49 know that HIV can be transmitted by breastfeeding and that the risk of mother-to-child transmission (MTCT) can be reduced by taking special drugs dur- ing pregnancy. Both aspects of MTCT are known to 76 percent of women and over 64 percent of men age 15-49. The level of awareness is somewhat lower among women and men age 50 and over. Given that most HIV infections in Swaziland are contracted through heterosexual contact, information on the proportion of women and men who have mul- tiple partners or engage in higher-risk sex (i.e., sexual intercourse with a non-marital, non-cohabiting part- ner) is important for planning prevention pro- grammes. The 2006-07 SDHS results indicate that 2 percent of women and 23 percent of men age 15-49 had two or more partners during the 12 months pre- ceding the survey. For older adults, the proportions are about half (1 percent of women and 10 percent of men). Sexual intercourse with a non-marital non- cohabiting partner is more common than sexual inter- course with multiple partners; 44 percent of women and 58 percent of men age 15-49 who had sex in the 12 months preceding the survey reported having had sex with a non-marital, non-cohabiting partner. Such behaviour is less common among those age 50 and over; 12 percent of women and 10 percent of men in this age group who had sex in the 12 months preced- ing the survey reported having had sex with a non- marital, non-cohabiting partner. Among respondents age 15-49 who engaged in higher-risk sexual inter- course, 55 percent of women and 68 percent of men reported using a condom at the last higher-risk sexual intercourse. The 2006-07 SDHS also obtained information on the coverage of HIV testing. Among adults age 15- 49, 36 percent of women and 17 percent of men have been tested for HIV at some time, and received the results of the test. Twenty-two percent of women and 9 percent of men received their results within the 12 months preceding the survey. Summary of Findings | xxvii HIV PREVALENCE Results from the HIV testing component in the 2006-07 SDHS indicate that 26 percent of Swazi adults age 15-49 are infected with HIV. Among women, the HIV rate is 31 percent, com- pared with 20 percent among men. HIV preva- lence peaks at 49 percent for women age 25-29, which is almost five times the rate among wo- men age 15-19 and more than twice the rate ob- served among women age 45-49. HIV preva- lence increases from 2 percent among men in the 15-19 age group to 45 percent in the age group 35-39 and then decreases to 28 percent among men age 45-49. HIV prevalence for women and men age 50 or over is 12 percent and 18 percent, respectively. Among the population age 2-14 years, 4 percent of girls and boys are infected. HIV prevalence is higher in urban than in rural areas (31 percent and 24 percent, respec- tively, for women and men age 15-49). By re- gion, Hhohho has the highest prevalence rate (29 percent), followed by Lubombo (26 percent) and Manzini (25 percent), while Shiselweni (23 per- cent) has the lowest HIV prevalence rate. More than 700 cohabiting couples were tes- ted for HIV in the 2006-07 SDHS. Results indi- cated that for 55 percent of these couples, both partners tested negative for HIV; 29 percent of couples, both partners tested positive for HIV; and in 16 percent of couples, the results were discordant, that is, one partner was infected and the other was not. In 8 percent of couples, the male partner was infected and the woman was not, while in another 9 percent of couples, the woman was infected and the man was not. ORPHANS AND VULNERABLE CHILDREN Seventy-eight percent of children under age 18 in the households sampled for the SDHS were not living with both parents, and 34 percent were not living with either parent. Twenty-three percent of children under age 18 were orphaned, that is, one or both parents were dead. The per- centage increases rapidly with age, from 7 per- cent among children under age five to 37 percent among children age 15-17 years. Overall, 12 percent of children under age 18 were consid- ered vulnerable, i.e., they lived in a household in which at least one adult had been chronically ill dur- ing the year preceding the survey or at least one par- ent living in the household or elsewhere had suffered from a chronic illness. Three in ten children in Swa- ziland are considered orphaned or vulnerable. YOUTH Interviews with children age 12-14 were de- signed to obtain information about risk factors asso- ciated with HIV infection. Respondents were asked questions about home care and protection, media ex- posure, knowledge and attitudes about sex, and knowledge of AIDS. One important topic was identifying the person who cares for the child at home. Grandmothers play an important role in taking care of children, even when both parents are still alive (21 percent). Their role is even greater when the child is orphaned (37 percent). Children benefit from knowing about the physi- ology of human reproduction and the ways a person can protect against sexual or reproductive diseases and problems. Six in ten youth (61 percent) said that they know the meaning of having sex. Knowledge is higher among girls and urban residents. Sex and sex- ual abuse are topics of discussion between some youth and their parents or guardians (37 percent and 43 percent, respectively). Girls are much more likely to have talked about sex with a parent or other care- giver than boys (47 percent compared with 25 per- cent). Almost all children age 12-14 have heard of HIV/AIDS (97 percent) and there are no major varia- tions by background characteristics. Overall, 64 per- cent of children mentioned abstaining from sex as a way of reducing the chances of getting AIDS, 47 per- cent mentioned the use of condoms, and 19 percent mentioned avoiding blood transfusions. One in six children (15 percent) said that the chances of con- tracting AIDS can be reduced or avoided by being faithful to one sexual partner. Half of children age 12-14 in the survey know of a place for AIDS testing. Girls and children living in urban areas are more likely than other children to know where to go for the AIDS testing. xxviii | Map of Swaziland Introduction | 1 INTRODUCTION 1 Henry Ginindza and Rachel Masuku 1.1 HISTORY, GEOGRAPHY, AND ECONOMY History Swaziland became independent of British colonial rule in September 1968. It is one of the few countries in the world operating under monarchy rule. The King is the Head of State and his mother, the Indlovukazi or Queen Mother, is the mother of the nation. Swaziland’s first constitution in over 30 years was ratified by King Mswati III in July 2005 and it became effective in February 2006. The majority of the population is ethnic Swazi, mixed with a small number of Zulus and non- Africans. Geography The Kingdom of Swaziland is the smallest landlocked country in Southern Africa measuring approximately 17,000 km2. The country enjoys a tropical to near-temperate climate along the western highlands, which rises to an altitude of over 1,800 metres above sea level, while the lowveld areas are generally hot. Swaziland lies in a summer rainfall region. Economy Although manufacturing contributes a growing share to Swaziland’s GDP, the economy is largely agricultural because most industries process agricultural produce. These include sugar processing, wood pulp production, food canning and so on. Other agriculture products include corn, citrus fruits, livestock, and pineapple, among others. The performance of the Swazi economy has been stagnant over the last five years, averaging an annual growth rate of around two percent. This has been largely due to fluctuations in the performance of the agricultural sector brought about by changes in climatic conditions as well as changes in prices in the world market. Persistent drought and disease have affected production, resulting in failure to meet export quota requirements 1.2 POPULATION The population census is the major source of historical demographic data. The first detailed population census was conducted in 1966 and since then, censuses have been conducted every ten years, i.e., 1986, 1997, and 2007. Table 1.1 shows that in 1976 the population of Swaziland was about half a million. Two decades later in 1997, the population had almost doubled. The high growth rate of the population is brought about by high fertility and declining mortality levels. According to the 1997 Population and Housing Census, life expectancy at birth is 60 years. 2 | Introduction The population per square kilometre almost doubled over the 20-year period, from 29 persons in 1976 to 54 persons in 1997. The proportion of urban residents increased significantly from 15 percent in 1976 to 23 percent in 1997. 1.3 POPULATION, FAMILY PLANNING, AND HIV POLICIES AND PROGRAMMES The National Population Policy was adopted in 2003 to focus on the implementation of the social development component of the National Development Strategy. The overall objective of the policy is to improve the quality of life by influencing demographic trends and responding to emerging challenges such as HIV/AIDS. The specific objectives of the population policy are to: • Improve the health and welfare status of the population • Curb the further spread of HIV/AIDS • Reduce the social and economic impact of HIV/AIDS • Reduce the level of fertility • Control the influx of illegal immigrants • Promote gender equality and equity at all levels and spheres of society • Promote the use of natural resources • Contribute towards increasing food security at the household and national levels • Reduce rural-urban migration • Improve the availability, quality, and timeliness of population-related data and research for use in policy and planning • Contribute towards reducing the levels of unemployment and poverty • Contribute towards the improvement of the accessibility, quality, and relevance of high school and tertiary education • Improve knowledge of the role that cultural beliefs and practices play in population-related issues • Promote the creation of a legal framework that is responsive to population concerns • Improve the availability, quality, and accessibility of population-related IEC • Promote the integration of population concerns in development planning at all levels • Contribute towards addressing the special needs of children, youth, the elderly, and persons with disabilities • Integrate population and family life education into the school curricula • Improve the quality of housing and related services The National Policy on HIV/AIDS As the HIV/AIDS epidemic affects all sectors, its control demands a well coordinated response. It is necessary to have policies that provide a framework, direction, and general principles for the national response, including prevention, care, and support to those infected and affected by the epidemic, and mitigation of its impact. Table 1.1 Basic demographic indicators Selected demographic indicators for Swaziland, 1976, 1986, and 1997 Population and Housing Censuses Indicator 1976 PHC 1986 PHC 1997 PHC Population 494,534 681,059 929,718 Intercensal growth rate (percent) 2.5 3.3 2.9 Density (pop./km2) 29 39 54 Percent urban 15.2 22.8 23.1 Crude birth rate 51.6 48.3 36.39 Crude death rate 18.5 13.0 7.6 Total fertility rate 5.2 6.4 4.5 Infant mortality rate u 99 78 Male 205 u u Female 180 u u Life expectancy at birth (years) 46 56 60 u = No information Source: CSO. nd. 1976 Population and Housing Census, Vol. 1; CSO. nd. 1986 Population and Housing Census, Vol. 4; CSO. nd. 1997 Population and Housing Census Vol. 4. Introduction | 3 The National Multisectoral HIV and AIDS Policy was adopted in July 2006 with the goal of providing the framework, direction and general principles for interventions. The policy thus strengthens and expands efforts to manage and co-ordinate the response, promotes prevention interventions, provides effective treatment, care and support to people living with HIV/AIDS and mitigates the impact of the epidemic. The National Emergency Response Council on HIV and AIDS (NERCHA) is guided by the ‘three ones principles’: one coordinating body, one strategic plan, and one monitoring and evaluation framework. 1.4 OBJECTIVES Main Objective The principal objective of the 2006-07 Swaziland Demographic and Health Survey (SDHS) was to provide up-to-date information on fertility, childhood mortality, marriage, fertility preferences, awareness, and use of family planning methods, infant feeding practices, maternal and child health, maternal mortality, HIV/AIDS-related knowledge and behaviour and prevalence of HIV and anaemia. More specifically the 2006-07 SDHS was aimed at achieving the following; • Determine key demographic rates, particularly fertility, under-five mortality, and adult mor- tality rates • Investigate the direct and indirect factors which determine the level and trends of fertility • Measure the level of contraceptive knowledge and practice of women and men by method • Determine immunization coverage and prevalence and treatment of diarrhoea and acute res- piratory diseases among children under five • Determine infant and young child feeding practices and assess the nutritional status of chil- dren 6–59 months, women age 15–49 years, and men aged 15-49 years • Estimate prevalence of anaemia • Assess knowledge and attitudes of women and men regarding sexually transmitted infections and HIV/AIDS, and evaluate patterns of recent behaviour regarding condom use • Identify behaviours that protect or predispose the population to HIV infection • Examine social, economic, and cultural determinants of HIV • Determine the proportion of households with orphans and vulnerable children (OVCs) • Determine the proportion of households with sick people taken care at household level • Determine HIV prevalence among males and females age 2 years and older • Determine the use of iodized salt in households • Describe care and protection of children age 12-14 years, and their knowledge and attitudes about sex and HIV/AIDS. This information is intended to provide data to assist policymakers and programme implementers to monitor and evaluate existing programmes and to design new strategies for demographic, social and health policies in Swaziland. The survey also provides data to monitor the country’s achievement towards the Millenium Development Goals. 1.5 ORGNISATION OF THE SURVEY The 2006-07 Swaziland Demographic and Health Survey (SDHS) is a national-level sample survey designed to provide information on various demographic and maternal and child health issues in Swaziland. The SDHS was implemented by the Central Statistical Office (CSO) at the request of the Ministry of Health and Social Welfare (MOHSW). The majority of the local costs of the survey were provided by the Government of Swaziland. Macro International Inc. (Macro) provided technical assistance 4 | Introduction to the SDHS as part of the worldwide USAID-funded MEASURE Demographic and Health Surveys (DHS) programme. The Human Sciences Research Council (HSRC) of South Africa assisted during the design phase of the survey. Through a subcontract with Macro, the Global Clinical and Viral Laboratory (GCVL) of South Africa provided support for the training and laboratory processing for the HIV testing component of the survey. Funds to support Macro’s and GCVL’s assistance and to defray some local costs were provided by USAID and the Centres for Disease Control and Prevention (CDC)-Global AIDS Programme operating under the President’s Emergency Plan for AIDS Relief (PEPFAR). Other organisations supporting SDHS included the National Emergency Response Council on HIV/AIDS (NERCHA), HIV/AIDS Prevention and Care (HAPAC), UNFPA, UNICEF, Italian Cooperation, the World Health Organisation, UNAIDS, and the Population Services International (PSI). 1.6 SAMPLE DESIGN The 2006-07 SDHS was designed to provide estimates of health and demographic indicators at the national level, for urban-rural areas, and for the four regions of Manzini, Hhohho, Lubombo, and Shiselweni. Standard DHS sampling policy recommends a minimum of 1,000 to 1,200 women per major domain. To meet this criterion, the number of households selected in each of the various domains, particularly urban areas, was not proportional to the actual size of the population in the domain. As a result, the SDHS sample is not self-weighting at the national level, and weights must be applied to the data to obtain the national-level estimates. The 2006-07 SDHS sample points (clusters) were selected from a list of enumeration areas (EAs) defined in the 1997 Swaziland Population and Housing Census. A total of 275 clusters were drawn from the census sample frame, 111 in the urban areas and 164 in the rural areas. CSO staff conducted an exhaustive listing of households in each of the SDHS clusters in August and September 2005. From these lists, a systematic sample of households was drawn for a total of 5,500 households. All women and men age 15-49 identified in these households were eligible for individual interview. In addition, a sub-sample of half of these households (2,750 households) was selected randomly in which all boys and girls age 12-14 and persons age 50 and older were eligible for individual interview. In the SDHS households where youth and older adults were interviewed, all individuals age 6 months and older were eligible for anaemia testing and all individuals age 2 and older were eligible for HIV testing. In the SDHS households where only women and men age 15-49 were interviewed, children age 6 months to 5 years were eligible for the anaemia testing and women and men age 15-49 were eligible for anaemia and HIV testing. During the household listing, field staff used Global Positioning System (GPS) receivers to establish and record the geographic coordinates of each of the SDHS clusters. 1.7 QUESTIONNAIRES Five types of questionnaires were used for the SDHS: the Household Questionnaire, the Woman’s Questionnaire, the Man’s Questionnaire, the Youth Questionnaire, and the Older Adult Questionnaire. The contents of the questionnaires were based on questionnaires developed for the MEASURE DHS programme. The Youth Questionnaire was adapted from the 2002 Nelson Mandela/HSRC Study of HIV/AIDS in South Africa. The SDHS questionnaires were developed in collaboration with a wide range of stakeholders. After the SDHS survey instruments were drafted, they were translated into and printed in the local language, Siswati, for pretesting. The Household Questionnaire was used to list all the usual members and visitors in the selected households. Basic information was collected on the characteristics of each person listed, including age, sex, education, and relationship to the head of the household. The Household Questionnaire was also used Introduction | 5 to identify persons eligible for the individual interview. In addition, information was collected about the dwelling, such as the source of water; type of toilet facilities; materials used to construct the house; ownership of various consumer goods; use of bed nets; and care and free external support received by chronically ill household members and orphans and vulnerable children. The results of anthropometric measurement and anaemia testing were recorded in the Household Questionnaire, as was the information on the consent of eligible household members for the HIV testing. The Woman’s Questionnaire was used to collect information from all women age 15-49 and covered the following topics: • Background characteristics (age, education, religion, etc.) • Birth history • Knowledge and use of family planning methods • Antenatal and delivery care • Infant feeding practices including patterns of breastfeeding • Vaccinations • Childhood illnesses and treatment • Marriage and sexual activity • Fertility preferences • Husband’s background and woman’s work status • Adult (maternal) mortality • HIV/AIDS-related knowledge, attitudes, and behaviour. The Man’s Questionnaire was shorter than the Woman’s Questionnaire, but covered many of the same topics, excluding the reproductive history and sections dealing with maternal and child health. The Older Adult Questionnaire obtained limited information on the background characteristics of the popu- lation age 50 and over and on HIV/AIDS knowledge, attitudes, and risk behaviour. The Youth Question- naire included questions on knowledge and attitudes about sex, and factors exposing youth to risk of abuse. 1.8 ANAEMIA AND HIV TESTING Haemoglobin testing is the primary method of anaemia diagnosis. In the SDHS, haemoglobin measurement was performed in the field by non-medical personnel. Prior to collecting the blood specimen, all participants age 12 and older and were asked to give informed consent to the testing. Prior to asking the consent of unmarried youth age 12-17, consent was obtained from the parent or other adult responsible for the child at the time of the survey. For children age 6 months-11 years, consent was asked only from the parent or guardian. The consent statement explained the purpose of the test, informed prospective subjects tested and/or their caretakers how the test would be done, advised them that the results would be available as soon as the test was completed, and requested permission for the test to be carried out. 1.9 PRETEST, TRAINING, AND FIELDWORK Pretest Two pretests were conducted for the 2006-07 SDHS. The first was aimed at testing the flow of the questions and the translation from English to Siswati. Given the fact that this was the first SDHS to be conducted in the country, this pretest was also viewed as a pilot exercise for the survey organising committee. The first pretest was conducted in August-September 2005. Pretest activities started with a 6 | Introduction training of trainers. The trainers were drawn from the CSO, the MOHSW, NERCHA, and the Ministry of Agriculture. Macro staff assisted with the training of trainers and Macro and HSRC staff assisted with the pretest training. Eight women and 16 men participated in the field staff training. All but five of the participants had worked in the SDHS as household listers. The SDHS trainers and several guest lecturers gave talks to introduce specific topics in the survey, such as sexual and reproductive health, water and sanitation, malaria, nutrition, and HIV/AIDS. The pretest was conducted in both urban and rural areas to help gauge how respondents’ reception of the SDHS teams might vary in different localities. On average, the Household Questionnaire took one hour to complete, the Woman’s Questionnaire took two hours, the Man’s Questionnaire took one hour, the Youth Questionnaire took 20 minutes, and the Older Adult Questionnaire took 30 minutes. The second pretest was carried out in April-May 2006 after the review of the HIV testing protocol was completed at CDC Atlanta. This pretest combined interviews and collection of blood samples for anaemia and HIV tests. Training A total of 83 persons, 38 males and 45 females, were trained to be the 2006-07 SDHS fieldworkers. They were grouped in two classes. Many of the trainees had participated in both the first and second pretest. The training followed the standard DHS training procedures, including instructions on how to conduct interviews and how to fill in all five questionnaires, classroom demonstration and practice in administering the questionnaires, and tests. The participants also had a chance to practice interviewing in actual households and discuss their experience before the fieldwork began. With respect to the biomarker data collection, the staff responsible for the anaemia and HIV testing received extensive classroom training plus additional field practice. As part of the training, they were given thorough training in informed consent procedures, how to take finger stick blood spot samples, and how to handle and package the dried blood spots. All staff received training in universal precautions and the disposal of hazardous waste. During the training, there were special lectures on the HIV/AIDS epidemic. Fieldwork Fieldwork for the 2006-07 SDHS was carried out by 10 mobile interviewing teams, each consisting of one supervisor, one field editor, three to four female interviewers, and one or two male interviewers. Two or three of the interviewers on each team were assigned to take the blood samples for the anaemia and HIV testing. Fieldwork commenced in July 2006 and was completed in February 2007. 1.10 HIV TESTING The SDHS HIV testing protocol involved the collection of at least three blood spots from a finger prick (generally the same prick used to obtain the blood drop for anaemia testing) on a special filter paper card. The HIV testing in the SDHS was anonymous, i.e., it was conducted in such fashion that the results could not be linked to individual respondents. A unique random identification number (bar-code) was assigned to each eligible respondent consenting to the testing, and labels containing that number were affixed to the filter paper card, the questionnaire, and a field tracking form at the time of the collection of the sample. No other identifiers were linked to the dried blood spot (DBS) samples from SDHS respondents during the HIV testing. Introduction | 7 Because of the anonymous nature of the testing approach in the SDHS, it was not possible to provide information on the results from the HIV testing conducted during the SDHS. In lieu of providing the SDHS test results, written and verbal information was provided on counselling and testing (VCT) sites where free confidential counselling and HIV testing were available during the survey. In addition, any person (whether or not they participated in the SDHS) approaching an SDHS team with a request about VCT was provided with information on the sites, in an effort to increase VCT usage in Swaziland. 1.11 DATA PROCESSING All questionnaires for the SDHS were returned to CSO central office for data processing. The processing operation consisted of office editing, coding of open-ended questions, data entry, double-entry verification, and resolving inconsistencies found by computer programmes developed for the SDHS. The SDHS data entry and editing programmes used CSPro, a computer software package specifically designed for processing survey data such as that produced by DHS surveys. Data processing commenced in August 2006 and was completed in April 2007. The HIV testing was carried out at the NRL between August 2006 and June 2007. 1.12 RESPONSE RATES Table 1.2 shows the response rates for the SDHS 2006-07. The response rates are important because they may affect the reliability of the results. Of a total of 5,500 households selected in the sample, 5,086 were occupied at the time of the fieldwork. This difference between the number of selected households and the number of occupied households is due to structures being vacated or destroyed. Suc- cessful interviews were conducted in 4,843 house- holds, yielding a response rate of 95 percent. In the households interviewed in the survey, a total of 5,301 eligible women age 15-49 were identified. Interviews were completed with 4,987 of these women, yielding a 94 percent response rate. In the same households, a total of 4,675 eligible men age 15-49 were identified and interviews were com- pleted with 4,156 of these men, yielding a male response rate of 89 percent. The response rates are slightly lower in the urban sample than in the rural sample, and lower among men than women. The principal reasons for non-response among both eligible men and women were refusal and the failure to find individuals at home despite repeated visits to the households. Men have lower response rates than women due to higher refusal rates, and more frequent and longer absence from the households, principally due to employment and their lifestyle (see Appendix A). Table 1.3 shows the results of the household and individual interviews in households selected for youth and older adults. A total of 2,750 households were selected in the sample, of which 2,543 were occupied at the time of the fieldwork. This difference between the number of selected households and the number of occupied households is due to structures being vacated or destroyed. Successful interviews were conducted in 2,410 households, yielding a response rate of 95 percent. Table 1.2 Results of the household and individual interviews Number of households, number of interviews, and response rates, according to residence (unweighted), Swaziland 2006-07 Residence Result Urban Rural Total Household interviews Households selected 2,220 3,280 5,500 Households occupied 2,028 3,058 5,086 Households interviewed 1,880 2,963 4,843 Household response rate1 92.7 96.9 95.2 Interviews with women age 15-49 Number of eligible women 1,682 3,619 5,301 Number of eligible women interviewed 1,544 3,443 4,987 Eligible women response rate2 91.8 95.1 94.1 Interviews with men age 15-49 Number of eligible men 1,638 3,037 4,675 Number of eligible men interviewed 1,441 2,715 4,156 Eligible men response rate2 88.0 89.4 88.9 1 Households interviewed/households occupied 2 Respondents interviewed/eligible respondents 8 | Introduction In the households selected for the youth and older adult survey, a total of 477 eligible girls and 439 eligible boys age 12-14 were identified. Interviews were completed with 459 girls and 411 boys, yielding response rates of 96 percent and 94 percent, respectively. The response rates for girls are the same for urban and rural areas. For boys, the response rate is slightly lower in urban than in rural areas (89 percent compared with 94 percent). A total of 693 eligible women age 50 and over were identified. Interviews were completed with 661 of these women, yielding a 95 percent response rate. In the same households, a total of 492 eligible men age 50 and over were identified and interviews were completed with 456 of these men, yielding a male response rate of 93 percent. The response rates are slightly lower in urban than in rural areas, and lower among men than women. Table 1.3 Results of the household and individual interviews in households selected for youth and older adults survey Number of households, number of interviews, and response rates for subsample selected for the youth and older adults survey, according to residence (unweighted), Swaziland 2006-07 Residence Result Urban Rural Total Household interviews Households selected 1,110 1,640 2,750 Households occupied 1,020 1,523 2,543 Households interviewed 937 1,473 2,410 Household response rate1 91.9 96.7 94.8 Interviews with female youth 12-14 Number of eligible female youth 78 399 477 Number of eligible female youth interviewed 75 384 459 Eligible female youth response rate2 96.2 96.2 96.2 Interviews with male youth 12-14 Number of eligible male youth 62 377 439 Number of eligible male youth interviewed 55 356 411 Eligible male youth response rate2 88.7 94.4 93.6 Interviews with women 50+ Number of eligible women 50+ 114 579 693 Number of eligible women 50+ interviewed 104 557 661 Eligible women 50+ response rate2 91.2 96.2 95.4 Interviews with men 50+ Number of eligible men 50+ 111 381 492 Number of eligible men 50+ interviewed 96 360 456 Eligible men 50+ response rate2 86.5 94.5 92.7 1 Households interviewed/households occupied 2 Respondents interviewed/eligible respondents Household Population and Housing Characteristics | 9 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS 2 Petronella Mamba This chapter presents a description of the demographic and socioeconomic characteristics of the population in the households sampled in the 2006-07 SDHS. For the purpose of the 2006-07 SDHS, a household was defined as a person or a group of persons, related or unrelated, who live together and share a common source of food. The Household Questionnaire (see Appendix E) included a schedule collecting basic demographic and socioeconomic information (e.g., age, sex, education attainment, and current school attendance) for all usual residents and visitors of the household who spent the night preceding the interview. This method of data collection allows the analysis of the results for either the de jure (usual residents) or de facto (those who are there at the time of the survey) populations. The household questionnaire also obtained information on housing facilities (e.g., sources of water supply and sanitation facilities) and household possessions. The information presented in this chapter is intended to facilitate interpretation of the key demographic, socioeconomic, and health indices presented later in the report. It is also intended to assist in the assessment of the representativeness of the survey sample. 2.1 POPULATION BY AGE AND SEX Age and sex are important demographic variables and are the primary basis of demographic classification. Needs and services for a given population mostly depend on its age and sex structure. Age and sex structure have a strong bearing on the population’s fertility, mortality, and nuptiality patterns. Table 2.1 gives the age and sex distribution of the de facto population by urban and rural residence. Eight in ten of the Swazi population live in the rural areas. Table 2.1 Household population by age, sex, and residence Percent distribution of the de facto household population by five-year age groups, according to sex and residence, Swaziland 2006-07 Urban Rural Total Age Male Female Total Male Female Total Male Female Total <5 12.0 12.0 12.0 16.2 14.2 15.2 15.3 13.8 14.5 5-9 10.0 10.3 10.2 16.6 14.0 15.2 15.2 13.2 14.1 10-14 10.3 10.1 10.2 17.1 15.8 16.4 15.7 14.6 15.1 15-19 9.8 10.9 10.4 14.4 11.8 13.0 13.5 11.6 12.5 20-24 11.8 12.6 12.2 8.9 8.9 8.9 9.5 9.7 9.6 25-29 11.7 10.8 11.2 5.5 5.8 5.6 6.8 6.8 6.8 30-34 8.6 8.8 8.7 3.6 4.7 4.2 4.7 5.5 5.1 35-39 7.4 6.8 7.1 3.2 4.2 3.7 4.1 4.7 4.4 40-44 5.2 5.2 5.2 2.3 3.7 3.0 2.9 4.0 3.5 45-49 4.3 3.8 4.0 2.3 3.4 2.9 2.7 3.5 3.1 50-54 3.7 3.0 3.3 2.3 3.3 2.8 2.6 3.3 2.9 55-59 2.5 2.3 2.4 1.8 2.2 2.0 2.0 2.2 2.1 60-64 1.7 1.3 1.5 2.3 2.7 2.5 2.2 2.4 2.3 65-69 0.6 0.9 0.8 1.3 1.8 1.6 1.1 1.7 1.4 70-74 0.1 0.5 0.3 1.1 1.3 1.2 0.9 1.1 1.0 75-79 0.1 0.4 0.3 0.6 0.9 0.8 0.5 0.8 0.7 80 + 0.3 0.3 0.3 0.5 1.3 0.9 0.5 1.1 0.8 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 2,259 2,488 4,747 8,345 9,520 17,865 10,604 12,008 22,612 10 | Household Population and Housing Characteristics There are more women than men in Swaziland (53 percent and 47 percent, respectively). The sex ratio (proportion of men out of 100 women) is 88. This is consistent with findings of the 1986 census (89) and the 1997 census (90). The sex ratio in the rural areas is lower than that in the urban areas (88 percent compared with 91 percent), which may be due to higher rural-urban migration among men than among women. Swaziland’s population is young, with 44 percent of the total population under 15 years of age, and less than 4 percent is 65 years or older. Figure 2.1 illustrates the age structure of the household population in a population pyramid. The two bottom bars of the pyramid that represent population 0-9 years are smaller than the bar for the next older age. This means that although fertility levels are still high, resulting in a wide base, the country has experienced significant fertility declines. The bars between age 20 and 45 for both males and females narrow rapidly with increasing age, reflecting high rates of mortality, probably due to AIDS-related factors. The proportion of women in the older age groups is much higher than the proportion of men. Because there is no evidence of significantly higher emigration among men, one may conclude that men have higher mortality levels. 2.2 HOUSEHOLD COMPOSITION Information on key aspects of the composition of households, including the sex of the head of the household and the size of the household, is presented in Table 2.2. These characteristics are important because they are associated with the welfare of the household. Female-headed households are, for example, typically poorer than male-headed households. Economic resources are often more limited in larger households. Moreover, where the size of the household is large, crowding also can lead to health problems. Table 2.2 presents the distribution of households by the sex of heads of household and by household size in urban and rural areas. The average size of households according to residence is also Figure 2.1 Population Pyramid 80+ 75-79 70-74 65-69 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5-9 0-4 0246810 0 2 4 6 8 10 SDHS 2006-07 Male Percent Female Age Household Population and Housing Characteristics | 11 given. Households in Swaziland are almost as likely to be headed by a woman as by a man (48 percent compared with 52 percent). There has been an increasing trend in the proportion of households headed by a woman, from 40 percent in the 1986 census to 43 percent in the 1997 census. Rural households are more likely than urban households to be headed by a woman; 52 percent of households in rural areas are headed by a woman compared to 39 percent in urban areas. Both rural and urban areas showed an increase in the proportion of households headed by a woman. The 1997 census found that 49 percent of households in the rural areas and 31 percent of households in the urban areas were headed by a woman. Generally, the size of a household has a negative correlation with socioeconomic status. However, it has also been found that there are significant benefits in having other members in a household. The overall mean size of households in 2006-07 is 4.6 persons; rural households are larger than urban households (5.4 and 3.0 persons per household, respectively). The mean household size is 5.3 persons in 1997, a drop from the 6.0 persons per household reported in the 1986 census. There are marked differences between rural and urban households. In the rural areas, 16 percent of households have nine or more members compared to only 3 percent in the urban areas. As expected, a large proportion (35 percent) of households in the urban areas has only one member. In contrast, one-person households account for only one-fifth of all households in Swaziland. 2.3 EDUCATION OF THE HOUSEHOLD POPULATION Education is a key determinant of the lifestyle and status an individual enjoys in a society. Studies have consistently shown that educational attainment has a positive effect on health behaviours and attitudes. Re- sults from the 2006-07 SDHS can be used to look at educational attainment among household members and school attendance, repetition, and drop-out rates among youth. In the analysis presented below, the official age for entry into the primary level is age six years. The official primary level of schooling consists of seven years (Grades 1-7) while the number of years assumed for completion of secondary school is five years. 2.3.1 Educational Attainment The 2006-07 SDHS collected education data on the highest level completed for males and females age six years and over. Table 2.3.1 and 2.3.2 show that, in general, the proportion of males and females with no education has been declining over time and the proportion attaining higher levels has been increasing. There are slight differentials between sexes in the levels of education attained, with men generally having higher levels. For example, the proportion of women with no education is 13 percent compared with 12 percent of men, and 5 percent of women have had higher than secondary level of education, compared to 6 percent of men. Table 2.2 Household composition Percent distribution of households by sex of head of household and by household size and by mean size of household according to residence, Swaziland 2006-07 Residence Characteristic Urban Rural Total Household headship Male 61.0 47.9 52.1 Female 39.0 52.1 47.9 Total 100.0 100.0 100.0 Number of usual members 0 0.2 0.4 0.3 1 35.4 12.5 19.9 2 19.1 9.6 12.6 3 12.5 10.2 11.0 4 12.1 12.7 12.5 5 7.0 12.2 10.5 6 5.3 10.4 8.7 7 3.0 9.1 7.1 8 2.2 6.6 5.1 9+ 3.3 16.4 12.2 Total 100.0 100.0 100.0 Mean size of households 3.0 5.4 4.6 Number of households 1,565 3,278 4,843 Note: Table is based on de jure household members, i.e., usual residents. 12 | Household Population and Housing Characteristics There are large differentials by residence. For both rural men and women, urban residents consistently have higher levels of education. For instance, the median years completed for urban women is 7.8 years, whereas for women in rural areas it is only 4.7 years. For men, the corresponding proportion is 8.0 and 3.9 years, respectively. Across regions, men and women in Manzini are better educated than those in other regions. On the other hand, men and women in Lubombo are the least educated. As expected, educational attainment is positively related to the wealth status of the household. Women and men in wealthier households are better educated than those in poorer households. Table 2.3.1 Educational attainment of the female household population Percent distribution of the de facto female household populations age six and over by highest level of schooling attended or completed and median years completed, according to background characteristics, Swaziland 2006-07 Background characteristic No education Some primary Completed primary1 Some secondary Completed secondary2 Tertiary Don't know/ missing Total Number Median years completed Age 6-9 16.2 82.7 0.2 0.2 0.0 0.0 0.7 100.0 1,274 0.3 10-14 2.3 85.3 6.6 5.4 0.0 0.0 0.4 100.0 1,754 3.7 15-19 2.7 27.4 14.6 50.1 4.3 0.6 0.3 100.0 1,392 7.3 20-24 6.0 16.8 10.4 44.7 16.9 5.0 0.3 100.0 1,161 8.4 25-29 6.9 14.4 9.8 34.4 22.6 11.2 0.7 100.0 821 9.1 30-34 8.2 16.3 12.4 34.2 15.7 12.5 0.8 100.0 663 8.6 35-39 11.1 22.3 11.8 29.3 11.5 12.7 1.2 100.0 569 7.7 40-44 15.5 25.7 8.3 29.2 8.3 11.0 1.9 100.0 483 6.9 45-49 22.4 25.0 11.9 25.1 6.3 8.1 1.2 100.0 415 6.2 50-54 28.4 30.9 13.6 16.7 2.7 6.8 0.9 100.0 392 4.8 55-59 26.8 36.1 11.1 14.5 2.0 7.9 1.6 100.0 262 3.9 60-64 47.7 36.0 5.5 5.0 1.1 3.0 1.9 100.0 293 0.4 65+ 59.3 28.8 5.7 3.1 0.2 1.3 1.6 100.0 559 0.0 Residence Urban 7.7 27.6 9.0 29.1 13.5 12.5 0.7 100.0 2,134 7.8 Rural 15.0 45.5 8.9 22.1 5.2 2.5 0.8 100.0 7,906 4.7 Region Hhohho 14.1 37.3 9.1 25.5 7.0 6.2 0.8 100.0 2,615 5.7 Manzini 9.3 39.9 9.5 25.1 9.5 6.0 0.7 100.0 3,167 6.0 Shiselweni 13.0 45.6 8.8 23.6 5.8 2.7 0.5 100.0 2,313 4.9 Lubombo 19.8 46.0 7.6 18.7 4.1 2.7 1.1 100.0 1,944 4.0 Wealth quintile Lowest 26.8 52.1 7.7 10.8 2.0 0.0 0.5 100.0 1,946 2.4 Second 18.3 49.7 9.6 17.9 3.2 0.2 1.1 100.0 1,988 3.8 Middle 10.3 46.5 10.2 26.1 5.4 0.7 0.7 100.0 1,988 5.3 Fourth 8.0 37.2 9.0 33.4 9.0 2.7 0.8 100.0 2,037 6.5 Highest 4.7 24.2 7.9 29.2 14.5 18.8 0.8 100.0 2,081 8.8 Total 13.4 41.7 8.9 23.6 6.9 4.6 0.8 100.0 10,040 5.3 Note: Total includes three females with information missing on age 1 Completed 7th grade at the primary level 2 Completed 5th grade at the secondary level Household Population and Housing Characteristics | 13 Table 2.3.2 Educational attainment of the male household population Percent distribution of the de facto male household populations age six and over by highest level of schooling attended or completed and median years completed, according to background characteristics, Swaziland 2006-07 Background characteristic No education Some primary Completed primary1 Some secondary Completed secondary2 Tertiary Don't know/ missing Total Number Median years completed Age 6-9 19.9 79.3 0.1 0.0 0.0 0.0 0.7 100.0 1,314 0.1 10-14 3.4 89.7 5.0 1.8 0.1 0.0 0.1 100.0 1,661 3.0 15-19 3.4 39.7 13.8 39.4 3.4 0.1 0.2 100.0 1,427 6.5 20-24 5.1 20.0 9.9 39.0 20.1 5.3 0.7 100.0 1,011 8.6 25-29 7.8 17.4 7.5 30.1 25.7 11.0 0.5 100.0 719 8.9 30-34 9.2 18.9 10.1 21.4 24.7 14.7 0.9 100.0 496 9.3 35-39 11.5 20.5 10.1 23.3 13.9 19.6 1.2 100.0 431 8.4 40-44 19.1 16.3 10.2 22.0 11.7 20.0 0.6 100.0 306 8.3 45-49 20.4 22.9 9.2 21.1 10.6 15.4 0.3 100.0 285 6.7 50-54 24.9 22.7 7.6 21.7 7.6 13.6 1.9 100.0 273 6.2 55-59 23.5 33.7 11.6 17.3 1.7 11.2 1.0 100.0 210 4.6 60-64 35.0 29.8 11.6 10.8 3.2 7.4 2.2 100.0 233 3.5 65+ 55.5 24.4 8.6 8.7 0.6 1.3 0.9 100.0 318 0.0 Residence Urban 7.7 29.5 7.1 24.3 16.0 14.9 0.5 100.0 1,946 8.0 Rural 13.5 50.9 8.1 18.0 6.1 2.8 0.6 100.0 6,738 3.9 Region Hhohho 13.2 41.0 7.9 21.2 9.2 6.8 0.7 100.0 2,285 5.3 Manzini 7.8 43.5 9.0 20.4 11.1 7.5 0.6 100.0 2,669 5.8 Shiselweni 13.1 51.1 7.6 19.5 6.1 2.5 0.1 100.0 1,940 4.0 Lubombo 16.6 50.9 6.6 15.6 5.2 4.0 1.0 100.0 1,790 3.4 Wealth quintile Lowest 24.3 58.5 6.2 8.7 1.4 0.0 0.8 100.0 1,646 2.0 Second 13.3 54.1 9.9 17.4 4.5 0.3 0.4 100.0 1,613 3.7 Middle 10.2 52.3 8.2 20.9 6.7 1.1 0.5 100.0 1,759 4.4 Fourth 9.6 42.0 8.3 25.4 10.3 3.8 0.7 100.0 1,782 5.7 Highest 5.0 26.4 7.0 23.4 17.1 20.5 0.5 100.0 1,884 9.0 Total 12.2 46.1 7.9 19.4 8.3 5.5 0.6 100.0 8,684 4.7 Note: Total includes one male with information missing on age 1 Completed 7th grade at the primary level 2 Completed 5 th grade at the secondary level 2.3.2 School Attendance Rates Table 2.4 presents primary school and secondary school net and gross attendance ratios (NAR and GAR) for the school year that started in 2005 by household residence and regions. The NAR for primary school is the percentage of the primary-school-age (6-12 years) population that is attending primary school. The NAR for secondary school is the percentage of the secondary-school-age (13-17 years) population that is attending secondary school. By definition, the NAR cannot exceed 100 percent. The GAR for primary school is the total number of primary school students, of any age, expressed as a percentage of the official primary-school-age population. The GAR for secondary school is the total number of secondary school students, of any age, expressed as a percentage of the official secondary- school-age population. If there are significant numbers of overage and underage students at a given level of schooling, the GAR can exceed 100 percent. Youth are considered to be attending school currently if they attended formal academic school at any point during the given school year. 14 | Household Population and Housing Characteristics Table 2.4 School attendance ratios Net attendance ratios (NAR) and gross attendance ratios (GAR) for the de facto household population by sex and level of schooling; and the gender parity index (GPI), according to background characteristics, Swaziland 2006-07 Net attendance ratio1 Gross attendance ratio2 Background characteristic Male Female Total Gender Parity Index3 Male Female Total Gender Parity Index3 PRIMARY SCHOOL Residence Urban 87.7 84.6 86.1 0.96 113.0 99.7 106.1 0.88 Rural 82.1 85.7 83.9 1.05 125.0 118.6 121.8 0.95 Region Hhohho 82.0 89.2 85.6 1.09 123.7 117.5 120.6 0.95 Manzini 89.5 87.6 88.5 0.98 126.5 118.0 122.2 0.93 Shiselweni 75.5 80.5 77.9 1.07 112.7 110.5 111.6 0.98 Lubombo 83.8 84.4 84.1 1.01 131.5 116.2 123.8 0.88 Wealth quintile Lowest 73.4 79.8 76.6 1.09 118.3 113.7 116.0 0.96 Second 84.2 84.8 84.5 1.01 126.6 117.4 121.9 0.93 Middle 83.7 88.6 86.1 1.06 126.1 120.7 123.5 0.96 Fourth 85.6 88.4 87.0 1.03 127.5 117.5 122.5 0.92 Highest 91.6 88.0 89.8 0.96 117.2 106.9 112.0 0.91 Total 82.9 85.6 84.2 1.03 123.3 115.7 119.5 0.94 SECONDARY SCHOOL Residence Urban 52.2 54.2 53.2 1.04 83.1 83.9 83.5 1.01 Rural 27.1 38.1 32.5 1.41 54.1 55.5 54.8 1.03 Region Hhohho 32.1 43.5 37.7 1.36 58.5 63.9 61.1 1.09 Manzini 35.7 42.7 39.3 1.20 67.9 61.3 64.5 0.90 Shiselweni 30.3 39.1 34.6 1.29 57.8 58.6 58.2 1.01 Lubombo 22.7 35.1 28.6 1.55 45.1 53.8 49.2 1.19 Wealth quintile Lowest 11.7 20.2 15.9 1.73 27.2 29.9 28.5 1.10 Second 25.8 29.8 27.8 1.16 51.8 45.0 48.5 0.87 Middle 26.6 43.4 34.8 1.63 58.0 60.5 59.2 1.04 Fourth 35.7 48.5 42.1 1.36 64.2 76.6 70.4 1.19 Highest 59.5 61.2 60.4 1.03 97.4 87.7 92.3 0.90 Total 30.7 40.5 35.6 1.32 58.2 59.8 59.0 1.03 1 The NAR for primary school is the percentage of the primary-school-age (6-12 years) population that is attending primary school. The NAR for secondary school is the percentage of the secondary-school-age (13-17 years) population that is attending secondary school. By definition the NAR cannot exceed 100 percent. 2 The GAR for primary school is the total number of primary school students, expressed as a percentage of the official primary-school-age population. The GAR for secondary school is the total number of secondary school students, expressed as a percentage of the official secondary-school-age population. If there are significant numbers of overage and underage students at a given level of schooling, the GAR can exceed 100 percent. 3 The GPI for primary school is the ratio of the primary school NAR (GAR) for females to the NAR (GAR) for males. The GPI for secondary school is the ratio of the secondary school NAR (GAR) for females to the NAR (GAR) for males. The gender parity index (GPI) assesses sex-related differences in school attendance rates and is calculated by dividing the GAR for females by the GAR for males. A GPI of less than one indicates a gender disparity in favour of males, i.e., a higher proportion of males than females attends that level of schooling. A GPI greater than one indicates a gender disparity in favour of females. A GPI of one indicates parity or equality between the rates of participation for males and females. Household Population and Housing Characteristics | 15 Table 2.4 shows that for both boys and girls at primary school level, the overall net attendance ratio (NAR) is 84 percent; 83 percent for males and 86 percent for females. For males, those living in urban areas are much more likely to attend school than rural residents (88 percent compared with 82 percent). For females, however, attendance in the rural and urban areas is similar (86 percent and 85 percent, respectively). It is worth noting that the GARs for both boys and girls at primary-school level are much higher than 100, signifying that the number of primary-school students of any age is larger than the official primary-school-age population. Across all subgroups of population, the GAR for boys is higher than for girls. The GPI for the GAR is consistently less than one, implying more boys are outside the primary- school age range. At secondary level, the NARs are very low (36 percent); 31 percent for boys and 41 percent for girls. The ratio is lowest for boys in the rural areas (27 percent), in Lubombo (23 percent), and in the lowest wealth quintile (12 percent). The GPI for NAR at secondary school is 1.32, implying that girls of secondary-school age are more likely than boys in the same age group to attend secondary school. GAR at secondary level is only 59 percent, which implies that there is a significant number of secondary-school- age boys and girls who are not attending secondary school. Figure 2.2 illustrates age-specific attendance rates. This includes the percentage of population age 5-24 years who attend school, regardless of the level attended (primary, secondary, or higher). 2.3.3 Grade Repetition and Dropout Rates Repetition and dropout rates presented in Table 2.5 describe the flow of pupils through the educational system in Swaziland at the primary level. The repetition rates indicate the percentage of pupils who attended a particular grade during the school year that started in 2004 who again attended that Figure 2.2 Age-specific Attendance Rates of the De-Facto Population 5 to 24 Years ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ') ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 0 20 40 60 80 100 Percent Male Female) ' SDHS 2006-07 16 | Household Population and Housing Characteristics same class during the following school year. The dropout rates show the percentage of pupils in a grade during the school year that started in 2005 who no longer attended school the following school year. The repetition rates generally decline from 23 percent in Grade 1 to 9 percent in Grade 7. There are differentials across subgroups of students, with boys, rural students, and students from households in the lowest wealth quintiles consistently having higher repetition rates than other students. For example, the repetition rates for boys range from 26 percent to 12 percent, while for girls the corresponding rates range from 20 percent to 7 percent. Students in Hhohho have the lowest repetition rates, while those in Lubombo have the highest repetition rates. Table 2.5 Grade repetition and dropout rates Repetition and dropout rates for the de facto household population age 5-24 who attended primary school in the previous school year by school grade, according to background characteristics, Swaziland 2006-07 School grade Background characteristic 1 2 3 4 5 6 7 REPETITION RATE1 Sex Male 26.4 13.4 18.6 14.8 12.5 13.4 11.8 Female 19.7 8.6 13.9 11.3 10.7 10.4 6.7 Residence Urban 22.9 9.2 8.6 8.1 5.4 9.2 8.8 Rural 23.2 11.4 17.6 13.8 12.7 12.3 9.4 Region Hhohho 25.3 4.8 8.0 7.5 5.6 16.0 10.2 Manzini 19.9 16.3 16.5 14.1 10.7 9.0 7.6 Shiselweni 23.2 11.8 19.3 13.5 16.2 8.7 11.1 Lubombo 25.7 10.3 22.0 17.1 14.7 14.9 9.0 Wealth quintile Lowest 25.6 13.6 19.6 17.6 15.1 9.8 11.2 Second 22.7 12.5 15.2 11.5 13.3 14.0 10.8 Middle 23.5 12.7 17.4 11.1 14.8 16.2 6.9 Fourth 19.6 5.6 18.8 13.5 7.2 10.3 12.8 Highest 23.4 9.6 7.2 11.1 5.7 6.9 5.7 Total 23.2 11.2 16.3 13.0 11.6 11.8 9.3 DROPOUT RATE2 Sex Male 3.4 5.0 3.1 3.2 3.1 7.7 8.0 Female 2.6 2.1 2.1 6.4 4.3 6.9 9.4 Residence Urban 0.7 2.2 0.0 6.6 3.7 9.1 8.9 Rural 3.4 3.8 3.0 4.5 3.7 7.0 8.6 Region Hhohho 2.8 0.6 1.0 0.8 3.0 3.9 4.1 Manzini 0.4 1.5 0.7 1.5 2.6 4.6 7.3 Shiselweni 6.7 10.8 6.9 14.3 6.2 15.7 13.9 Lubombo 2.7 1.3 1.4 2.3 3.0 4.8 8.9 Wealth quintile Lowest 3.5 5.7 3.7 4.5 3.3 4.9 6.8 Second 4.2 4.6 3.6 7.5 2.4 10.2 9.8 Middle 2.5 2.4 3.1 5.1 4.8 7.0 10.1 Fourth 3.2 2.3 0.4 2.8 5.2 7.8 9.7 Highest 0.2 2.0 0.8 4.0 2.9 5.7 5.8 Total 3.0 3.7 2.6 4.8 3.7 7.3 8.6 1 The repetition rate is the percentage of students in a given grade in the previous school year who are repeating that grade in the current school year. 2 The dropout rate is the percentage of students in a given grade in the previous school year who are not attending school in the current school year. Household Population and Housing Characteristics | 17 The dropout rates are higher for boys than for girls in the lower grades, but boys’ dropout rates are lower than girls’ rates. in the higher grades. This may be due to more girls dropping out due to various reasons including a pregnancy. For both boys and girls, the dropout rates increase with grade. Also notable is that children in Shiselweni are more likely to drop out of school than children in other regions. In sum, in Swaziland on average one in three children who started Grade 1 will not complete Grade 7. 2.4 HOUSEHOLD ENVIRONMENT The physical characteristics of the dwelling in which a household lives are important determinants of the health status of household members, especially children. They can also be used as indicators of the socio-economic status of households. The 2006-07 SDHS respondents were asked a number of questions about their household environment, including questions on the source of drinking water: type of sanitation facility; type of flooring; walls; and roof; and number of rooms in the dwelling. The results are presented both in terms of households and of their usual members. 2.4.1 Drinking Water Increasing access to improved drinking water is one of the Millennium Development Goals that Swaziland, along with other nations worldwide. has adopted (United Nations General Assembly, 2001). Table 2.6 includes a number of indicators that are useful in monitoring household access to improved drinking water (WHO and UNICEF, 2005). The source of drinking water is an indicator of whether it is suitable for drinking. Sources which are likely to provide water suitable for drinking are identified as improved sources in Table 2.6. They include a piped source within the dwelling or plot, public tap, tube well or borehole, protected well or spring, and rainwater.1 Lack of ready access to a water source may limit the quantity of suitable drinking water that is available to a household. Even if the water is obtained from an improved source, water that must be fetched from a source that is not immediately accessible to the household may be contaminated during transport or storage. Another factor in considering the accessibility of water sources is the fact that the burden of going for water often falls disproportionately on female members of the household. Finally, home water treatment can be effective in improving the quality of household drinking water. Seven in ten households in Swaziland obtain water from improved sources. This proportion represents an improvement from the 1997 Population and Housing Census (56 percent). There is a wide variation between urban and rural households (92 percent and 59 percent, respectively). In the ten years since the 1997 Census, rural areas have shown a greater improvement in access to safe water than urban areas. According to the 1997 Population and Housing Census, the proportion of households with safe drinking water is 89 percent in urban areas and 40 percent in rural areas. Water is available on the premises for 76 percent of households in the urban areas and 27 percent of households in rural areas. Overall, one in four households take 30 or more minutes to obtain water; 4 percent in the urban areas compared with 34 percent in the rural areas. Getting water is a chore predominantly done by women. Water is collected by women for more than half of the population; 46 percent by women age 15 or older and 7 percent by female children under age 15. This is particularly true in the rural areas, where for 54 percent of the population’s water is collected by women age 15 or older and 8 percent by female children under age 15. 1 The categorisation into improved and non-improved follows that proposed by the WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation (WHO and UNICEF, 2004). 18 | Household Population and Housing Characteristics Households were further asked if they treat the water before drinking it. A total of 86 percent of households do not treat the water in any way. The most common form of treatment is the use of bleach or chlorine (8 percent). Table 2.6 Household drinking water Percent distribution of households and de jure population by source, time to collect, and person who usually collects drinking water; and percentage of households and the de jure population by treatment of drinking water, according to residence, Swaziland 2006-07 Households Population Characteristic Urban Rural Total Urban Rural Total Source of drinking water Improved source 92.2 59.1 69.8 91.9 56.4 63.9 Piped water into dwelling/yard/plot 72.6 22.5 38.7 72.4 19.8 30.9 Public tap/standpipe 15.2 18.7 17.6 14.1 18.5 17.6 Tube well or borehole 2.5 9.0 6.9 3.0 8.6 7.4 Protected dug well 1.2 5.8 4.3 1.3 6.6 5.5 Protected spring 0.5 2.2 1.7 0.9 2.3 2.0 Rainwater 0.1 0.7 0.5 0.2 0.6 0.5 Non-improved source 7.4 40.6 29.9 7.8 43.3 35.8 Unprotected dug well 2.9 12.3 9.3 2.9 13.5 11.2 Unprotected spring 0.8 4.6 3.4 1.2 5.3 4.4 Tanker truck/cart with small tank 1.5 1.6 1.6 1.3 1.2 1.2 Surface water 2.2 22.0 15.6 2.3 23.3 18.9 Other 0.3 0.3 0.3 0.3 0.2 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Percentage using any improved source of drinking water 92.3 59.1 69.8 91.9 56.4 63.9 Time to obtain drinking water (round trip) Water on premises 75.9 27.3 43.0 75.4 24.0 34.8 Less than 30 minutes 19.4 38.0 32.0 20.2 40.0 35.8 30 minutes or longer 4.3 34.2 24.5 3.9 35.5 28.8 Don't know/missing 0.3 0.6 0.5 0.5 0.5 0.5 Total 100.0 100.0 100.0 100.0 100.0 100.0 Person who usually collects drinking water Adult female 15+ 14.5 48.9 37.8 16.7 54.2 46.3 Adult male 15+ 7.0 11.7 10.2 3.9 7.4 6.7 Female child under age 15 0.7 6.3 4.5 1.4 8.1 6.7 Male child under age 15 0.9 3.3 2.5 1.5 3.5 3.1 Any person 0.4 1.6 1.2 0.5 2.0 1.7 Other 0.4 0.8 0.6 0.2 0.9 0.7 Water on premises 75.9 27.3 43.0 75.4 24.0 34.8 Missing 0.3 0.2 0.2 0.3 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Water treatment prior to drinking1 Boiled 2.0 3.4 3.0 2.0 2.9 2.7 Bleach/chlorine 7.9 8.3 8.2 8.6 8.1 8.2 Strained through cloth 0.0 0.1 0.1 0.0 0.1 0.1 Ceramic, sand, or other filter 0.5 0.5 0.5 0.4 0.4 0.4 Other 0.0 0.3 0.2 0.1 0.3 0.2 No treatment 84.9 86.6 86.0 84.7 87.5 86.9 Percentage using an appropriate treatment method2 10.4 11.7 11.2 10.8 11.0 11.0 Number 1,565 3,278 4,843 4,705 17,598 22,302 1 Respondents may report multiple treatment methods so the sum of treatment may exceed 100 percent. 2 Appropriate water treatment methods include boiling, bleaching, straining, filtering, and solar disinfecting. Household Population and Housing Characteristics | 19 2.4.2 Household Sanitation Facilities Ensuring adequate sanitation facilities is another of the Millennium Development Goals which Swaziland shares with other countries. A household is classified as having an improved toilet if the toilet is used only by members of one household (i.e., it is not shared) and if the facility used by the household separates the waste from human contact (WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation, 2004). Proper sanitation facilities lead to improved hygiene practices, and ultimately low infant mortality rates. Normally, one would expect the urban areas to have better sanitation facilities, but according to the 2006-07 SDHS, rural households in Swaziland have better sanitation facilities than urban households; 52 percent of the rural households use improved, not shared facilities compared with 44 percent of urban households (Table 2.7). The proportion of households using a flush toilet has declined from 20 percent in the 1997 Population and Housing Census to 15 percent in the 2006-07 SDHS. The majority of households in the urban areas (53 percent) use shared facilities, compared with 22 percent in the rural areas. The proportion of households who have no toilet facility declined from 24 percent in 1997 to 18 percent in the 2006-07 SDHS. Table 2.7 Household sanitation facilities Percent distribution of households and de jure population by type of toilet/latrine facilities, according to residence, Swaziland 2006-07 Households Population Type of toilet/ latrine facility Urban Rural Total Urban Rural Total Improved, not shared facility 43.5 52.1 49.4 55.6 56.7 56.5 Flush/pour flush 34.2 5.4 14.7 39.2 3.7 11.2 Ordinary pit latrine 7.2 34.0 25.4 13.0 37.7 32.5 Ventilated improved pit (VIP) latrine 2.1 12.7 9.3 3.4 15.3 12.8 Non-improved facility 56.4 47.8 50.7 44.3 43.3 43.6 Any facility shared with other households 53.1 21.9 32.0 42.3 18.3 23.4 No facility/bush/field 2.5 25.6 18.2 1.7 24.7 19.9 Other 0.1 0.1 0.1 0.0 0.1 0.1 Missing 0.7 0.2 0.4 0.3 0.2 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Number 1,565 3,278 4,843 4,705 17,598 22,302 2.4.3 Housing Characteristics Table 2.8 presents information on a number of characteristics of the dwelling in which SDHS households live. These characteristics reflect the household’s socioeconomic situation. They also may influence environmental conditions—for example, in the case of the use of biomass fuels, exposure to indoor pollution—that have a direct bearing on the health and welfare of household members. The use of electricity as an energy source usually goes hand in hand with improved housing structures and a better standard of living. There have been a number of initiatives by the Government of Swaziland to bring electricity to rural areas in recent years. Clinics and schools are the first targets of the programme. Thirty-five percent of households reported having electricity. There are marked differences between urban and rural; only 22 percent of rural households reported using electricity compared with 63 percent of urban households. 20 | Household Population and Housing Characteristics Table 2.8 Housing characteristics Percent distribution of households and de jure population by housing characteristics and percentage using solid fuel for cooking; and among those using solid fuels, percent distribution by type of fire/stove, according to residence, Swaziland 2006-07 Households Population Housing characteristic Urban Rural Total Urban Rural Total Electricity Yes 63.4 21.7 35.2 65.2 20.2 29.7 No 36.6 78.3 64.8 34.8 79.8 70.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 Flooring material Earth, sand 1.2 3.8 3.0 1.4 3.2 2.8 Dung 0.4 12.0 8.2 0.5 12.0 9.6 Wood/planks 0.4 0.0 0.1 0.5 0.0 0.1 Parquet or polished wood 0.3 0.0 0.1 0.4 0.0 0.1 Vinyl or asphalt strips 2.7 0.1 1.0 3.1 0.1 0.7 Ceramic tiles 9.6 2.5 4.8 12.6 2.2 4.4 Cement 80.3 80.4 80.3 75.9 81.4 80.2 Carpet 4.9 1.1 2.3 5.4 1.0 1.9 Other/missing 0.2 0.1 0.1 0.3 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Wall material Cane/palm/trunks 0.1 0.2 0.2 0.3 0.2 0.2 Mud 4.7 12.0 9.7 4.6 11.7 10.2 Bamboo with mud 1.1 3.3 2.5 1.1 3.0 2.6 Stone with mud 4.6 11.5 9.3 5.1 11.8 10.3 Plywood 0.4 0.1 0.2 0.3 0.1 0.1 Reused wood 0.4 0.3 0.3 0.7 0.4 0.4 Cement 51.7 29.9 37.0 52.6 30.2 34.9 Stone with lime/ cement 2.2 5.9 4.7 2.3 6.3 5.4 Bricks 3.9 3.2 3.4 3.2 2.7 2.8 Cement blocks 27.5 28.2 28.0 26.9 28.3 28.0 Mud blocks 2.7 4.7 4.0 2.5 4.9 4.4 Wood planks/shingles 0.4 0.2 0.3 0.3 0.1 0.1 Other 0.3 0.4 0.3 0.3 0.2 0.2 Missing 0.1 0.1 0.1 0.1 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Roofing material Grass 0.4 16.9 11.5 0.5 17.4 13.8 Wood planks 0.2 0.0 0.1 0.2 0.0 0.0 Corrugated iron 81.7 80.6 80.9 82.6 80.5 80.9 Asbestos 10.5 0.3 3.6 8.6 0.1 1.9 Tiles 5.6 1.9 3.1 6.6 1.8 2.8 Slate 0.1 0.0 0.0 0.2 0.0 0.0 Concrete 1.0 0.1 0.4 1.0 0.0 0.2 Other 0.1 0.2 0.2 0.2 0.1 0.1 Missing 0.2 0.1 0.2 0.2 0.1 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Rooms used for sleeping One 54.2 25.7 34.9 35.5 14.6 19.0 Two 25.3 32.2 30.0 30.3 29.6 29.8 Three or more 20.3 41.8 34.8 34.1 55.6 51.0 Missing 0.2 0.3 0.3 0.1 0.2 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Place for cooking In the house 83.9 30.0 47.4 78.9 24.5 35.9 In a separate building 11.4 57.2 42.4 15.8 63.4 53.3 Outdoors 3.6 12.4 9.5 4.9 12.0 10.5 Other 0.0 0.1 0.1 0.0 0.1 0.0 Missing 1.2 0.4 0.6 0.4 0.1 0.2 Total 100.0 100.0 100.0 100.0 100.0 100.0 Continued… Household Population and Housing Characteristics | 21 Table 2.8—Continued Households Population Housing characteristic Urban Rural Total Urban Rural Total Cooking fuel Electricity 41.0 8.1 18.7 40.0 5.2 12.6 Gas 29.8 9.4 16.0 29.3 6.4 11.2 Coal 0.8 0.1 0.3 0.7 0.1 0.2 Charcoal 0.9 0.2 0.4 0.7 0.3 0.4 Wood 11.4 78.3 56.7 17.4 86.2 71.7 Paraffin 14.7 3.5 7.1 11.3 1.6 3.7 No food cooked in household 1.2 0.4 0.6 0.4 0.1 0.2 Other 0.1 0.1 0.1 0.2 0.1 0.1 Missing 0.0 0.1 0.0 0.0 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Percentage using solid fuel for cooking 13.2 78.6 57.4 18.8 86.6 72.3 Number of households 1,565 3,278 4,843 4,705 17,598 22,302 The type of material used for flooring is an indicator of socioeconomic status, and to some extent determines the household’s vulnerability to exposure to disease-causing agents. Four in five households have a cement floor. Cow dung is more prevalent in rural than in urban areas (12 percent compared with less than 1 percent), while ceramic tiles are more often in urban than in rural households (10 percent compared with 3 percent). Good-quality walls ensure that household members are protected from harsh weather conditions, and therefore exposure to hazardous factors. According to the 2006-07 SDHS, cement is the most often used material for outer walls (37 percent), followed by cement blocks (28 percent). Urban houses are much more likely than rural houses to have cement walls (52 percent compared with 30 percent). The most common form of roofing for both rural and urban areas is corrugated iron (81 percent). Grass is the second most preferred type of roofing material and is almost exclusively used in rural areas (17 percent). Asbestos is used in 4 percent of households, mostly in urban areas. Tiles are also used as roofing material; they are used in 6 percent of households in urban areas and 2 percent of households in rural areas. Congestion is an undesirable condition that can have adverse health effects. The number of rooms used for sleeping is therefore an indicator of the level of crowding. Most households have either one room (35 percent) or three rooms or more (35 percent) for sleeping. However, in terms of the population, more than half (51 percent) of all the people are living in households where there are three or more rooms for sleeping. One may conclude that overcrowding is not yet a problem in Swaziland. Describing the cooking place as being indoors or outdoors is helpful in understanding the level of food exposure to harmful elements. Nine in ten households cook indoors, with 47 percent cooking in the house and 42 percent in a separate building. The latter is more common in rural areas (57 percent) compared than in urban areas (11 percent). The type of fuel used for cooking may have a direct effect on people’s health status, and is also an indicator of a household’s socioeconomic status. Fifty-seven percent of households, or 72 percent of the population, use wood for cooking. There are marked differentials in cooking fuel between rural and urban households; 78 percent of rural households use wood, while 41 percent of urban households use electricity and 30 percent use gas. A significant proportion of households in urban areas (15 percent) use paraffin for cooking. 22 | Household Population and Housing Characteristics Windows are important for health reasons as they provide ventilation and light. The results of the 2006-07 SDHS show that most houses in Swaziland have some type of windows (96 percent); the proportion is lower in rural areas than in urban areas (95 percent compared with 99 percent). Glass and curtains are common in both urban and rural areas (86-96 percent). Wooden windows are more common in rural than in urban areas (15 percent compared with 6 percent). Table 2.9 Type of windows Percent distribution of households and de jure population using different types of windows, according to residence and region, Swaziland 2006-07 Households Population Type of windows Urban Rural Total Urban Rural Total Any windows 99.1 94.8 96.2 99.1 95.7 96.4 Windows with glass 96.1 87.5 90.3 96.4 89.5 90.9 Windows with screens 5.6 1.9 3.1 5.7 2.1 2.8 Windows with curtains 94.9 86.0 88.9 95.0 88.2 89.6 Wooden windows 5.7 14.5 11.6 5.9 15.8 13.7 Other windows 3.7 4.0 3.9 4.0 4.2 4.1 Number of households 1,565 3,278 4,843 4,705 17,598 22,302 2.5 HOUSEHOLD POSSESSIONS The availability of durable consumer goods is a good indicator of a household’s socioeconomic status. Moreover, each particular item has specific benefits. For instance, having access to a radio or a television exposes household members to innovative ideas; a refrigerator prolongs the wholesomeness of foods; and a means of transport allows greater access to many services away from the local area. Table 2.10 shows the availability of selected consumer goods by residence. The most popular household effects are a watch/clock and a radio (each 77 percent). These items are available in urban areas and in rural areas. Stoves are also available in the majority of households—62 percent of households have a stove. One in three households has a TV set and a refrigerator. The majority of households in Swaziland have access to communication network, because 60 percent have access to a mobile phone. The most common means of transportation is a car or truck (19 percent), followed by bicycle (10 percent). The use of a motorcycle/scooter, tractor, and animal-drawn cart is very limited. Six in ten households in Swaziland own agricultural land. However, as expected, there is a significant variation between rural and urban households; 17 percent of urban households have agricultural land compared with 80 percent of rural households. This differential is also observed on ownership of farm animals; 74 percent of households in rural areas have farm animals compared with 9 percent in urban areas. More than half (52 percent) of households reported a household member with a bank account. Access to formal monetary services varies by residence; 69 percent in urban areas compared with 43 percent in rural areas. Household Population and Housing Characteristics | 23 Table 2.10 Household possessions Percentage of households and de jure population possessing various household effects, means of transportation, agricultural land and livestock/farm animals by residence, Swaziland 2006-07 Households Population Possession Urban Rural Total Urban Rural Total Household effects Radio 82.1 73.7 76.5 84.0 75.7 77.4 Television 54.4 26.4 35.4 62.1 28.0 35.2 Mobile telephone 75.1 53.3 60.3 78.0 58.1 62.3 Non-mobile telephone 20.1 8.7 12.4 28.0 9.8 13.6 Refrigerator 51.1 26.4 34.4 58.8 28.4 34.8 Stove 87.8 49.7 62.0 88.4 49.8 57.9 Watch/clock 83.8 73.5 76.9 87.9 77.4 79.6 Means of transport Bicycle 9.4 9.5 9.5 12.4 10.9 11.2 Motorcycle/scooter 1.6 0.5 0.9 1.7 0.5 0.8 Animal drawn cart 0.3 2.9 2.1 0.2 4.3 3.4 Car/truck 24.5 15.6 18.5 32.2 17.1 20.3 Tractor 0.6 4.9 3.5 0.7 6.7 5.5 Ownership of agricultural land 16.5 79.5 59.1 21.6 86.1 72.5 Ownership of farm animals1 8.8 74.2 53.1 15.2 84.3 69.7 Household member with bank account 69.3 43.3 51.7 72.4 44.6 50.4 Number 1,565 3,278 4,843 4,705 17,598 22,302 1 Cattle, cows, bulls, horses, donkeys, mules, goats, sheep, or chickens 2.6 WEALTH INDEX The wealth index is a background characteristic that is used throughout the report as a proxy for long-term standard of living of the household. It is based on the data on the household’s ownership of consumer goods; dwelling characteristics; type of drinking water source; toilet facilities; and other characteristics that are related to a household’s socioeconomic status. To construct the index, each of these assets was assigned a weight (factor score) generated through principal component analysis, and the resulting asset scores were standardized in relation to a standard normal distribution with a mean of zero and standard deviation of one (Gwatkin et al., 2000). Each household was then assigned a score for each asset, and the scores were summed for each household. Individuals were ranked according to the total score of the household in which they resided. The sample was then divided into quintiles from one (lowest) to five (highest). A single asset index was developed on the basis of data from the entire country sample and this index is used in all the tabulations presented. Table 2.11 shows the distribution of the de jure household population into five wealth levels (quintiles) based on the wealth index by residence and region. This distribution indicates the degree to which wealth is evenly (or unevenly) distributed by geographic areas. Six in ten of the population residing in urban areas are from the richest quintile. On the other hand, half of rural residents are in the two poorest quintiles. In terms of households, 55 percent of households in the urban areas are in the highest quintile, compared with 13 percent of households in the rural areas. 24 | Household Population and Housing Characteristics Table 2.11 Wealth quintiles Percent distribution of the de jure population and households by wealth quintiles, according to residence and region, Swaziland 2006-07 Wealth quintile Residence/ region Lowest Second Middle Fourth Highest Total Number of population POPULATION Residence Urban 1.3 3.5 11.1 25.9 58.2 100.0 4,705 Rural 25.0 24.4 22.4 18.4 9.8 100.0 17,598 Region Hhohho 15.2 18.2 21.2 20.3 25.1 100.0 5,830 Manzini 10.9 19.1 19.3 24.2 26.6 100.0 6,900 Shiselweni 26.8 22.8 23.8 19.0 7.6 100.0 5,121 Lubombo 32.6 20.7 15.2 14.1 17.5 100.0 4,451 Total 20.0 20.0 20.0 20.0 20.0 100.0 22,302 HOUSEHOLD Residence Urban 0.9 2.7 11.5 29.9 55.0 100.0 1,565 Rural 24.7 23.2 20.9 18.2 12.9 100.0 3,278 Region Hhohho 12.2 14.9 18.9 23.2 30.7 100.0 1,370 Manzini 9.4 14.6 17.4 26.0 32.5 100.0 1,537 Shiselweni 26.3 21.9 21.9 19.3 10.7 100.0 931 Lubombo 26.6 17.1 13.5 16.7 26.2 100.0 1,005 Total 17.0 16.6 17.9 22.0 26.5 100.0 4,843 Among the four regions, households in Manzini and Hhohho are more likely to fall in the highest wealth quintile than those living in the other regions. In these regions, the proportion of households or population increases with the wealth quintile. Hhohho has the most even distribution of wealth, which may be explained by the fact that this region houses the Government headquarters, where the income of civil servants and other professionals is relatively evenly distributed. On the other hand, the Shiselweni and Lubombo regions have the highest proportion of the population in the lowest wealth quintile (26-27 percent). In Shiselweni, the proportion of households or population decreases with an increase in the wealth quintile. Lubombo presents the most skewed distribution of wealth, with 53 percent of its population in the two poorest quintiles and 18 percent of the population in the highest quintile. The situation in Lubombo may be explained by the fact that the only significant economic activity in this region is the sugar industry. 2.7 BIRTH REGISTRATION The registration of births is the inscription of the facts of the birth into an official log kept at the registrar’s office. A birth certificate is issued at the time of registration or later as proof of the registration of the birth. Birth registration is basic to ensuring a child’s legal status and, thus, basic rights and services (UNICEF, 2006; United Nations General Assembly, 2002). The registration of vital events in most developing countries is a function of a number of socioeconomic factors. Registration of births is mandatory in Swaziland. However, for most families this is delayed until such time as it may be needed as a requirement to start schooling. Not all children who are registered may have a birth certificate since some certificates may have been lost or were never issued. However, all children with a certificate have been registered. Household Population and Housing Characteristics | 25 Table 2.11 presents the percentage of children under five years of age whose births were officially registered and the percentage who had a birth certificate at the time of the survey. Overall, 30 percent of births in the past five years were registered and 20 percent had a birth certificate. Older children are more likely than younger children to have been registered; 35 percent of children age 2-4 years were registered compared with 22 percent of children under age 2. The SDHS reveals that there is no difference in birth registration according to sex. Children in urban areas are more likely to be registered and have a birth certificate than rural children (38 percent compared with 28 percent). According to administrative regions, Hhohho has the highest proportion of registered births (35 percent), followed by Manzini (31 percent), and Shiselweni (28 percent). Lubombo has the lowest birth registration coverage (24 percent). Coverage of birth registration increases with wealth status; it ranges from 18 percent for children in the lowest wealth quintile to 50 percent for children in the highest wealth quintile. Table 2.12 Birth registration of children under age five Percentage of de jure children under five years of age whose births are registered with the civil authorities, according to background character- istics, Swaziland 2006-07 Percentage of children whose births are registered Background characteristic Had a birth certificate Did not have a birth certificate Total registered Number of children Age <2 11.7 10.0 21.6 1,265 2-4 25.4 9.8 35.1 1,953 Sex Male 19.9 9.9 29.8 1,600 Female 20.1 9.7 29.8 1,618 Residence Urban 29.2 8.9 38.1 555 Rural 18.1 10.0 28.1 2,664 Region Hhohho 24.9 10.1 35.0 835 Manzini 22.2 8.3 30.5 966 Shiselweni 16.5 11.5 28.0 779 Lubombo 14.5 9.8 24.3 638 Wealth quintile Lowest 8.9 8.9 17.7 745 Second 13.9 10.8 24.7 754 Middle 21.4 10.1 31.5 627 Fourth 22.1 10.8 32.8 601 Highest 41.8 8.4 50.2 491 Total 20.0 9.8 29.8 3,219 Characteristics of Respondents | 27 CHARACTERISTICS OF RESPONDENTS 3 Henry Ginindza and Petronella Mamba This chapter provides a profile of the respondents who were interviewed in the 2006-07 Swaziland DHS, including youth age 12-14, men and women age 15-49, and older adults age 50 and older. First, information is presented on a number of basic characteristics including age at the time of the survey, religion, marital status, residence, education, literacy, and media access. Then, the chapter explores adults’ employment status, occupation, and earnings. An analysis of these variables provides the socioeconomic context within which demographic and reproductive health issues are examined in the subsequent chapters. 3.1 CHARACTERISTICS OF SURVEY RESPONDENTS Table 3.1.1 presents the distribution of women and men age 15-49 by age, religion, marital status, urban-rural residence, region, education level, and wealth quintile. For both sexes, the proportion in each age group tends to decrease with increasing age. A high proportion of the respondents are young adults age 15-24 (47 percent of women and 53 percent of men). The majority of respondents are Zionists (37 percent each of women and men). One in four women and 18 percent of men are Protestants. A large proportion of men reported no religion. Half of women have never been married, compared with 66 percent of men. Thirty-two percent of women and 23 percent of men are currently married. Ten percent of women and 6 percent of men are living together with their partners, 3 percent of women and 4 percent of men are divorced or separated, and 6 percent of women and 1 percent of men are widowed. Women and men are distributed almost equally by urban-rural residence and across regions. Women are as likely as men to have no education. However, men are slightly more likely than women to have reached higher education levels. For instance, 21 percent of men have attended high school compared with 18 percent of women. Overall, slightly over 50 percent of the population have attained secondary education and above. Table 3.1.2 shows the background characteristics of the 465 girls and 409 boys age 12-14 interviewed in the survey. The boys and girls are almost evenly distributed by age. About 88 percent of these boys and girls live in rural areas. Between 24-29 percent of boys and girls live in the Hhohho, Manzini, or Shiselweni regions, and only 20-21 percent live in the Lubombo region. The distribution of the respondents by education is skewed, because of the small numbers of boys and girls age 12-14 who have no education and who have secondary or higher education. While girls age 12-14 are less likely than boys to be in lower primary school (28 percent of girls compared with 43 percent of boys), they are more likely than boys of the same age to attend higher primary school (61 percent of girls compared with 50 percent of boys). Table 3.1.3 shows the distribution of women and men age 50 and older who were interviewed in the survey. It is interesting to note that 59 percent of women and 56 percent of men are 60 years or older. As in the case of younger populations, 87 percent of women and 82 percent of men live in the rural areas. Also, three in ten each of women and men are in Manzini, between 22-29 percent of women and men are in Hhohho and Shiselweni, and 19 percent or less are in Lubombo. Education among older adults is limited; 41 percent of women and 38 percent of men in this age group have had no education. 28 | Characteristics of Respondents Table 3.1.1 Background characteristics of respondents: Women and men 15-49 Percent distribution of women and men age 15-49 by selected background characteristics, Swaziland 2006-07 Women Men Background characteristic Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number Age 15-19 25.5 1,274 1,265 31.8 1,323 1,257 20-24 21.0 1,046 1,027 21.3 886 878 25-29 14.6 729 732 15.0 624 639 30-34 12.3 616 630 10.4 431 449 35-39 10.1 503 508 8.8 367 395 40-44 8.8 438 442 6.5 269 284 45-49 7.7 383 383 6.2 256 254 Religion Charismatic 17.8 887 898 10.8 448 443 Protestant 24.1 1,203 1,191 17.8 741 752 Roman Catholic 4.7 232 244 5.1 212 205 Pentecostal 3.8 191 195 2.4 99 103 Zionist 36.7 1,832 1,820 37.1 1,542 1,533 Apostolic Sect 7.3 365 356 6.1 252 248 None 4.0 197 203 18.7 775 780 Other 1.5 76 77 2.0 86 90 Marital status Never married 49.9 2,487 2,486 65.8 2,734 2,680 Married 31.9 1,589 1,581 23.3 970 1,032 Living together 9.5 473 488 6.0 249 249 Divorced/separated 3.2 161 159 3.5 145 141 Widowed 5.6 277 273 1.4 58 54 Residence Urban 26.7 1,330 1,544 28.4 1,181 1,441 Rural 73.3 3,657 3,443 71.6 2,975 2,715 Region Hhohho 26.9 1,340 1,263 26.5 1,099 1,019 Manzini 33.0 1,647 1,475 32.5 1,349 1,186 Shiselweni 20.7 1,033 1,083 20.3 843 838 Lubombo 19.4 966 1,166 20.8 865 1,113 Education No education 8.1 402 413 7.6 316 332 Lower primary 7.2 360 374 11.3 470 457 Higher primary 25.4 1,268 1,262 23.6 980 971 Secondary 33.9 1,693 1,647 28.6 1,191 1,180 High school 17.9 894 894 20.5 852 838 Tertiary 7.4 370 397 8.3 347 378 Wealth quintile Lowest 15.7 785 778 14.5 601 585 Second 17.3 862 857 16.0 665 639 Middle 19.4 968 934 20.6 856 787 Fourth 22.3 1,111 1,059 22.9 953 922 Highest 25.3 1,262 1,359 26.0 1,081 1,223 Total 15-49 100.0 4,987 4,987 100.0 4,156 4,156 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. Total includes three women and two men with information missing on religion. Characteristics of Respondents | 29 Table 3.1.2 Background characteristics of respondents: Youth age 12-14 Percent distribution of girls and boys age 12-14 by selected background characteristics, Swaziland 2006-07 Girls age 12-14 Boys age 12-14 Background characteristic Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number Age 12 31.3 146 152 33.8 138 139 13 33.1 154 146 31.9 130 130 14 35.5 165 161 34.3 140 142 Residence Urban 12.2 57 75 11.7 48 55 Rural 87.8 408 384 88.3 361 356 Region Hhohho 25.2 117 113 27.1 111 110 Manzini 29.5 137 130 28.7 117 112 Shiselweni 24.8 116 110 23.9 98 95 Lubombo 20.5 95 106 20.2 83 94 Education No education * 7 9 * 13 13 Lower primary 27.7 129 125 42.6 174 176 Higher primary 60.6 282 280 49.5 202 201 Secondary + (10.2) 47 45 * 20 21 Total 12-14 100.0 465 459 100.0 409 411 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that an estimate is based on fewer than 25 unweighted cases. Table 3.1.3 Background characteristics of respondents: Older adults age 50+ Percent distribution of women and men age 50 and over by selected background characteristics, Swaziland 2006-07 Women age 50+ Men age 50+ Background characteristic Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number Age 50-54 24.6 164 164 26.0 116 126 55-59 16.8 112 111 18.0 80 87 60+ 58.7 392 386 56.0 249 243 Residence Urban 12.9 86 104 17.7 79 96 Rural 87.1 583 557 82.3 365 360 Region Hhohho 25.0 167 158 28.6 127 125 Manzini 31.3 209 193 30.6 136 133 Shiselweni 27.8 186 186 22.0 98 100 Lubombo 15.9 107 124 18.9 84 98 Education No education 41.4 277 275 38.4 171 175 Lower primary 22.5 151 141 18.1 80 80 Higher primary 22.2 149 147 19.2 85 84 Secondary 9.0 60 60 13.5 60 63 High school + (1.0) 31 37 3.1 47 53 Total 50+ 100.0 669 661 100.0 444 456 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. Figure in parentheses is based on 25-49 unweighted cases. 30 | Characteristics of Respondents 3.2 EDUCATIONAL ATTAINMENT BY BACKGROUND CHARACTERISTICS Tables 3.2.1 and 3.2.2 present an overview of the relationship between the respondent’s level of education and other background characteristics. Table 3.2.1 shows that younger people are better educated than older people; while 19 percent of women age 15-24 have attended high school, the corresponding proportion for women age 45-49 is only 5 percent. Urban women, women in Manzini and Hhohho, and women in the highest wealth quintile are better educated than other women. Among urban women, 24 percent have completed secondary education compared with 16 percent of their rural counterparts. Also, women in urban areas are four times more likely to have attained tertiary education (16 percent) compared with women in rural areas (4 percent). Across regions, women in Lubombo are the least educated. Table 3.2.1 Educational attainment by background characteristics: Women 15-49 Percent distribution of women age 15-49 by highest level of schooling attended or completed, and median years completed, according to background characteristics, Swaziland 2006-07 Highest educational level Background characteristic No education Lower primary Higher primary Secondary High school Tertiary Total Median years completed Number of women Age 15-24 3.7 6.0 27.4 41.9 18.6 2.4 100.0 7.9 2,320 15-19 2.1 6.5 33.4 45.2 12.3 0.5 100.0 7.5 1,274 20-24 5.6 5.3 20.2 37.9 26.2 4.8 100.0 8.6 1,046 25-29 7.1 5.0 19.7 28.0 28.7 11.5 100.0 9.0 729 30-34 8.5 6.2 23.7 28.8 19.6 13.3 100.0 8.6 616 35-39 10.9 10.7 24.8 27.1 12.9 13.7 100.0 7.8 503 40-44 16.5 11.8 25.1 25.3 10.9 10.4 100.0 6.6 438 45-49 22.4 11.0 28.3 24.2 5.4 8.6 100.0 6.0 383 Residence Urban 5.1 3.9 17.5 32.7 24.4 16.4 100.0 9.3 1,330 Rural 9.2 8.4 28.3 34.4 15.6 4.2 100.0 7.4 3,657 Region Hhohho 8.1 5.7 23.4 35.9 17.4 9.5 100.0 8.3 1,340 Manzini 5.1 5.5 24.5 34.1 21.3 9.5 100.0 8.5 1,647 Shiselweni 5.9 7.0 28.3 36.2 18.5 4.2 100.0 7.8 1,033 Lubombo 15.4 12.5 26.8 28.6 12.4 4.4 100.0 6.5 966 Wealth quintile Lowest 18.9 18.1 33.5 22.5 7.0 0.0 100.0 5.4 785 Second 13.2 10.2 33.3 32.0 10.8 0.5 100.0 6.5 862 Middle 5.8 6.8 31.7 38.4 15.9 1.3 100.0 7.5 968 Fourth 4.6 3.7 21.8 42.4 23.9 3.7 100.0 8.7 1,111 Highest 2.6 1.8 13.4 31.5 26.0 24.7 100.0 10.1 1,262 Total 8.1 7.2 25.4 33.9 17.9 7.4 100.0 8.0 4,987 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. Table 3.2.2 shows that the pattern for men is similar to that for women. For example, urban men, men in Manzini and Hhohho, and men in the highest wealth quintile are also better educated than other men. The median years of schooling completed for women and men is also similar—8.0 years for women and 7.9 years for men. Characteristics of Respondents | 31 Table 3.2.2 Educational attainment by background characteristics: Men 15-49 Percent distribution of women age 15-49 by highest level of schooling attended or completed, and median years completed, according to background characteristics, Swaziland 2006-07 Highest educational level Background characteristic No education Lower primary Higher primary Secondary High school Tertiary Total Median years completed Number of men Age 15-24 3.7 12.7 28.5 34.4 18.4 2.4 100.0 7.5 2,209 15-19 3.2 14.5 34.8 37.3 10.0 0.2 100.0 6.8 1,323 20-24 4.3 9.9 19.1 30.0 30.9 5.7 100.0 8.7 886 25-29 8.2 6.8 16.9 27.6 29.8 10.7 100.0 9.0 624 30-34 8.2 12.0 16.8 16.3 32.6 14.1 100.0 9.6 431 35-39 12.2 11.6 19.3 22.1 14.9 20.0 100.0 8.2 367 40-44 18.1 8.4 19.5 20.6 12.8 20.7 100.0 8.1 269 45-49 21.9 11.8 19.4 20.2 12.2 14.4 100.0 6.6 256 Residence Urban 6.2 6.2 15.1 27.1 27.2 18.1 100.0 9.5 1,181 Rural 8.2 13.3 27.0 29.3 17.8 4.5 100.0 7.2 2,975 Region Hhohho 8.5 8.6 20.6 30.9 21.8 9.6 100.0 8.4 1,099 Manzini 5.2 8.2 23.0 28.0 24.2 11.4 100.0 8.5 1,349 Shiselweni 5.6 14.6 25.9 31.2 19.0 3.6 100.0 7.4 843 Lubombo 12.2 16.5 25.9 24.3 14.5 6.6 100.0 6.4 865 Wealth quintile Lowest 17.5 25.1 32.7 18.3 6.4 0.0 100.0 4.7 601 Second 8.7 14.2 31.7 31.1 14.0 0.3 100.0 6.6 665 Middle 6.0 12.9 29.7 32.9 17.1 1.4 100.0 7.1 856 Fourth 5.9 8.1 21.4 33.6 25.2 5.7 100.0 8.4 953 Highest 4.2 3.3 10.7 25.2 30.8 25.7 100.0 11.0 1,081 Total 15-49 7.6 11.3 23.6 28.6 20.5 8.3 100.0 7.9 4,156 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. 3.3 LITERACY The ability to read and write is an important personal asset, allowing individuals increased opportunities in life. Knowing the distribution of the literate population can facilitate programme managers, especially in health and family planning, in designing messages to reach women and men. The 2006-07 SDHS assessed the ability to read among women and men who had never been to school or who had attended only the primary level by asking respondents to read a simple and short sentence.2 Tables 3.3.1 and 3.3.2 show the percent distribution of female and male respondents, by level of literacy and percent literate according to background characteristics. Overall, 91 percent of Swazi women age 15-49 are literate. Six in ten women have attended secondary or higher education and among those who have no education or had attended or are attending primary school, 32 percent can read a whole sentence or part of a sentence. Younger women, women in urban areas, and those who are in the highest wealth quintile are more likely to be literate than other women. The literacy rate is lowest among women in Lubombo. Table 3.3.2 shows that the pattern for men is similar to that for women. For example, literacy rates are higher for younger men, urban men, and men in the highest wealth quintile than for other men. Across regions, men in Lubombo also have the lowest literacy rate. 2 The sentence is: “The radio programme on health issues is brought to you by Muhle Dlamini.” 32 | Characteristics of Respondents Table 3.3.2 Literacy: Men 15-49 Percent distribution of men age 15-49 by level of schooling attended and level of literacy, and percentage literate, according to background characteristics, Swaziland 2006-07 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all No card with required language Blind/ visually impaired Missing Total Percentage literate1 Number Age 15-19 47.5 34.5 8.6 8.9 0.1 0.1 0.4 100.0 90.5 1,323 20-24 66.6 17.3 7.3 8.0 0.3 0.0 0.4 100.0 91.3 886 25-29 68.1 17.2 7.0 7.0 0.3 0.1 0.2 100.0 92.3 624 30-34 63.0 19.4 7.9 9.3 0.0 0.2 0.3 100.0 90.2 431 35-39 57.0 22.6 10.0 10.0 0.0 0.0 0.4 100.0 89.6 367 40-44 54.0 19.8 8.7 17.0 0.5 0.0 0.0 100.0 82.5 269 45-49 46.9 21.5 11.1 19.1 0.0 0.8 0.6 100.0 79.5 256 Residence Urban 72.5 15.3 5.3 6.6 0.2 0.0 0.1 100.0 93.2 1,181 Rural 51.5 27.3 9.5 10.9 0.2 0.2 0.5 100.0 88.3 2,975 Region Hhohho 62.3 21.0 8.3 7.8 0.1 0.2 0.3 100.0 91.6 1,099 Manzini 63.6 20.8 6.3 8.5 0.3 0.2 0.3 100.0 90.7 1,349 Shiselweni 53.8 26.5 10.3 9.0 0.0 0.0 0.3 100.0 90.6 843 Lubombo 45.4 29.7 9.4 14.7 0.2 0.0 0.5 100.0 84.5 865 Wealth quintile Lowest 24.7 39.8 12.7 21.6 0.1 0.1 0.8 100.0 77.3 601 Second 45.4 32.3 9.6 11.9 0.2 0.3 0.4 100.0 87.3 665 Middle 51.4 28.0 10.2 9.9 0.2 0.1 0.1 100.0 89.6 856 Fourth 64.6 19.0 8.0 7.5 0.2 0.1 0.7 100.0 91.5 953 Highest 81.7 10.8 3.8 3.5 0.1 0.0 0.0 100.0 96.4 1,081 Total 15-49 57.5 23.9 8.3 9.7 0.2 0.1 0.4 100.0 89.6 4,156 1 Refers to men who attended secondary school or higher and men who can read a whole sentence or part of a sentence Table 3.3.1 Literacy: Women 15-49 Percent distribution of women age 15-49 by level of schooling attended and level of literacy, and percentage literate, according to background characteristics, Swaziland 2006-07 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all Blind/ visually impaired Missing Total Percentage literate1 Number Age 15-19 58.0 33.0 4.7 3.4 0.0 0.9 100.0 95.7 1,274 20-24 68.9 20.0 4.3 6.4 0.1 0.4 100.0 93.2 1,046 25-29 68.2 18.2 5.5 7.5 0.2 0.5 100.0 91.9 729 30-34 61.6 24.1 6.6 6.7 0.1 0.9 100.0 92.3 616 35-39 53.7 26.4 9.6 9.8 0.2 0.3 100.0 89.7 503 40-44 46.6 29.3 8.2 15.5 0.2 0.2 100.0 84.1 438 45-49 38.3 30.5 10.5 18.3 1.9 0.4 100.0 79.4 383 Residence Urban 73.5 16.2 4.4 4.9 0.0 1.1 100.0 94.0 1,330 Rural 54.1 29.3 6.9 9.0 0.3 0.4 100.0 90.3 3,657 Region Hhohho 62.8 23.0 6.8 7.0 0.0 0.4 100.0 92.6 1,340 Manzini 64.9 23.6 5.0 5.3 0.2 1.1 100.0 93.5 1,647 Shiselweni 58.9 28.0 5.9 6.1 0.8 0.2 100.0 92.8 1,033 Lubombo 45.3 31.2 7.8 15.3 0.2 0.2 100.0 84.3 966 Wealth quintile Lowest 29.5 36.3 12.8 20.9 0.1 0.3 100.0 78.6 785 Second 43.3 35.8 8.1 11.5 0.9 0.4 100.0 87.2 862 Middle 55.6 31.2 6.3 5.7 0.3 0.8 100.0 93.1 968 Fourth 70.0 21.6 3.7 4.3 0.1 0.3 100.0 95.3 1,111 Highest 82.2 12.0 2.9 2.1 0.0 0.8 100.0 97.1 1,262 Total 59.3 25.8 6.2 7.9 0.3 0.6 100.0 91.3 4,987 1 Refers to women who attended secondary school or higher and women who can read a whole sentence or part of a sentence Characteristics of Respondents | 33 The 2006-07 SDHS also assessed literacy among youth age 12-14 and older adults age 50 and older. Table 3.3.3 shows the extent of literacy among children age 12-14. Overall, 97 percent of girls age 12-14 and 88 percent of boys of the same age are literate. For both girls and boys, there are virtually no differences in literacy level across age, urban-rural residence, and region. Table 3.3.3 Literacy: Youth age 12-14 Percent distribution of girls and boys age 12-14 by level of schooling attended and level of literacy, and percentage literate, according to background characteristics, Swaziland 2006-07 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all No card with required language Missing Total Percentage literate1 Number GIRLS AGE 12-14 Age 12 0.6 75.0 21.3 3.1 0.0 0.0 100.0 96.9 146 13 7.8 79.9 8.7 3.0 0.0 0.6 100.0 96.4 154 14 20.8 68.8 7.2 2.8 0.0 0.5 100.0 96.8 165 Residence Urban 21.2 74.0 2.9 0.6 0.0 1.3 100.0 98.1 57 Rural 8.6 74.5 13.4 3.3 0.0 0.2 100.0 96.5 408 Region Hhohho 7.8 79.1 9.8 3.4 0.0 0.0 100.0 96.6 117 Manzini 13.4 71.7 13.4 0.9 0.0 0.5 100.0 98.5 137 Shiselweni 10.4 69.8 17.6 2.2 0.0 0.0 100.0 97.8 116 Lubombo 8.1 78.2 6.4 6.3 0.0 1.0 100.0 92.7 95 Total 12-14 10.2 74.4 12.1 3.0 0.0 0.4 100.0 96.7 465 BOYS AGE 12-14 Age 12 0.0 65.5 16.1 17.8 0.6 0.0 100.0 81.6 138 13 3.9 74.5 14.5 7.1 0.0 0.0 100.0 92.9 130 14 10.5 63.7 14.5 10.6 0.0 0.7 100.0 88.7 140 Residence Urban 8.6 66.7 11.8 12.9 0.0 0.0 100.0 87.1 48 Rural 4.4 67.9 15.5 11.8 0.2 0.3 100.0 87.7 361 Region Hhohho 2.2 67.4 16.1 13.4 0.0 0.8 100.0 85.8 111 Manzini 3.9 71.9 13.2 11.1 0.0 0.0 100.0 88.9 117 Shiselweni 8.1 67.4 18.5 5.1 0.9 0.0 100.0 94.0 98 Lubombo 5.9 62.7 12.1 19.2 0.0 0.0 100.0 80.8 83 Total 12-14 4.9 67.7 15.0 11.9 0.2 0.2 100.0 87.6 409 1 Refers to girls and boys who attended secondary school or higher and those who can read a whole sentence or part of a sentence Table 3.3.4 shows the extent of literacy among older adults age 50 and older. In this age group, men are more likely than women to be literate (61 percent compared with 51 percent, respectively). As is the case with the population under age 50, among older adults age 50 and older, those who are younger, living in urban areas, and in the highest wealth quintile are more likely to be literate than other respondents. The literacy rate is lowest among women age 50 and older in Lubombo (36 percent) and men age 50 and older in Shiselweni (55 percent). 34 | Characteristics of Respondents Table 3.3.4 Literacy: Older adults age 50+ Percent distribution of women and men age 50+ by level of schooling attended and level of literacy, and percentage literate, according to background characteristics, Swaziland 2006-07 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all No card with required language Blind/ visually impaired Missing Total Percentage literate1 Number WOMEN AGE 50+ Age 50-54 22.1 30.1 15.6 26.9 0.8 3.9 0.6 100.0 67.9 164 55-59 26.2 31.6 13.4 24.2 0.0 4.5 0.0 100.0 71.3 112 60+ 6.5 15.4 16.4 49.5 0.0 12.2 0.0 100.0 38.3 392 Residence Urban 34.8 34.7 12.5 14.0 0.0 3.9 0.0 100.0 82.1 86 Rural 10.5 19.8 16.2 43.5 0.2 9.6 0.2 100.0 46.5 583 Region Hhohho 16.9 26.1 12.5 36.5 0.7 6.7 0.5 100.0 55.5 167 Manzini 18.3 21.1 21.2 33.0 0.0 6.5 0.0 100.0 60.5 209 Shiselweni 9.0 21.4 15.1 39.9 0.0 14.7 0.0 100.0 45.5 186 Lubombo 7.5 16.8 11.2 57.6 0.0 6.9 0.0 100.0 35.5 107 Total 50+ 13.6 21.7 15.7 39.7 0.2 8.9 0.1 100.0 51.1 669 MEN AGE 50+ Age 50-54 43.4 17.1 12.6 20.3 0.9 4.9 0.8 100.0 73.1 116 55-59 31.1 27.9 17.0 17.3 0.0 6.7 0.0 100.0 76.0 80 60+ 12.8 21.3 16.6 40.6 0.0 8.4 0.3 100.0 50.7 249 Residence Urban 51.9 22.8 13.3 7.9 0.0 4.1 0.0 100.0 88.0 79 Rural 18.0 21.1 16.1 36.2 0.3 7.9 0.4 100.0 55.2 365 Region Hhohho 25.2 17.6 17.2 34.2 0.0 5.2 0.7 100.0 59.9 127 Manzini 34.0 18.3 14.0 22.9 0.8 10.1 0.0 100.0 66.2 136 Shiselweni 14.6 23.3 16.8 36.7 0.0 8.6 0.0 100.0 54.7 98 Lubombo 17.3 29.9 14.6 33.3 0.0 4.0 0.9 100.0 61.8 84 Total 50+ 24.0 21.4 15.6 31.2 0.2 7.2 0.4 100.0 61.0 444 1 Refers to women and men who attended secondary school or higher and those who can read a whole sentence or part of a sentence 3.4 ACCESS TO MASS MEDIA Information access is essential in increasing people’s knowledge and awareness of what is taking place around them, which may eventually affect their perceptions and behaviour. It is important to know the types of persons who are more or less likely to be reached by the media for purposes of planning programmes intended to spread information about health and family planning. In the survey, exposure to the media was assessed by asking how often a respondent reads a newspaper, watches television, or listens to a radio. Tables 3.4.1 to 3.4.4 show the percentage of females, males, youth, and older adults who were exposed to different types of media at least once a week by age, urban-rural residence, region, level of education, and wealth quintile. Characteristics of Respondents | 35 Table 3.4.1 Exposure to mass media: Women 15-49 Percentage of women age 15-49 who are exposed to specific media on a weekly basis, by background characteristics, Swaziland 2006-07 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media at least once a week No media at least once a week Number Age 15-19 60.8 37.9 77.1 25.6 10.6 1,274 20-24 51.5 40.0 73.2 22.5 14.7 1,046 25-29 50.9 38.4 74.1 25.6 14.7 729 30-34 51.6 41.1 75.6 24.1 13.1 616 35-39 46.5 36.8 71.8 21.9 17.1 503 40-44 46.0 32.7 74.7 19.9 15.8 438 45-49 36.7 28.1 71.2 15.7 22.2 383 Residence Urban 63.9 57.4 75.5 36.0 9.3 1,330 Rural 47.2 30.3 74.1 18.4 16.2 3,657 Region Hhohho 57.1 42.2 77.2 27.2 11.6 1,340 Manzini 57.7 41.9 75.0 25.9 12.2 1,647 Shiselweni 42.4 30.4 73.1 18.2 17.3 1,033 Lubombo 43.7 31.0 71.2 18.0 18.8 966 Education No education 4.9 12.2 58.8 1.0 38.6 402 Lower primary 20.3 17.1 67.8 5.6 26.1 360 Higher primary 36.3 24.0 70.7 10.2 19.7 1,268 Secondary 59.5 40.2 79.3 26.0 9.8 1,693 High school 75.7 51.6 79.0 36.7 5.5 894 Tertiary 91.9 85.1 78.1 62.4 0.6 370 Wealth quintile Lowest 25.1 5.1 55.7 1.5 36.3 785 Second 35.1 9.5 69.8 4.3 21.2 862 Middle 48.4 18.3 75.8 9.7 14.2 968 Fourth 58.9 46.5 83.2 28.2 7.6 1,111 Highest 75.6 83.6 80.7 55.2 2.2 1,262 Total 51.7 37.5 74.5 23.1 14.4 4,987 Tables 3.4.1 and 3.4.2 show that in general, men are more exposed to mass media than women. The most popular mass media is radio broadcast, with 75 percent of women and 83 percent of men listening to the radio at least once a week. The next most accessed media is newspaper, with 52 percent of women and 61 percent of men reading a newspaper at least once a week. Thirty-eight percent of women and 45 percent of men watch television at least once a week. Overall, only 23 percent of women and 31 percent of men have access to all three types of media. Further, 14 percent of women and 7 percent of men have no contact with any media at any time of the week. Younger women and men, those who live in urban areas, and those who live in Hhohho and Manzini are more likely to be exposed to mass media than other respondents. Access to mass media increases with education and wealth. 36 | Characteristics of Respondents Table 3.4.2 Exposure to mass media: Men 15-49 Percentage of men age 15-49 who are exposed to specific media on a weekly basis, by background characteristics, Swaziland 2006-07 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media at least once a week No media at least once a week Number Age 15-19 59.0 42.5 82.5 26.5 6.4 1,323 20-24 63.3 46.6 81.9 30.7 6.7 886 25-29 66.4 46.7 85.6 32.0 5.1 624 30-34 63.5 51.8 86.0 39.0 4.8 431 35-39 58.4 43.1 82.7 30.9 8.9 367 40-44 55.8 44.3 83.0 33.0 10.2 269 45-49 53.2 40.9 78.9 31.9 15.0 256 Residence Urban 74.8 60.8 81.7 45.1 4.4 1,181 Rural 55.4 38.8 83.6 25.0 8.2 2,975 Region Hhohho 65.4 45.6 83.4 33.2 6.0 1,099 Manzini 67.2 49.6 82.9 35.0 6.5 1,349 Shiselweni 50.9 40.9 85.0 25.5 7.9 843 Lubombo 55.0 41.2 80.9 25.8 8.6 865 Education No education 8.0 21.9 74.7 3.1 21.7 316 Lower primary 21.5 25.7 80.3 7.5 15.5 470 Higher primary 47.4 32.9 83.1 17.5 9.1 980 Secondary 72.7 46.4 84.9 34.2 4.3 1,191 High school 86.9 60.1 85.8 49.2 1.3 852 Tertiary 95.9 84.9 80.9 66.9 0.7 347 Wealth quintile Lowest 31.9 15.5 77.6 8.2 17.5 601 Second 50.4 24.6 80.4 14.5 9.7 665 Middle 56.0 29.2 86.7 17.3 6.6 856 Fourth 65.5 50.1 84.2 32.6 5.5 953 Highest 83.2 82.2 83.7 62.0 1.5 1,081 Total 15-49 60.9 45.0 83.0 30.7 7.1 4,156 Table 3.4.3 shows the exposure to mass media among girls and boys age 12-14 interviewed in the survey. As in the case for women and men age 15-49, radio is the most popular media (60 percent of girls and 71 percent of boys). Overall, only 14 percent of girls age 12-14 and 15 percent of boys age 12-14 have access to all three types of media. In contrast, 19 percent of girls age 12-14 and 16 percent of boys age 12-14 have no contact with any media at least once a week. The pattern of exposure to mass media for youth age 12-14 is similar to that of women and men age 15-49. Younger girls and boys, those who live in urban areas, and those who live in Hhohho and Manzini are more likely to be exposed to mass media than other respondents. Access to mass media increases with education. Characteristics of Respondents | 37 Table 3.4.3 Exposure to mass media: Youth age 12-14 Percentage of girls and boys age 12-14 who are exposed to specific media on a weekly basis, by background characteristics, Swaziland 2006-07 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media at least once a week No media at least once a week Number GIRLS AGE 12-14 Age 12 48.5 30.5 63.5 14.3 15.3 146 13 48.2 29.0 67.4 11.1 18.3 154 14 47.7 36.1 50.5 14.9 22.1 165 Residence Urban 43.8 63.4 49.3 19.0 13.5 57 Rural 48.7 27.6 61.7 12.7 19.5 408 Region Hhohho 50.2 33.9 63.8 10.6 10.7 117 Manzini 54.0 33.6 55.3 16.5 20.2 137 Shiselweni 48.1 30.3 59.4 13.8 22.0 116 Lubombo 37.1 29.3 63.9 12.2 22.6 95 Education No education * * * * * 7 Lower primary 39.9 17.4 62.7 8.5 24.0 129 Higher primary 60.9 36.9 70.1 18.3 9.4 282 Secondary + (0.0) (46.7) (0.0) (0.0) (53.3) 47 Total 12-14 48.1 32.0 60.2 13.5 18.7 465 BOYS AGE 12-14 Age 12 42.5 31.6 76.8 16.8 14.6 138 13 41.6 32.4 67.8 14.5 19.2 130 14 34.0 37.7 67.5 12.4 15.4 140 Residence Urban (55.3) (71.4) (79.7) (41.9) (8.8) 48 Rural 37.2 29.0 69.6 11.0 17.3 361 Region Hhohho 41.0 26.0 75.3 12.8 16.5 111 Manzini 53.7 36.8 69.4 20.9 13.9 117 Shiselweni 25.7 39.6 67.0 13.7 19.3 98 Lubombo 32.4 33.9 71.0 9.1 16.1 83 Education No education * * * * * 13 Lower primary 35.0 26.1 71.2 8.3 18.7 174 Higher primary 48.8 37.8 78.5 22.3 12.4 202 Secondary + * * * * * 20 Total 12-14 39.3 33.9 70.8 14.6 16.3 409 Note: Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that an estimate is based on fewer than 25 unweighted cases. Table 3.4.4 presents the exposure to mass media among older adults age 50 and older interviewed in the survey. As in the case of younger respondents, radio is the most popular media (60 percent of women and 77 percent of men). Overall, only 9 percent of women and 14 percent of men age 50 and older have access to all three types of media. In addition, 35 percent of women and 20 percent of men age 50 and older have no contact with any media at any time of the week. 38 | Characteristics of Respondents The pattern of exposure to mass media for older adults is similar to that of younger women and men, except there is no clear pattern by age. Women and men who live in urban areas and those who live in Hhohho and Manzini are more likely to be exposed to mass media than other respondents. Access to mass media increases with education. Table 3.4.4 Exposure to mass media: Older adults age 50+ Percentage of women and men age 50+ who are exposed to specific media on a weekly basis, by background characteristics, Swaziland 2006-07 Background characteristic Reads a newspaper at least once a week Watches television at least once a week Listens to the radio at least once a week All three media at least once a week No media at least once a week Number WOMEN AGE 50+ Age 50-54 14.7 14.8 53.4 3.7 40.8 254 55-59 22.2 28.8 63.3 11.9 30.7 233 60+ 23.1 19.4 62.6 12.1 31.3 179 Residence Urban 40.2 53.6 71.1 22.9 14.7 86 Rural 16.6 16.2 57.8 6.9 37.5 583 Region Hhohho 25.1 28.9 64.3 13.9 29.2 167 Manzini 23.0 24.5 64.2 11.0 31.2 209 Shiselweni 12.8 12.5 54.6 3.9 40.0 186 Lubombo 16.7 16.8 51.2 6.0 40.2 107 Education No education 3.7 12.0 50.0 1.5 45.8 277 Lower primary 13.7 10.7 52.2 0.8 42.2 151 Higher primary 25.1 23.4 68.1 6.1 24.1 149 Secondary 61.5 47.9 89.9 41.0 5.6 60 High school + (85.2) (89.7) (81.3) (66.8) 2.0 31 Total 50+ 19.7 21.0 59.5 9.0 34.6 669 MEN AGE 50+ Age 50-54 25.4 19.5 77.5 11.3 19.0 165 55-59 30.2 33.0 75.8 16.1 18.7 174 60+ 30.0 21.0 76.7 13.4 22.4 105 Residence Urban 56.8 52.6 88.5 37.6 4.7 79 Rural 22.3 19.3 74.1 8.6 22.9 365 Region Hhohho 35.2 26.8 83.5 18.3 11.7 127 Manzini 32.3 32.8 80.0 16.4 16.3 136 Shiselweni 17.0 13.5 70.9 4.8 27.4 98 Lubombo 25.1 24.1 67.3 12.6 28.2 84 Education No education 3.0 9.9 67.2 0.6 31.5 171 Lower primary 15.3 9.2 69.8 0.8 25.6 80 Higher primary 36.5 29.2 87.5 15.3 9.7 85 Secondary 57.8 41.5 85.8 21.9 7.8 60 High school + (91.4) (80.4) (90.8) (70.3) (0.0) 47 Total 50+ 28.4 25.2 76.6 13.7 19.7 444 Note: Figures in parentheses are based on 25-49 unweighted cases. Characteristics of Respondents | 39 3.5 EMPLOYMENT Male and female respondents age 15 and older were asked whether they were employed at the time of the survey and if not, whether they were employed in the 12 months preceding the survey. The measurement of employment, however, is difficult. The difficulty arises largely because some work, especially work on family farms, in family businesses, or in the informal sector, is often not perceived as employment, and hence not reported as such. To avoid underestimating respondents’ employment, the 2006-07 SDHS asked respondents several questions to probe for their employment status and to ensure complete coverage of employment in both the formal and informal sectors. Respondents were asked a number of questions to elicit their current employment status and continuity of employment in the 12 months prior to the survey. Employed individuals are those who said that they are currently working (i.e., worked in the past 7 days) and those who worked at any time during the 12 months prior to the survey. Tables 3.5.1 to 3.5.3 show the percent distribution of 2006-07 SDHS adult respondents according to current and recent employment. Table 3.5.1 and Figure 3.1 show that four in ten women age 15-49 are currently employed, 4 percent are not currently employed but worked in the past 12 months, and 53 percent did not work in the past 12 months. The proportion of women who are currently employed varies by age, ranging from 9 percent for women age 15-19 to 57 percent or higher for women age 30 and older. Women who have never been married, have no children, and are living in rural areas are less likely than other women to be employed. Women who are divorced or separated are more likely to be employed than women who have never married or are currently married or living together (62 percent for divorced or separated women, 49 percent for women who are married or living together, and 28 percent for never married women). More than half of women who have three or more living children are employed compared with 20 percent of women who have no living children. Women who reside in urban areas are much more likely to be employed than rural women. In three of the four regions, Hhohho, Manzini, and Lubombo, four in ten women are employed. In Shiselweni, only 27 percent of women are currently employed. Employment of women does not vary much by education and wealth status except if they have tertiary education or are in the highest wealth quintile. The age pattern of employment for men is the same as that for women; older men are more likely than younger men to be employed. However, the pattern of employment for men by marital status is the reverse of that for women; married men, those living with women as spouses, are more likely to be employed than men who are divorced or separated (81 percent compared with 69 percent). The likelihood that a man holds a job increases with the number of living children he has; 33 percent of men who have no living children are currently employed, and at least 74 percent of men who have children are employed. This may be due to the social expectation that mature men provide for their families by working. As in the case of women, men in Hhohho, Manzini, and Lubombo are more likely to be employed than those in Shiselweni (52 percent or higher compared with 38 percent). The relationship between employment and education for men is U-shaped; high among men with the least education, declining with increasing education to reach the lowest for men with secondary education, and increasing to the highest proportion for men with tertiary education. The likelihood of men being employed increases with wealth; 36 percent of men in the lowest wealth quintile are employed compared with 64 percent of men in the highest wealth quintile. 40 | Characteristics of Respondents Table 3.5.1 Employment status: Women 15-49 Percent distribution of women age 15-49 by employment status, according to background characteristics, Swaziland 2006-07 Employed in the 12 months preceding the survey Background characteristic Currently employed1 Not currently employed Not employed in the 12 months preceding the survey Missing/ don't know Total Number of women Age 15-19 9.1 2.4 83.0 5.5 100.0 1,274 20-24 33.9 5.8 57.1 3.2 100.0 1,046 25-29 48.5 4.9 43.9 2.8 100.0 729 30-34 57.2 5.5 35.5 1.9 100.0 616 35-39 61.6 3.3 32.6 2.6 100.0 503 40-44 58.8 3.9 35.1 2.1 100.0 438 45-49 62.4 3.9 32.0 1.6 100.0 383 Marital status Never married 28.3 3.9 63.9 3.9 100.0 2,487 Married or living together 48.7 4.3 43.7 3.2 100.0 2,062 Divorced/separated/widowed 62.4 4.8 32.3 0.5 100.0 438 Number of living children 0 19.6 4.1 71.4 4.9 100.0 1,601 1-2 45.2 4.7 47.7 2.4 100.0 1,754 3-4 54.9 3.2 39.5 2.3 100.0 887 5+ 52.3 4.2 40.6 2.9 100.0 745 Residence Urban 56.2 2.4 38.9 2.5 100.0 1,330 Rural 33.8 4.8 57.8 3.6 100.0 3,657 Region Hhohho 42.6 3.3 50.0 4.1 100.0 1,340 Manzini 43.7 2.8 50.9 2.6 100.0 1,647 Shiselweni 26.7 4.0 65.7 3.6 100.0 1,033 Lubombo 42.9 8.0 46.0 3.0 100.0 966 Education No education 43.1 5.6 46.8 4.5 100.0 402 Lower primary 36.7 4.8 55.6 2.9 100.0 360 Higher primary 35.6 4.1 57.1 3.2 100.0 1,268 Secondary 34.7 4.0 57.7 3.7 100.0 1,693 High school 40.5 4.9 51.6 3.1 100.0 894 Tertiary 74.7 1.5 22.4 1.3 100.0 370 Wealth quintile Lowest 29.0 5.4 59.8 5.8 100.0 785 Second 28.7 4.6 63.1 3.5 100.0 862 Middle 33.2 4.6 59.9 2.4 100.0 968 Fourth 42.0 4.5 51.0 2.5 100.0 1,111 Highest 57.0 2.5 37.5 3.0 100.0 1,262 Total 15-49 39.8 4.2 52.8 3.3 100.0 4,987 1 "Currently employed" is defined as having done work in the past seven days. Includes persons who did not work in the past seven days but who are regularly employed and were absent from work for leave, illness, vacation, or any other such reason. Characteristics of Respondents | 41 Comparisons between women and men show that women in general are less likely to be employed in the past 12 months than men (44 percent of men compared with 53 percent). The same is true for current employment; 40 percent for women and 50 percent for men. Employment information on older adults age 50 and over is presented in Table 3.5.3. As expected, men are more likely to be employed than women (29 percent and 20 percent, respectively). For both women and men, employment decreases with age. While in general urban residents are more likely to be employed than rural residents, the difference for men is much more significant than for women. For men, the proportion working is 72 percent in urban areas and 20 percent in rural areas. For women, these proportions are 38 percent and 18 percent, respectively. For both men and women over 50 years old, the level of employment increases with an improvement in the level of education. Curently employed 40% Not currently employed 4% Not employed in the 12 53% Missing/ don't know 3% Figure 3.1 Women’s Employment Status (past 12 months) SDHS 2006-07 Not employed in the 12 months preceding the survey 53% 42 | Characteristics of Respondents Table 3.5.2 Employment status: Men 15-49 Percent distribution of men age 15-49 by employment status, according to background characteristics, Swaziland 2006-07 Employed in the 12 months preceding the survey Background characteristic Currently employed1 Not currently employed Not employed in the 12 months preceding the survey Total Number of men Age 15-19 16.6 3.2 80.2 100.0 1,323 20-24 45.3 8.2 46.5 100.0 886 25-29 72.4 7.0 20.6 100.0 624 30-34 77.2 6.3 16.5 100.0 431 35-39 81.0 3.7 15.3 100.0 367 40-44 79.2 3.9 16.9 100.0 269 45-49 70.5 6.3 23.2 100.0 256 Marital status Never married 35.3 5.6 59.1 100.0 2,734 Married or living together 81.1 5.2 13.7 100.0 1,219 Divorced/separated/widowed 69.4 5.3 25.4 100.0 203 Number of living children 0 32.7 5.3 61.9 100.0 2,500 1-2 77.7 6.0 16.3 100.0 835 3-4 73.8 5.9 20.3 100.0 424 5+ 79.7 4.3 15.9 100.0 397 Residence Urban 69.6 3.2 27.2 100.0 1,181 Rural 42.8 6.3 50.8 100.0 2,975 Region Hhohho 55.3 5.2 39.5 100.0 1,099 Manzini 51.6 4.9 43.5 100.0 1,349 Shiselweni 38.4 4.6 57.0 100.0 843 Lubombo 54.1 7.4 38.4 100.0 865 Education No education 64.7 7.0 28.3 100.0 316 Lower primary 51.0 7.4 41.6 100.0 470 Higher primary 43.2 5.9 50.9 100.0 980 Secondary 41.0 4.5 54.6 100.0 1,191 High school 54.2 5.9 40.0 100.0 852 Tertiary 80.5 2.4 17.2 100.0 347 Wealth quintile Lowest 36.1 8.7 55.2 100.0 601 Second 39.0 7.0 54.0 100.0 665 Middle 45.0 6.9 48.1 100.0 856 Fourth 57.0 3.9 39.0 100.0 953 Highest 63.9 2.9 33.2 100.0 1,081 Total 15-49 50.4 5.4 44.1 100.0 4,156 1 "Currently employed" is defined as having done work in the past seven days. Includes persons who did not work in the past seven days but who are regularly employed and were absent from work for leave, illness, vacation, or any other such reason. Characteristics of Respondents | 43 Table 3.5.3 Employment status: Older adults age 50+ Percent distribution of women and men age 50+ by employment status, according to background characteristics, Swaziland 2006-07 Employed in the 12 months preceding the survey Background characteristic Currently employed Not currently employed Not employed in the 12 months preceding the survey Missing/ don't know Total Number WOMEN Age 50-54 35.8 11.7 50.5 2.1 100.0 164 55-59 34.3 14.2 51.5 0.0 100.0 112 60+ 9.9 12.6 76.4 1.1 100.0 392 Residence Urban 37.7 7.9 54.4 0.0 100.0 86 Rural 17.8 13.4 67.5 1.3 100.0 583 Region Hhohho 25.9 11.6 60.7 1.7 100.0 167 Manzini 24.4 12.4 62.9 0.3 100.0 209 Shiselweni 12.6 12.7 74.2 0.5 100.0 186 Lubombo 17.1 14.7 65.1 3.1 100.0 107 Education No education 11.6 12.7 73.8 1.9 100.0 277 Lower primary 15.6 11.6 71.8 1.0 100.0 151 Higher primary 23.6 15.7 60.1 0.6 100.0 149 Secondary 41.8 9.9 48.3 0.0 100.0 60 High school + (64.7) (8.5) (26.8) (0.0) 100.0 31 Total 50+ 20.3 12.7 65.8 1.2 100.0 669 MEN Age 50-54 53.9 15.6 29.2 1.2 100.0 116 55-59 38.5 10.8 49.7 1.1 100.0 80 60+ 15.1 15.9 66.9 2.1 100.0 249 Residence Urban 71.8 5.4 21.6 1.3 100.0 79 Rural 20.3 17.0 61.0 1.8 100.0 365 Region Hhohho 33.1 15.0 49.5 2.5 100.0 127 Manzini 33.3 14.3 51.1 1.2 100.0 136 Shiselweni 15.8 16.1 66.2 2.0 100.0 98 Lubombo 33.3 14.4 51.2 1.0 100.0 84 Education No education 17.9 12.3 67.8 2.1 100.0 171 Lower primary 18.9 27.1 50.1 4.0 100.0 80 Higher primary 27.1 16.8 55.0 1.0 100.0 85 Secondary 46.7 10.9 42.3 0.0 100.0 60 High school + (70.2) (5.3) (24.5) (0.0) 100.0 47 Total 50+ 29.4 14.9 54.0 1.7 100.0 444 Note: Figures in parentheses are based on 25-49 unweighted cases. 1 "Currently employed" is defined as having done work in the past seven days. Includes persons who did not work in the past seven days but who are regularly employed and were absent from work for leave, illness, vacation, or any other such reason. 44 | Characteristics of Respondents 3.6 OCCUPATION Respondents who were currently employed were asked to state their occupation, and the results are presented in Tables 3.6.1 to 3.6.3. There are six occupational categories under which those currently employed are classified. They are: professional, technical, and managerial; clerical; sales and services; skilled manual; unskilled manual; and agricultural. The vast majority of currently employed women work in sales and services (56 percent). The next most popular occupations for women are skilled manual labour (14 percent); professional, technical, and managerial (12 percent); and agriculture (9 percent). Seven percent of women are employed in clerical jobs. Table 3.6.1 Occupation: Women 15-49 Percent distribution of women age 15-49 employed in the 12 months preceding the survey by occupation, according to background characteristics, Swaziland 2006-07 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agriculture Missing Total Number of women Age 15-19 1.5 2.0 69.6 10.5 0.7 8.3 7.6 100.0 147 20-24 5.3 7.8 63.7 14.1 1.6 6.7 0.9 100.0 415 25-29 12.9 7.8 53.7 16.6 1.6 6.7 0.7 100.0 389 30-34 14.3 6.7 53.2 15.9 1.0 8.1 0.8 100.0 386 35-39 16.9 9.1 47.2 14.4 1.4 10.8 0.3 100.0 326 40-44 15.5 4.9 58.8 10.6 0.4 9.6 0.3 100.0 274 45-49 14.9 4.0 52.5 15.6 0.5 11.8 0.8 100.0 254 Marital status Never married 8.1 8.2 60.4 15.2 1.3 5.2 1.7 100.0 803 Married or living together 15.5 5.9 52.8 13.7 1.3 10.0 0.9 100.0 1,094 Divorced/separated/widowed 10.2 4.9 56.8 14.9 0.2 12.4 0.6 100.0 294 Number of living children 0 9.4 6.1 59.2 14.5 1.4 6.9 2.6 100.0 379 1-2 15.1 7.8 53.7 14.5 1.1 6.5 1.2 100.0 875 3-4 13.4 9.2 53.1 14.6 0.7 8.7 0.2 100.0 516 5+ 6.6 1.3 62.0 13.7 1.3 14.4 0.7 100.0 421 Residence Urban 15.5 11.2 50.3 16.8 1.1 4.3 0.9 100.0 780 Rural 10.2 4.1 59.3 13.1 1.1 11.0 1.2 100.0 1,411 Region Hhohho 11.8 9.1 61.9 8.7 0.9 6.7 1.0 100.0 615 Manzini 14.5 6.8 50.6 22.2 1.3 3.6 1.0 100.0 766 Shiselweni 12.8 5.5 51.7 16.9 1.8 11.3 0.0 100.0 317 Lubombo 8.2 3.8 60.3 7.7 0.8 17.0 2.2 100.0 492 Education No education 0.7 0.0 63.6 9.5 2.1 23.4 0.6 100.0 196 Lower primary 0.5 0.0 64.7 14.5 1.2 18.3 0.7 100.0 149 Higher primary 1.1 2.2 66.2 17.0 0.1 12.0 1.4 100.0 504 Secondary 4.4 3.8 66.1 16.4 2.2 5.9 1.3 100.0 654 High school 12.2 11.2 51.8 18.4 0.9 4.0 1.6 100.0 406 Tertiary 63.3 22.6 11.5 2.6 0.0 0.0 0.0 100.0 282 Wealth quintile Lowest 0.5 2.4 60.8 13.4 0.4 21.3 1.2 100.0 270 Second 1.4 2.4 65.4 13.6 0.7 14.9 1.6 100.0 288 Middle 4.2 3.8 60.7 17.4 1.8 9.9 2.2 100.0 366 Fourth 7.4 4.8 56.8 21.0 1.7 6.8 1.4 100.0 517 Highest 27.4 12.3 48.1 9.0 0.8 2.2 0.2 100.0 751 Total 12.1 6.6 56.1 14.4 1.1 8.6 1.1 100.0 2,191 Characteristics of Respondents | 45 Women’s occupations vary with their background characteristics. In general, professional, technical, and managerial jobs attract women with the highest education and in the highest wealth quintile, and work in sales and services attracts younger women and women with lower than tertiary education. Women in Lubombo are more likely than women in other regions to work in agriculture. While women tend to be concentrated in selected types of occupations, men’s choice of work is more varied. Men are almost evenly employed in sales and services and as skilled manual labourers (29 percent and 32 percent, respectively). Seventeen percent of men work in agriculture and 13 percent are employed in professional, technical, and managerial jobs. Men’s occupations also vary with their background characteristics. In general, professional, technical, and managerial jobs attract married men, men in urban areas, men with the highest education, and men in the highest wealth quintile. On the other hand, young men, rural men, men in Lubombo, men with less education, and those in the lower wealth quintiles tend to work in agriculture. Table 3.6.2 Occupation: Men 15-49 Percent distribution of men age 15-49 employed in the 12 months preceding the survey by occupation, according to background characteristics, Swaziland 2006-07 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agriculture Missing Total Number of men Age 15-19 0.0 1.2 45.4 16.3 8.1 28.3 0.6 100.0 262 20-24 4.6 2.5 32.2 32.3 7.3 20.2 0.8 100.0 474 25-29 11.2 2.5 28.4 37.8 4.4 14.7 1.0 100.0 495 30-34 18.6 4.7 25.1 35.5 4.6 10.7 0.9 100.0 360 35-39 20.5 3.3 23.6 32.2 4.2 16.2 0.0 100.0 311 40-44 24.4 4.5 21.8 31.3 2.7 15.3 0.0 100.0 224 45-49 18.1 5.6 20.4 35.7 5.8 14.4 0.0 100.0 197 Marital status Never married 7.1 2.6 33.8 31.6 5.8 18.6 0.6 100.0 1,118 Married or living together 20.1 4.0 22.5 32.5 4.9 15.5 0.5 100.0 1,052 Divorced/separated/widowed 4.8 3.1 33.5 37.1 5.3 15.3 0.8 100.0 152 Number of living children 0 6.7 2.0 33.6 30.9 5.9 20.2 0.7 100.0 951 1-2 17.9 3.6 25.8 33.5 4.7 13.5 0.9 100.0 700 3-4 19.0 4.6 22.1 31.5 6.9 15.9 0.0 100.0 337 5+ 13.4 4.7 26.9 34.8 3.6 16.2 0.3 100.0 334 Residence Urban 18.4 4.8 26.1 31.8 4.3 14.2 0.4 100.0 860 Rural 9.6 2.3 30.1 32.6 6.0 18.6 0.7 100.0 1,462 Region Hhohho 14.7 4.7 28.1 34.0 5.3 12.9 0.3 100.0 665 Manzini 15.1 2.9 29.4 36.7 5.4 9.8 0.6 100.0 762 Shiselweni 10.0 2.6 28.4 26.5 9.3 22.9 0.4 100.0 363 Lubombo 9.1 2.5 28.4 28.0 2.7 28.4 1.0 100.0 532 Education No education 0.5 0.0 34.2 28.8 5.3 30.8 0.4 100.0 227 Lower primary 2.2 0.9 33.5 22.4 7.5 33.6 0.0 100.0 275 Higher primary 2.9 0.9 31.1 37.8 8.0 18.7 0.6 100.0 481 Secondary 5.1 2.6 28.2 39.9 5.8 17.3 1.0 100.0 541 High school 13.0 7.2 31.0 35.2 4.2 8.5 0.9 100.0 511 Tertiary 63.5 6.1 12.1 16.3 0.3 1.7 0.0 100.0 287 Wealth quintile Lowest 2.2 0.8 30.6 30.0 5.9 28.3 2.2 100.0 269 Second 2.5 2.3 29.0 34.3 8.7 22.8 0.3 100.0 306 Middle 5.2 1.9 32.3 34.3 7.1 18.8 0.4 100.0 444 Fourth 6.7 2.9 25.7 40.6 5.6 18.1 0.4 100.0 580 Highest 30.8 5.7 27.8 24.5 2.5 8.3 0.4 100.0 722 Total 15-49 12.8 3.2 28.6 32.3 5.4 17.0 0.6 100.0 2,322 46 | Characteristics of Respondents Table 3.6.3 shows the percent distribution of 2006-07 SDHS respondents age 50 and over according to current and recent employment. Analysis of the data is limited due to the small number of women and men age 50 years and over who were employed at the time of the survey. The most popular type of work for women age 50 years and over is in sales and services (54 percent), while for men it is working as a skilled manual labourer (33 percent). Table 3.6.3 Occupation: Older adults age 50+ Percent distribution of women and men age 50+ employed in the 12 months preceding the survey by occupation, according to background characteristics, Swaziland 2006-07 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agriculture Missing Total Number of men WOMEN Age 50-54 12.4 7.3 53.1 13.2 0.0 14.0 0.0 100.0 78 55-59 15.9 8.0 54.1 11.8 0.0 10.2 0.0 100.0 54 60+ 1.3 1.4 54.4 26.0 0.0 13.6 3.4 100.0 88 Residence Urban (13.9) (19.7) (44.4) (15.0) (0.0) (7.0) (0.0) 100.0 39 Rural 7.7 1.9 55.9 18.7 0.0 14.2 1.7 100.0 181 Region Hhohho 14.3 6.0 50.1 16.7 0.0 12.9 0.0 100.0 63 Manzini 6.7 7.7 61.2 16.3 0.0 8.1 0.0 100.0 77 Shiselweni (3.9) (3.4) (52.3) (26.3) (0.0) (14.1) (0.0) 100.0 47 Lubombo (10.1) (0.0) (46.3) (12.7) (0.0) (22.0) (8.8) 100.0 34 Education No education 0.0 0.0 50.8 26.3 0.0 18.5 4.5 100.0 67 Lower primary (1.4) (0.0) (72.6) (11.0) (0.0) (15.0) (0.0) 100.0 41 Higher primary 3.1 0.8 61.8 22.1 0.0 12.3 0.0 100.0 58 Secondary 31.6 20.1 34.8 8.6 0.0 4.9 0.0 100.0 54 High school + * * * * * * * 100.0 23 Total 50+ 8.8 5.1 53.9 18.0 0.0 12.9 1.4 100.0 221 MEN Age 50-54 13.7 5.3 27.4 34.7 3.0 15.9 0.0 100.0 80 55-59 (11.8) (8.6) (23.2) (42.6) (2.9) (10.9) (0.0) 100.0 39 60+ 10.0 0.0 25.5 27.3 1.8 33.8 1.5 100.0 77 Residence Urban 19.2 4.7 21.9 43.6 2.2 8.3 0.0 100.0 61 Rural 8.6 3.5 27.6 28.8 2.7 28.0 0.8 100.0 136 Region Hhohho 10.4 4.4 25.9 34.8 4.5 20.0 0.0 100.0 61 Manzini 20.0 1.4 27.3 37.7 0.0 11.8 1.8 100.0 65 Shiselweni (3.2) (5.0) (26.5) (26.3) (7.0) (32.1) (0.0) 100.0 31 Lubombo (7.8) (6.4) (23.0) (29.7) (0.0) (33.1) (0.0) 100.0 40 Education No education 3.0 0.0 35.4 25.9 4.8 30.9 0.0 100.0 51 Lower primary (3.2) (0.0) (28.3) (37.1) (3.1) (28.3) (0.0) 100.0 37 Higher primary (11.3) (0.0) (23.7) (44.9) (0.0) (20.1) (0.0) 100.0 37 Secondary 23.6 11.0 17.7 31.1 1.9 13.1 1.6 100.0 70 High school + (44.6) (14.3) (10.9) (17.0) (0.0) (9.5) (0.0) 100.0 36 Total 50+ 11.9 3.9 25.8 33.4 2.5 21.9 0.6 100.0 197 Note: " Currently employed" includes women and men who indicated they were currently working at the time of the survey. Figures in parentheses are based on 25-49 unweighted cases. An asterisk indicates that an estimate is based on fewer than 25 unweighted cases. Characteristics of Respondents | 47 Younger women are more likely to work in professional, technical, and managerial jobs, while older women tend to work as skilled manual labourers. Urban men are more likely than rural men to work as a professional, technician, or manager, while rural men are more likely than urban men to work in agriculture. Whereas 19 percent of urban men work in professional, technical, and managerial jobs, only 9 percent of men in rural areas have this type of occupation. On the other hand, 28 percent of rural men work in agriculture compared with only 8 percent of men in urban areas. 3.7 EARNINGS AND TYPE OF EMPLOYMENT Table 3.7 presents the percent distribution of employed women and men age 15-49, by type of earnings and employer characteristics, according to type of employment (agricultural or non-agricultural). The majority of men and women, whether in agricultural or non-agricultural employment, receive cash income (91 percent of women and 88 percent of men). Men and women who work in agriculture are less likely than those who work in non-agricultural jobs to receive cash. For example, the proportion of women who earn cash in agricultural employment is 79 percent compared with 93 percent in non- agricultural work. Women working in agriculture are more likely not to be paid at all than to be paid in cash and in-kind. In fact, 10 percent of women and 25 percent of men who work in agriculture do not receive payment for their work. Table 3.7 Type of employment: Women and men age 15-49 Percent distribution of women and men age 15-49 employed in the 12 months preceding the survey by type of earnings, type of employer, and continuity of employment, according to type of employment (agricultural or non-agricultural), Swaziland 2006-07 WOMEN MEN Employment characteristic Agricultural work Non-agricultural work Total Agricultural work Non-agricultural work Total Type of earnings Cash only 78.7 93.0 91.1 72.6 91.0 87.7 Cash and in-kind 6.0 3.2 3.4 1.9 3.7 3.4 In-kind only 4.8 0.7 1.1 0.4 0.8 0.8 Not paid 10.4 3.1 3.7 24.7 4.4 7.9 Missing 0.0 0.1 0.6 0.3 0.1 0.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 Type of employer Employed by family member 6.9 4.1 4.4 18.1 7.5 9.4 Employed by non-family member 55.8 60.1 59.4 64.0 74.0 72.0 Self-employed 37.3 35.4 35.3 17.6 18.3 18.2 Missing 0.0 0.4 0.9 0.3 0.2 0.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Continuity of employment All year 39.3 78.8 74.8 33.9 61.9 56.9 Seasonal 51.2 12.9 16.2 45.3 16.7 21.6 Occasional 9.5 8.1 8.3 20.3 21.3 21.1 Missing 0.0 0.2 0.7 0.4 0.2 0.4 Total 100.0 100.0 100.0 100.0 100.0 100.0 Number of respondents employed during the past 12 months 188 1,978 2,191 394 1,914 2,322 Note: Total includes 24 women and 14 men with information missing on type of employment who are not shown separately. 48 | Characteristics of Respondents Six in ten women and 72 percent of men are employed by a non-family member. Men are more likely than women to be employed by a family member (9 percent and 4 percent, respectively). The difference is more striking in agricultural work, with 18 percent of men working for a family member compared with 7 percent of women. Women are twice as likely as men to be self-employed (35 percent compared with 18 percent). Women are more likely to work throughout the year than men. Three in four women hold a job all year compared with 57 percent of men. The agricultural sector is more likely to provide seasonal employment for both men (45 percent) and women (51 percent). On the other hand, work in the non- agricultural sector is more likely to provide income throughout the year (79 percent of women and 62 percent of men). Figure 3.2 shows the distribution of women who have worked at any time during the 12 months preceding the survey by the type of earnings women receive (cash, in-kind, or both). Figure 3.2 Type of Earnings of Women Employed in the Past 12 Months SDHS 2006-07 Cash only 91% Cash and in-kind 3% Not paid 4% Missing 1% In-kind only 1% Fertility Levels, Trends, and Differentials | 49 FERTILITY LEVELS, TRENDS, AND DIFFERENTIALS 4 Dudu Dlamini 4.1 INTRODUCTION This chapter looks at a number of fertility indicators, including levels, patterns, and trends in both current and cumulative fertility; the length of birth intervals; and the age at which women initiate childbearing. Information on current and cumulative fertility is essential in monitoring population growth. The data on birth intervals are important since short intervals are strongly associated with childhood mortality. The age at which childbearing begins can also have a major impact on the health and well- being of both the mother and the child. Data on fertility were collected in several ways. Each woman was asked about all of the births she had had in her lifetime. To ensure completeness of the responses, the duration, the month and year of termination, and the result of the pregnancy were recorded for each pregnancy. In addition, questions were asked separately about sons and daughters who live with the mother, those who live elsewhere, and those who have died. Subsequently, a list of all births was recorded along with name, age if still alive, and age at death if dead. Finally, information was collected on whether women were pregnant at the time of the survey. 4.2 CURRENT FERTILITY The level of current fertility is one of the most important topics in this report because of its direct relevance to population policies and programmes. Current fertility can be measured using the age-specific fertility rate (ASFR), the total fertility rate (TFR), the general fertility rate (GFR), and the crude birth rate (CBR). The ASFR provides the age pattern of fertility, while the TFR refers to the number of live births that a woman would have had if she were subject to the current ASFRs throughout the reproductive ages (15- 49 years). The GFR is expressed as the number of live births per 1,000 women of reproductive age, and the CBR is expressed as the number of live births per 1,000 population. The measures of fertility presented in this chapter refer to the period of three years prior to the survey. This generates a sufficient number of births to provide robust and current estimates. Table 4.1 depicts measures of current fertility for the three years preceding the survey for the country as a whole and by urban and rural residence. These are the total fertility rate (TFR), general fertility rate (GFR), and crude birth rate (CBR). The survey results indicate that the TFR for Swaziland for the three years preceding the 2006-07 SDHS survey is 3.8 births per woman. As expected, fertility is considerably higher in rural areas (4.2 births per woman) than in urban areas (3.0 births per woman). As the ASFRs show, the pattern of higher rural fertility is prevalent in all age groups except age 40-49. The urban-rural difference in fertility is more pronounced Table 4.1 Current fertility Age-specific and total fertility rate, the general fertility rate, and the crude birth rate for the three years preceding the survey, by residence, Swaziland 2006-07 Residence Age group Urban Rural Total 15-19 89 118 111 20-24 163 219 202 25-29 124 184 165 30-34 113 182 159 35-39 83 105 99 40-44 31 30 30 45-49 0 5 4 TFR 3.0 4.2 3.8 GFR 110 146 136 CBR 31.9 31.0 31.1 Note: Age-specific fertility rates are per 1,000 women. Rates for age group 45-49 may be slightly biased due to truncation. Rates are for the period 1-36 months prior to interview. TFR: Total fertility rate expressed per woman GFR: General fertility rate expressed per 1,000 women CBR: Crude birth rate expressed per 1,000 population 50 | Fertility Levels, Trends, and Differentials for women in the age group 30-34 (113 births per 1,000 women in urban areas versus 182 births per 1,000 women in rural areas). The overall age pattern of fertility as reflected in the ASFRs indicates that childbearing begins early. Fertility is low among adolescents and increases to a peak of 202 births per 1,000 among women age 20-24 and declines thereafter. The CBR in Swaziland is 31.1 births per 1,000 population. The overall GFR is 136 and is higher in the rural areas than in urban areas (146 and 110 births per 1,000 women, respectively). Compared with other south-east African countries that have participated in the DHS programme, the TFR in Swaziland is the same as that in Zimbabwe and slightly higher than in Namibia and Lesotho. 4.3 FERTILITY DIFFERENTIALS Fertility is known to vary by residence, educational background, and other background characteristics of a woman. Table 4.2 shows several different indicators of fertility, mainly the total fertility rate, mean number of births to women age 40-49, and the percentage currently pregnant. The mean number of births to women age 40-49 is an indicator of cumulative fertility; it reflects the fertility performance of older women who are nearing the end of their reproductive period. If fertility remains stable over time, the two fertility measures, total fertility rate (TFR) and children ever born (CEB), tend to be very similar. The percentage pregnant provides a useful additional measure of current fertility, although it is recognized that it may not capture all pregnancies in an early stage. Table 4.2 indicates that there are variations in the TFR by residence, region, education, and wealth quintile. Fertility is highest in the Shiselweni region with a TFR of 4.3 births per woman and lowest in Hhohho at 3.6 births per woman. TFR decreases gradually with increasing level of education; better educated women have fewer children than less educated women. Women with no education have on average 4.9 children compared with 2.4 children for women with tertiary education. Fertility varies 6.7 6 5.9 5.7 5.5 4.9 3.8 3.8 3.6 3.5 Uganda 2006 Malawi 2004 Zambia 2001-02 Tanzania 2004 Mozam- bique 2003 Kenya 2003 Zimba- bwe 2005-06 Swaziland 2006-07 Namibia 2006-07 Lesotho 2004 0 1 2 3 4 5 6 7 8 Births per woman Figure 4.1 Total Fertility Rates for Selected Countries in Southeast Africa SDHS 2006-07 6.0 Fertility Levels, Trends, and Differentials | 51 widely according to household wealth. Women in the highest wealth quintile have 2.9 children fewer than women in the lowest quintile (2.6 and 5.5 births per woman, respectively). It has been noted that although the percentage currently pregnant is a useful measure of current fe- rtility, not all women who are pregnant are likely to be captured because they may not be aware that they are pregnant or may be reluctant to disclose a pregnancy in the early stages. Six percent of women reported that they were pregnant at the time of the survey. Women who have lower primary education and those in the lowest wealth quintile are the most likely to be pregnant. Table 4.2 also presents a crude assessment of trends in the various subgroups by comparing current fertility with a measure of completed fertility: the mean number of children ever born to women age 40-49. The mean number of children ever born to older women who are nearing the end of their reproductive period is an indicator of average completed fertility of women who began childbearing during the three decades preceding the survey. If fertility remained constant over time and the reported data on both children ever born and births during the three years preceding the survey are reasonably accurate, the TFR and the mean number of children ever born for women 40-49 are expected to be similar. When fertility levels have been falling, the TFR will be substantially lower than the mean number of children ever born among women age 40-49. The comparison suggests that fertility has fallen by 1.5 births during the past few decades, from 5.3 births per woman to 3.8. Fertility has declined in both urban and rural areas, in all regions, at all educational levels, and for all wealth quintiles. The difference between the level of current and completed fertility is highest in rural areas (1.6 births), the Hhohho and Lubombo regions (1.6 births), among women who have no education (1.6 births), and among women in the middle and fourth wealth quintiles (1.7 births). 4.4 FERTILITY TRENDS Table 4.3 uses information from the retrospective birth histories obtained from the SDHS respondents to examine trends in age-specific fertility rates for successive five-year periods before the survey. To calculate these rates, births were classified according to the period of time in which the birth occurred and the mother’s age at the time of birth. Because birth histories were not collected for women over age 50, the rates for older age groups become progressively more truncated for periods more distant from the survey date. For example, rates cannot be calculated for women age 45-49 for the period 5-9 years or more prior to the survey, because women in that age group would have been 50 years or older at the time of the survey. Table 4.2 Fertility by background characteristics Total fertility rate for the three years preceding the survey, percentage of women age 15-49 currently pregnant, and mean number of children ever born to women age 40-49 years, by background characteristics, Swaziland 2006-07 Background characteristic Total fertility rate Percentage of women age 15-49 currently pregnant Mean number of children ever born to women age 40-49 Residence Urban 3.0 5.6 4.0 Rural 4.2 5.6 5.8 Region Hhohho 3.6 5.1 5.2 Manzini 3.7 5.7 5.1 Shiselweni 4.3 5.7 5.5 Lubombo 4.0 5.9 5.6 Education No education 4.9 5.5 6.5 Lower primary 5.1 8.9 6.6 Higher primary 4.4 6.1 5.4 Secondary 3.9 4.6 5.0 High school 3.1 5.9 4.0 Tertiary 2.4 4.5 3.2 Wealth quintile Lowest 5.5 7.3 6.8 Second 4.9 6.1 6.0 Middle 3.9 5.2 5.6 Fourth 3.3 4.8 5.0 Highest 2.6 5.2 3.8 Total 3.8 5.6 5.3 Note: Total fertility rates are for the period 1-36 months prior to interview. Table 4.3 Trends in age-specific fertility rates Age-specific fertility rates for five-year periods preceding the survey, by mother's age at the time of the birth, Swaziland 2006-07 Number of years preceding survey Mother's age at birth 0-4 5-9 10-14 15-19 15-19 111 126 137 148 20-24 196 203 222 252 25-29 172 207 223 233 30-34 158 170 189 [191] 35-39 104 116 [161] - 40-44 32 [65] - - 45-49 [5] - - - Note: Age-specific fertility rates are per 1,000 women. Estimates in brackets are truncated. Rates exclude the month of interview. 52 | Fertility Levels, Trends, and Differentials Nonetheless, the results in Table 4.3 show that fertility has dropped substantially among all age groups over the past two decades. This decline is most obvious in the 15 years preceding the survey, with the largest decline observed between the two most recent five-year periods. Fertility decline is steepest among the cohort age 30-34, with a 60 percent decline between the period 15-19 years before the survey and the period 0-4 years before the survey. Another way to examine fertility trends is to compare current estimates with earlier surveys. Table 4.4 and Figure 4.2 show the ASFRs for the 1986 and 1997 Population and Housing Censuses (PHC), 1991 Demographic and Housing Survey (DHS), and 2006-07 SDHS. Estimates prior to the SDHS were calculated using information on total number of children ever born and surviving. The data indicate that fer- tility has been declining rapidly in Swaziland, with the TFR falling from 6.4 in the 1986 PHC to 4.5 in 1997 PHC, and to 3.8 in the 2006-07 SDHS. This translates to a decrease of 2.6 births since 1986. A substantial fertility decline has occurred in all age groups. Between 1986 and 1997, the largest decline took place among the younger age groups, but since 1997, larger declines in ASFR are observed among older women (Figure 4.2). Table 4.4 Trends in fertility Age-specific fertility rates (per 1,000 women) and total fertility rates for the three years preceding the survey, Swaziland 1986, 1991, 1997, and 2006-07 Age group 1986 PHC 1991 DHS 1997 PHC 2006-07 SDHS 15-19 145 134 90 111 20-24 289 253 211 202 25-29 279 252 210 165 30-34 246 208 169 159 35-39 178 147 133 99 40-44 89 88 64 30 45-49 45 35 31 4 TFR 6.4 5.6 4.5 3.8 DHS = Demographic and Housing Survey PHC = Population and Housing Census Sources: CSO, 1976 Population and Housing Census, Vol. 1; CSO, 1986 Population and Housing Census, Vol. 4; CSO, 1997 Population and Housing Census Vol. 4. % % % % % % % & & & & & & & + + + + + + + 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age group 0 50 100 150 200 250 300 Births per 1,000 women PHC 1986 DHS 1991 PHC 1997 SDHS 2006-07+ & % SDHS 2006-07 Figure 4.2 Trends in Fertility Fertility Levels, Trends, and Differentials | 53 4.5 CHILDREN EVER BORN AND LIVING Table 4.5 presents the distribution of all women and currently married women by number of children ever born, according to five-year age groups. The table also shows the mean number of children ever born. Data on the number of children ever born reflect the accumulation of births to women over their entire reproductive years and therefore have limited reference to current fertility levels, particularly when a country has experienced a decline in fertility. Moreover, the data are subject to recall error, which is typically greater for older than younger women. Nevertheless, the information on children ever born (or parity) is useful in looking at a number of issues. The parity data show how average family size varies across age groups. The percentage of women in their forties who have never had children also provides an indicator of the level of primary infertility, or the inability to bear children. Comparison of the differences in the mean number of children ever born and surviving reflects the cumulative effects of mortality levels during the period in which women have been bearing children. Table 4.5 shows the percent distribution of all women and currently married women by number of children ever born, mean number of children ever born, and mean number of children living. More than four-fifths of women age 15-19 (82 percent) have never given birth. However, this proportion declines to 11 percent for women age 25-29 and to 5 percent or less among women age 30 and above, indicating that childbearing among Swazi women is nearly universal. On average, Swazi women nearing the end of their reproductive years have attained a parity of 4.8 children. This is one child more than the total fertility rate, a difference brought about by the decline in fertility during the 1980s and 1990s. The same pattern is replicated for currently married women, except that the mean number of children ever born is higher for currently married women (3.6 children) than for all women (2.3 children). The difference in the mean number of children ever born between all women and currently married women is due to a substantial proportion of young and unmarried women in the former category who exhibit lower fertility. Table 4.5 Children ever born and living Percent distribution of all women and currently married women age 15-49 by number of children ever born, mean number of children ever born, and mean number of living children, according to age group, Swaziland 2006-07 Number of children ever born Age 0 1 2 3 4 5 6 7 8 9 10+ Total Number of women Mean number of children ever born Mean number of living children ALL WOMEN 15-19 81.5 16.0 2.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 1,274 0.21 0.19 20-24 29.2 41.0 23.6 4.9 1.3 0.0 0.0 0.0 0.0 0.0 0.0 100.0 1,046 1.08 0.97 25-29 10.8 23.7 28.8 21.5 9.1 4.9 0.9 0.3 0.0 0.0 0.0 100.0 729 2.14 1.90 30-34 3.7 11.1 23.8 26.6 13.0 10.7 7.2 2.0 1.5 0.1 0.2 100.0 616 3.17 2.92 35-39 4.8 7.3 11.9 16.2 18.0 14.0 12.4 7.5 4.6 1.9 1.4 100.0 503 4.17 3.86 40-44 3.3 5.1 8.9 11.0 14.1 12.9 13.4 10.5 10.2 4.1 6.4 100.0 438 5.19 4.65 45-49 4.6 6.9 5.7 8.3 13.2 11.5 12.5 11.4 7.9 9.2 8.7 100.0 383 5.50 4.76 Total 30.1 19.3 15.2 10.7 7.3 5.5 4.4 2.8 2.2 1.3 1.4 100.0 4,987 2.28 2.05 CURRENTLY MARRIED WOMEN 15-19 26.2 58.5 15.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 88 0.89 0.83 20-24 13.6 39.4 35.5 8.6 2.9 0.0 0.0 0.0 0.0 0.0 0.0 100.0 343 1.48 1.31 25-29 6.0 17.4 30.2 24.5 13.1 7.0 1.3 0.5 0.0 0.0 0.0 100.0 388 2.50 2.24 30-34 1.5 7.4 23.3 26.0 14.3 12.3 10.0 2.9 2.1 0.1 0.0 100.0 379 3.49 3.23 35-39 2.9 4.9 9.7 16.0 18.0 14.6 13.9 8.7 6.9 2.7 1.7 100.0 334 4.59 4.25 40-44 2.1 3.8 8.5 9.5 14.3 14.6 12.9 10.5 12.0 5.4 6.5 100.0 291 5.44 4.93 45-49 4.3 4.6 6.0 7.4 13.5 12.7 11.5 8.9 10.7 9.9 10.5 100.0 238 5.79 5.12 Total 6.1 15.6 20.0 15.6 12.1 9.5 7.5 4.5 4.4 2.3 2.4 100.0 2,062 3.58 3.25 54 | Fertility Levels, Trends, and Differentials 4.6 BIRTH INTERVALS A birth interval is defined as the length of time between two successive live births. The study of birth intervals is important in understanding the health status of young children. Information on birth intervals provides insight into birth spacing patterns, which affect fertility as well as maternal, infant, and childhood mortality. Research has shown that short birth intervals are closely associated with poor health of children, especially during in

View the publication