Rwanda - Demographic and Health Survey - 2006

Publication date: 2006

REPUBLIC OF RWANDA Rwanda Demographic and Health Survey 2005 Institut National de la Statistique Ministère des Finances et de la Planification Économique Kigali, Rwanda ORC Macro Calverton, Maryland, USA July 2006 The following people participated in data analysis and the preparation of this report: Jean Philippe Gatarayiha, Apolline Mukanyonga, Dr Eugénie Kayirangwa, Adeline Kabeja, Alphonse Rukundo, Athanasie Kabagwira, Gafishi N. Philippe, Rwabikumba Dévote, Rwakayiro Ignace, Muhoza Ananie, Kalinda Charles, Kayibanda Françoise, Fern Greenwell, Noah Bartlett, Mohamed Ayad, and Monique Barrère. This report presents the findings of the 2005 Rwanda Demographic and Health Survey (RDHS-III), carried out from February to July 2005 by the Direction de la Statistique (renamed Institut National de la Statistique du Rwanda in September 2005). Funding for the RDHS-III was provided by USAID, the Commission Nationale de Lutte contre le SIDA (CNLS) through the World Bank’s Multi-County AIDS Program (MAP), Unicef, UNFPA, DFID and GTZ. Assistance was also provided by other national organizations, such as the Treatment and Research AIDS Center (TRAC), the Laboratoire National de Référence and the Service National de Recensement (SNR). Technical assistance was provided by ORC Macro as part of the Demographic and Health Surveys project (MEASURE DHS). The objective of the MEASURE DHS project is to collect, analyze and disseminate demographic data, especially those related to fertility, family planning, maternal and child health, and HIV/AIDS. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID or other cooperating organizations. Additional information about the survey can be obtained from the Institut National de la Statistique du Rwanda (INSR), BP 6139, Kigali, Rwanda (Telephone: (250) 55104164; e-mail: snr@rwanda1.com). Additional information about the MEASURE DHS project can be obtained from ORC Macro, 11785 Beltsville Drive, Suite 300, Calverton, MD 20705, USA. Telephone: 301-572-0200; Fax: 301-572-0999; e-mail: reports@orcmacro.com; Internet: http://www.measuredhs.com). Recommended citation: Institut National de la Statistique du Rwanda (INSR) and ORC Macro. 2006. Rwanda Demographic and Health Survey 2005. Calverton, Maryland, U.S.A.: INSR and ORC Macro. Contents | iii CONTENTS Page TABLES AND FIGURES . ix FOREWORD . xix ACKNOWLEDGMENTS . xxi ABBREVIATIONS . xxiii SUMMARY OF FINDINGS . xxv MILLENNIUM DEVELOPMENT GOAL INDICATORS. xxxi MAP OF RWANDA.xxxii CHAPTER 1 INTRODUCTION 1.1 Country Profile .1 1.1.1 Geography .1 1.1.2 Economy.2 1.1.3 Population .3 1.1.4 Population Policy .4 1.1.5 Public Health Policy.4 1.2 Objectives and Methodology of the Survey.4 1.2.1 Objectives of the Survey .5 1.2.2 Questionnaires.5 1.2.3 Sample Design .7 1.2.4 Sample Coverage .7 1.2.5 Hemoglobin and HIV Testing .8 1.2.6 Training and Data Collection.9 1.2.7 Data Processing.10 CHAPTER 2 HOUSEHOLD CHARACTERISTICS 2.1 Household Population By Age and Sex .11 2.2 Household Size and Composition.12 2.3 School Attendance and Educational Attainment.13 2.4 Living Conditions.18 2.5 Birth Registration with Civil Authorities .21 iv │ Contents CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS 3.1 Background Characteristics of Respondents .23 3.2 Educational Attainment.25 3.3 Literacy.27 3.4 Exposure to Mass Media .28 3.5 Employment .31 CHAPTER 4 FERTILITY 4.1 Fertility Levels and Differentials .38 4.2 Fertility Trends.41 4.3 Parity and Primary Infertility.44 4.4 Birth Intervals .46 4.5 Age at First Birth .48 4.6 Teenage Fertility .49 CHAPTER 5 FAMILY PLANNING 5.1 Knowledge of Contraception .51 5.2 Use of Contraception.52 5.2.1 Ever Use of Contraception .52 5.2.2 Current Use of Contraception .54 5.3 Number of Children at First Use of Contraception .59 5.4 Knowledge of Fertile Period.59 5.5 Source of Contraception.60 5.6 Future Use of Contraception.61 5.7 Exposure to Family Planning Messages.64 5.8 Contact of Nonusers with Family Planning Providers.51 5.9 Opinions and Attitudes of Couples toward Family Planning .51 5.9.1 Discussion of Family Planning with Husband.51 5.9.2 Attitudes of Couples toward Family Planning.52 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY 6.1 Marital Status.71 6.2 Polygyny.72 6.3 Age at First Union.73 6.4 Age at First Sexual Intercourse .76 6.4 Recent Sexual Activity.78 6.6 Exposure to the Risk of Pregnancy .81 6.7 Menopause .82 Contents | v CHAPTER 7 FERTILITY PREFERENCES 7.1 Desire for (More) Children.85 7.2 Need for Family Planning Services .88 7.3 Ideal Number of Children.90 7.4 Fertility Planning Status .92 CHAPTER 8 MOTHER AND CHILD HEALTH 8.1 Antenatal Care.95 8.2 Delivery Care . 100 8.3 Postnatal Care . 105 8.4 Vaccination of Children . 106 8.5 Childhood Illnesses . 109 8.5.1 Acute Respiratory Infection (ARI) and Fever. 109 8.5.2 Diarrhea . 111 8.6 Problems in Accessing Health Care . 116 8.7 Tobacco Consumption. 117 CHAPTER 9 MALARIA 9.1 Introduction . 119 9.2 Malaria Prevention . 121 9.2.1 Household Possession of Mosquito Nets . 122 9.2.2 Use of Mosquito Nets by Children . 123 9.2.3 Use of Mosquito Nets by Women . 125 9.2.4 Intermittent Preventive Treatment during Pregnancy . 127 9.3 Treatment of Malaria in Children Under the Age of Five . 128 CHAPTER 10 BREASTFEEDING AND NUTRITION OF MOTHERS AND CHILDREN 10.1 Breastfeeding and Supplementation. 131 10.2 Micronutrient Intake And Anemia Prevalence . 137 10.3 Prevalence of Anemia Due to Iron Deficiency. 143 10.4 Nutritional Status of Children. 147 10.5 Nutritional Status of Women . 154 CHAPTER 11 INFANT AND CHILD MORTALITY 11.1 Definition, Methodology, and Data Quality . 157 11.2 Levels and Trends . 158 11.3 Differentials in Infant and Child Mortality . 160 11.4 Perinatal Mortality . 163 vi │ Contents 11.5 High-Risk Fertility Behavior. 165 CHAPTER 12 MATERNAL MORTALITY 12.1 Introduction . 167 12.2 Data Collection. 167 12.3 Data Quality. 168 12.4 Direct Estimates of Adult Mortality. 170 12.5 Direct Estimates of Maternal Mortality . 172 CHAPTER 13 DOMESTIC VIOLENCE 13.1 Methodology . 175 13.2 Domestic Violence . 177 13.2.1 Physical Violence Since Age 15 . 177 13.2.2 Violence during Pregnancy. 178 13.2.3 Marital Control Exercised by the Husband/Partner . 179 13.3 Spousal Violence . 181 13.3.1 Prevalence of Spousal Violence. 181 13.3.2 Frequency of Recent Spousal Violence. 183 13.3.3 Onset of Spousal Violence . 184 13.4 Consequences Of Violence And Help Seeking . 185 13.5 Violence by Spousal Characteristics and Women’s Status Indicators. 187 CHAPTER 14 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR 14.1 Knowledge, Opinions, And Attitudes . 192 14.1.1 Knowledge of HIV Transmission and Prevention Methods . 192 14.1.2 Stigmatization . 198 14.1.3 Opinions. 200 14.2 Higher-Risk Sexual Intercourse and Condom Use . 203 14.3 Testing and Counseling for HIV/AIDS. 206 14.4 Sexually Transmitted Infections (STIs). 209 14.5 Injections from a Health Worker. 211 14.6 Knowledge of HIV/AIDS and Sexual Behavior among Youth . 214 CHAPTER 15 HIV PREVALENCE AND ASSOCIATED FACTORS 15.1 HIV Testing Protocol. 225 15.2 Coverage of HIV Testing . 228 15.3 HIV Prevalence. 231 Contents | vii 15.3.1 HIV Prevalence Distribution According to Sociodemographic Variables. 231 15.3.2 HIV Prevalence by Demographic Variables . 233 15.3.3 HIV Prevalence by Sexual Behavior Characteristics. 234 15.3.4 HIV Prevalence among Youth. 236 15.3.5 HIV Prevalence and Other Risk Factors . 238 15.3.6 HIV Prevalence and Male Circumcision . 239 15.3.7 HIV Prevalence among Couples . 239 15.4 Sentinel Surveillance System and RDHS-III . 241 CHAPTER 16 ORPHANED AND VULNERABLE CHILDREN 16.1 Orphanhood and Children’s Living Arrangements. 243 16.2 Access to Essential Services . 246 16.3 Strengthening Family Capacities to Support And Protect Children. 247 16.3.1 Malnutrition . 247 16.3.2 Early Sexual Intercourse . 248 16.3.3 Succession Planning . 248 16.4 Protection of Vulnerable Children. 249 16.5 Care and Support . 250 16.5.1 Care and Support of the Chronically Ill. 250 16.5.2 Care and Support of OVC. 251 REFERENCES .253 APPENDIX A SAMPLE IMPLEMENTATION A.1 Introduction . 257 A.2 Sample Frame . 257 A.3 Sample Selection . 258 A.4 Sampling Probability . 259 A.5 Survey Results. 259 APPENDIX B ESTIMATES OF SAMPLING ERRORS. 267 APPENDIX C DATA QUALITY TABLES . 279 APPENDIX D RESULTS ACCORDING TO OLD PROVINCES . 285 APPENDIX E PERSONS INVOLVED IN THE 2005 RWANDA DEMOGRAPHIC AND HEALTH SURVEY. 323 APPENDIX F QUESTIONNAIRES . 327 Tables and Figures | ix TABLES AND FIGURES Page CHAPTER 1 INTRODUCTION Table 1.1 Results of the household and individual interviews. 8 CHAPTER 2 HOUSEHOLD CHARACTERISTICS Table 2.1 Household population by age, sex, and residence. 11 Table 2.2 Household composition. 13 Table 2.3.1 Educational attainment of household population: female. 14 Table 2.3.2 Educational attainment of household population: male. 15 Table 2.4 School attendance ratios . 16 Table 2.5 Housing characteristics. 19 Table 2.6 Household durable goods . 20 Table 2.7 Wealth quintiles. 21 Table 2.8 Birth registration of children under age five . 22 Figure 2.1 Population pyramid . 12 Figure 2.2 Age-specific attendance rates . 18 CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS Table 3.1 Age of respondents . 23 Table 3.2 Background characteristics of respondents . 24 Table 3.3.1 Educational attainment by background characteristics: women . 25 Table 3.3.2 Educational attainment by background characteristics: men. 26 Table 3.4.1 Literacy: women. 27 Table 3.4.2 Literacy: men. 28 Table 3.5.1 Exposure to mass media: women. 29 Table 3.5.2 Exposure to mass media: men. 30 Table 3.6 Employment status. 32 Table 3.7.1 Occupation: women. 33 Table 3.7.2 Occupation: men . 34 Table 3.8 Type of employment. 35 Table 3.9 Type of employer. 36 CHAPTER 4 FERTILITY Table 4.1 Current fertility . 38 Table 4.2 Fertility by background characteristics . 40 Table 4.3 Trends in fertility. 41 x | Tables and Figures Table 4.4 Trends in age-specific fertility rates. 42 Table 4.5.1 Children ever born and living : women. 44 Table 4.5.2 Children ever born and living : men. 45 Table 4.6 Birth intervals. 47 Table 4.7 Age at first birth . 48 Table 4.8 Median age at first birth by background characteristics. 49 Table 4.9 Teenage pregnancy and motherhood. 50 Figure 4.1 Age-Specific Fertility Rates, by Residence. 39 Figure 4.2 Total Fertility Rate and Mean Number of Children Ever Born to Women Age 40-49. 40 Figure 4.3 Trends in Age-Specific Fertility Rates, Rwanda 1992, 2000, and 2005 . 42 Figure 4.4 Age-Specific Fertility Rates for Five-Year Periods Preceding the Survey . 43 Figure 4.5 Trends in the Total Fertility Rate among Women Age 15-34, Rwanda 1992, 2000, and 2005 . 43 CHAPTER 5 FAMILY PLANNING Table 5.1.1 Knowledge of contraceptive methods: women. 51 Table 5.1.2 Knowledge of contraceptive methods: men . 52 Table 5.2 Ever use of contraception . 53 Table 5.3 Current use of contraception . 55 Table 5.4 Current use of contraception by background characteristics . 58 Table 5.5 Number of children at first use of contraception . 59 Table 5.6 Knowledge of the fertile period. 60 Table 5.7 Source of contraception. 61 Table 5.8 Future use of contraception . 62 Table 5.9 Reason for not intending to use contraception . 63 Table 5.10 Preferred method of contraception for future use. 64 Table 5.11.1 Exposure to family planning messages: women . 65 Table 5.11.2 Exposure to family planning messages: men. 66 Table 5.12 Contact of nonusers with family planning providers . 67 Table 5.13 Discussion of family planning with husband . 68 Table 5.14 Attitudes towards family planning . 69 Figure 5.1 Contraceptive Use among Currently Married Women Age 15-49. 56 Figure 5.2 Trends in Use of Modern Methods among Currently Married Women . 57 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY Table 6.1 Current marital status . 71 Table 6.2 Number of co-wives and wives . 73 Table 6.3 Age at first marriage . 74 Table 6.4 Median age at first marriage. 75 Table 6.5 Age at first sexual intercourse:. 76 Table 6.6 Median age at first sexual intercourse. 78 Table 6.7.1 Recent sexual activity: women. 79 Tables and Figures | xi Table 6.7.2 Recent sexual activity: men. 80 Table 6.8 Postpartum amenorrhea, abstinence, and insusceptibility. 81 Table 6.9 Median duration of postpartum insusceptibility by background characteristics . 82 Table 6.10 Menopause. 83 Figure 6.1 Percentage of Never-Married Women and Men, by Age . 72 Figure 6.2 Median Age at First Marriage among Women and Men, by Background Characteristics. 75 Figure 6.3 Median Age at First Intercourse and at First Union among Women 25-49, by Background Characteristics . 77 CHAPTER 7 FERTILITY PREFERENCES Table 7.1 Fertility preferences by number of living children . 86 Table 7.2 Desire to limit childbearing . 88 Table 7.3 Need for family planning among currently married women. 89 Table 7.4 Ideal number of children . 91 Table 7.5 Mean ideal number of children. 92 Table 7.6 Fertility planning status. 93 Table 7.7 Wanted fertility rates. 94 Figure 7.1 Proportion of Currently Married Women and Men Who Want No More Children, by Number of Living Children. 87 CHAPTER 8 MATERNAL AND CHILD HEALTH Table 8.1 Antenatal care. 96 Table 8.2 Number of antenatal care visits and timing of first visit . 97 Table 8.3 Components of antenatal care . 98 Table 8.4 Tetanus toxoid injections . 100 Table 8.5 Place of delivery . 101 Table 8.6 Assistance during delivery . 102 Table 8.7 Delivery characteristics . 104 Table 8.8 Postnatal care . 106 Table 8.9 Vaccinations by source of information. 107 Table 8.10 Vaccinations by background characteristics . 109 Table 8.11 Prevalence and treatment of symptoms of ARI and fever. 110 Table 8.12 Prevalence of diarrhea . 112 Table 8.13 Knowledge of ORS packets . 114 Table 8.14 Diarrhea treatment . 115 Table 8.15 Feeding practices during diarrhea . 116 Table 8.16 Problems in accessing health care . 117 Table 8.17 Use of smoking tobacco. 118 Figure 8.1 Trends in Antenatal Care and Delivery, Rwanda 1992, 2000, and 2005 . 97 Figure 8.2 Children Whose Delivery Was Assisted by Trained Personnel . 103 xii | Tables and Figures Figure 8.3 Trends in Vaccination Coverage among Children Age 12-23 Months, Rwanda 1992, 2000, and 2005 . 108 Figure 8.4 Prevalence of ARI, Fever, and Diarrhea, by Age . 113 CHAPTER 9 MALARIA Table 9.1 Household possession of mosquito nets . 122 Table 9.2 Use of mosquito nets by children. 124 Table 9.3 Use of mosquito nets by women. 126 Table 9.4 Use of Intermittent Preventive Treatment by women during pregnancy. 127 Table 9.5 Use of SP/Fansidar by women during pregnancy . 128 Table 9.6 Prevalence and prompt treatment of children with fever . 129 Table 9.7 Type and timing of antimalarial drugs taken by children with fever . 130 Figure 9.1 Household Ownership of Mosquito Nets . 123 Figure 9.2 Use of Mosquito Nets by Children Under Age 5, According to Province. 125 Figure 9.3 Pregnant Women Who Slept Under a Mosquito Net the Night Preceding the Survey. 126 CHAPTER 10 BREASTFEEDING AND NUTRITION OF MOTHERS AND CHILDREN Table 10.1 Initial breastfeeding. 132 Table 10.2 Breastfeeding status by age . 134 Table 10.3 Median duration and frequency of breastfeeding . 136 Table 10.4 Foods consumed by children in the day or night preceding the interview. 137 Table 10.5 Iodization of household salt . 138 Table 10.6 Micronutrient intake among children . 140 Table 10.7 Micronutrient intake among mothers . 142 Table 10.8 Prevalence of anemia in children . 144 Table 10.9 Prevalence of anemia in women . 145 Table 10.10 Prevalence of anemia in children by anemia status of mother . 146 Table 10.11 Prevalence of anemia in men. 147 Table 10.12 Nutritional status of children . 149 Table 10.13 Nutritional status of women . 156 Figure 10.1 Breastfeeding Practices Among Children Under Age 3 . 135 Figure 10.2 Percentage of Children Under Age 5 Who Are Stunted . 150 Figure 10.3 Percentage of Children Under Age 5 Who Are Wasted . 152 Figure 10.4 Trends in malnutrition among Children under 5 Years), Rwanda 1992, 2000, and 2005. 154 CHAPTER 11 INFANT AND CHILD MORTALITY Table 11.1 Early childhood mortality rates . 158 Table 11.2 Early childhood mortality rates by background characteristics. 161 Table 11.3 Early childhood mortality rates by demographic characteristics. 162 Table 11.4 Perinatal mortality. 164 Tables and Figures | xiii Table 11.5 High-risk fertility behavior . 166 Figure 11.1 Trends in Infant and Under-five Mortality, Rwanda 1992, 2000, and 2005. 159 Figure 11.2 Trends in Infant and Under-five Mortality from the RDHS-I, RDHS-II, and RDHS-III . 160 Figure 11.3 Under-five Mortality by Mother’s Background Characteristics . 161 Figure 11.4 Infant Mortality by Reproductive Behavior . 163 CHAPTER 12 MATERNAL MORTALITY Table 12.1 Data on siblings . 168 Table 12.2 Indicators on data quality. 169 Table 12.3 Estimates of age-specific female and male adult mortality. 170 Table 12.4 Maternal mortality . 173 Figure 12.1 Female Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group . 171 Figure 12.2 Male Mortality Rates for the Period 2000-2004 and Model Life Table Rates, by Age Group . 172 CHAPTER 13 DOMESTIC VIOLENCE Table 13.1 Experience of beatings or physical mistreatment . 177 Table 13.2 Perpetrators of violence . 178 Table 13.3 Violence during pregnancy . 179 Table 13.4 Marital control exercised by husband. 180 Table 13.5 Marital violence . 182 Table 13.6 Frequency of spousal violence . 184 Table 13.7 Onset of spousal violence . 185 Table 13.8 Physical consequences of spousal violence. 186 Table 13.9 Help seeking. 187 Table 13.10 Spousal violence, women's status, and spousal characteristics. 188 Figure 13.1 Percentage of Ever-Married Women who Have Ever Experienced Specific Forms of Violence from Their Husbands . 183 Figure 13.2 Prevalence of Spousal Violence, by Level of Education of Woman and Her Spouse and Alcohol Consumption of Spouse. 189 CHAPTER 14 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR Table 14.1 Knowledge of AIDS. 192 Table 14.2 Knowledge of HIV prevention methods. 193 Table 14.3.1 Comprehensive knowledge about AIDS : women . 195 Table 14.3.2 Comprehensive knowledge about AIDS : men . 196 Table 14.4 Knowledge of prevention of mother-to-child transmission of HIV. 197 Table 14.5.1 Accepting attitudes toward those living with HIV/AIDS: women. 199 Table 14.5.2 Accepting attitudes toward those living with HIV/AIDS: men. 200 xiv | Tables and Figures Table 14.6 Attitudes toward negotiating safer sexual relations with husband . 201 Table 14.7 Adult support of education about condom use to prevent AIDS . 202 Table 14.8.1 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: women . 204 Table 14.8.2 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months: men . 205 Table 14.9.1 Prior HIV testing and knowledge of results: women . 206 Table 14.9.2 Prior HIV testing and knowledge of results: men . 207 Table 14.10 Pregnant women counseled and tested for HIV. 209 Table 14.11 Self-reported prevalence of sexually-transmitted infections (STIs) and STI symptoms . 210 Table 14.12 Prevalence of injections . 212 Table 14.13 Comprehensive knowledge about AIDS and of a source of condoms among youth . 214 Table 14.14 Age at first sexual intercourse among youth. 216 Table 14.15 Condom use at first sexual intercourse among youth. 217 Table 14.16 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth . 218 Table 14.17 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months . 219 Table 14.18 Age-mixing in sexual relationships among women age 15-19 . 221 Table 14.19 Drunkenness during sexual intercourse among youth. 222 Table 14.20 Recent HIV tests among youth . 223 Figure 14.1 Perception and Beliefs about Abstinence and Faithfulness. 203 Figure 14.2 Women and Men Seeking Treatment for STIs . 211 Figure 14.3 Type of Facility where Received Last Medical Injection . 213 Figure 14.4 Percentage whose Last Injection was Given with a Syringe and Needle Taken from a New, Unopened Package. 213 Figure 14.5 Trends in Age at First Sex, Rwanda 2000 and 2005. 217 Figure 14.6 Abstinence, Being Faithful, and Condom Use (ABC) Among Young Women and Men . 220 CHAPTER 15 HIV PREVALENCE AND ASSOCIATED FACTORS Table 15.1 Coverage of HIV testing by residence and province. 228 Table 15.2 Coverage of HIV testing by background characteristics . 230 Table 15.3 HIV prevalence by age. 231 Table 15.4 HIV prevalence by background characteristics . 232 Table 15.5 HIV prevalence and confidence intervals . 233 Table 15.6 HIV prevalence by sociodemographic characteristics . 234 Table 15.7 HIV prevalence by sexual behavior characteristics. 235 Table 15.8 HIV prevalence among young people . 237 Table 15.9 HIV prevalence by other characteristics. 238 Table 15.10 Prior HIV testing by HIV status . 239 Table 15.11 HIV prevalence by male circumcision . 239 Table 15.12 HIV prevalence among couples. 240 Tables and Figures | xv Figure 15.1 HIV Prevalence by Sex and Age . 231 CHAPTER 16 ORPHANED AND VULNERABLE CHILDREN Table 16.1 Children’s living arrangements and orphanhood . 244 Table 16.2 Orphans and vulnerable children (OVC) . 245 Table 16.3 School attendance by survivorship of parents and by OVC status . 246 Table 16.4 Underweight orphans and vulnerable children. 247 Table 16.5 Sexual intercourse before age 15 among orphans and vulnerable children . 248 Table 16.6 Succession planning. 249 Table 16.7 Widows dispossessed of property. 250 Table 16.8 External support for chronically ill persons . 251 Table 16.9 External support for orphans and vulnerable children. 252 APPENDIX A SAMPLE IMPLEMENTATION Table A.1 Distribution of households and enumeration areas (EAs) by old province and according to residence (RGPH, 2002) . 257 Table A.2 Sample allocation by old province and according to residence. 258 Table A.3 Sample implementation: women . 260 Table A.4 Sample implementation: men. 261 Table A.5 Coverage of HIV testing among interviewed women by background characteristics . 262 Table A.6 Coverage of HIV testing among interviewed men by background characteristics . 263 Table A.7 Coverage of HIV testing among women who ever had sex by risk status variables . 264 Table A.8 Coverage of HIV testing among men who ever had sex by risk status variables . 265 APPENDIX B ESTIMATES OF SAMPLING ERRORS Table B.1 List of selected variables for sampling errors . 270 Table B.2 Sampling errors - National sample. 271 Table B.3 Sampling errors - Urban sample . 272 Table B.4 Sampling errors - Rural sample. 273 Table B.5 Sampling errors – City of Kigali. 274 Table B.6 Sampling errors – South Province. 275 Table B.7 Sampling errors – West Province. 276 Table B.8 Sampling errors – North Province. 277 Table B.9 Sampling errors – East Province. 278 APPENDIX C DATA QUALITY TABLES Table C.1 Household age distribution . 279 Table C.2.1 Age distribution of eligible and interviewed women . 280 Table C.2.2 Age distribution of eligible and interviewed men. 280 xvi | Tables and Figures Table C.3 Completeness of reporting . 281 Table C.4 Births by calendar years . 281 Table C.5 Reporting of age at death in days . 282 Table C.6 Reporting of age at death in months. 283 APPENDIX D RESULTS ACCORDING TO OLD PROVINCES Table D.2.3 Educational attainment of household population . 285 Table D.2.4 School attendance ratios . 286 Table D.2.7 Wealth quintiles. 286 Table D.2.8 Birth registration of children under age five . 287 Table D.3.3 Educational attainment . 287 Table D.3.4 Literacy. 288 Table D.3.5 Exposure to mass media. 288 Table D.3.6 Employment status. 289 Table D.3.6 Occupation. 289 Table D.4.2 Fertility by old province . 290 Table D.4.6 Birth Intervals. 290 Table D.4.8 Median age at first birth . 290 Table D.4.9 Teenage pregnancy and motherhood. 291 Table D.5.4 Current use of contraception by background characteristics . 291 Table D.5.11 Exposure to family planning messages . 292 Table D.6.2 Number of co-wives and wives . 292 Table D.6.4 Median age at first marriage. 293 Table D.6.6 Median age at first sexual intercourse. 293 Table D.6.7 Recent sexual activity. 294 Table D.6.9 Median duration of postpartum insusceptibility by background characteristics . 294 Table D.7.2 Desire to limit childbearing . 295 Table D.7.3 Need for family planning among currently married women. 295 Table D.7.5 Mean ideal number of children. 295 Table D.7.7 Wanted fertility rates. 296 Table D.8.1 Antenatal care. 296 Table D.8.3 Components of antenatal care . 297 Table D.8.4 Tetanus toxoid injections . 297 Table D.8.5 Place of delivery . 297 Table D.8.6 Assistance during delivery . 298 Table D.8.7 Delivery characteristics . 298 Table D.8.8 Postnatal care . 299 Table D.8.10 Vaccinations . 299 Table D.8.11 Prevalence and treatment of symptoms of ARI and fever. 300 Table D.8.12 Prevalence of diarrhea . 300 Table D.8.13 Knowledge of ORS packets . 300 Table D.8.14 Diarrhea treatment . 301 Table D.8.16 Problems in accessing health care . 301 Table D.9.1 Household possession of mosquito nets . 302 Table D.9.2 Use of mosquito nets by children. 302 Tables and Figures | xvii Table D.9.3 Use of mosquito nets by women. 303 Table D.9.4 Use of Intermittent Preventive Treatment by women during pregnancy. 303 Table D.9.6 Prevalence and prompt treatment of children with fever . 304 Table D.9.7 Type and timing of antimalarial drugs taken by children with fever . 304 Table D.10.1 Initial breastfeeding. 304 Table D.10.3 Median duration and frequency of breastfeeding . 305 Table D.10.5 Iodization of household salt . 305 Table D.10.6 Micronutrient intake among children . 305 Table D.10.7 Micronutrient intake among mothers . 306 Table D.10.8 Prevalence of anemia in children . 306 Table D.10.9 Prevalence of anemia in women . 307 Table D.10.11 Prevalence of anemia in men. 307 Table D.10.12 Nutritional status of children . 307 Table D.10.13 Nutritional status of women . 308 Table D.11.2 Early childhood mortality rates . 308 Table D.11.4 Perinatal mortality. 308 Table D.13.1 Experience of beatings or physical mistreatment . 309 Table D.13.3 Violence during pregnancy . 309 Table D.13.5 Marital violence . 309 Table D.13.6 Frequency of spousal violence . 310 Table D.14.1 Knowledge of AIDS. 310 Table D.14.2 Knowledge of HIV prevention methods. 310 Table D.14.3 Comprehensive knowledge about AIDS . 311 Table D.14.4 Knowledge of prevention of mother to child transmission of HIV . 311 Table D.14.5 Accepting attitudes toward those living with HIV/AIDS. 312 Table D.14.6 Attitudes toward negotiating safer sexual relations with husband . 312 Table D.14.7 Adult support of education about condom use to prevent AIDS . 313 Table D.14.8 Multiple sexual partners and higher-risk sexual intercourse in the past 12 months . 313 Table D.14.9 Coverage of prior HIV testing . 314 Table D.14.10 Pregnant women counseled and tested for HIV. 314 Table D.14.11 Self-reported prevalence of sexually-transmitted infections (STIs) and STI symptoms . 315 Table D.14.12 Prevalence of injections . 315 Table D.14.13 Comprehensive knowledge about AIDS and of a source of condoms among youth . 316 Table D.14.14 Age at first sexual intercourse among youth. 316 Table D.14.16 Premarital sexual intercourse and condom use during premarital sexual intercourse among youth . 317 Table D.14.17 Higher-risk sexual intercourse among youth and condom use at last higher-risk intercourse in the past 12 months . 317 Table D.14.19 Drunkenness during sexual intercourse among youth. 318 Table D.15.4 HIV prevalence. 318 Table D.15.8 HIV prevalence among young people . 318 Table D.15.12 HIV prevalence among couples. 319 Table D.16.1 Children’s living arrangements and orphanhood . 319 Table D.16.2 Orphans and vulnerable children (OVC) . 320 xviii | Tables and Figures Table D.16.3 School attendance by survivorship of parents and by OVC status . 320 Table D.16.4 Underweight orphans and vulnerable children. 321 Table D.16.6 Succession planning. 321 Table D.16.7 Widows dispossessed of property. 321 Table D.16.8 External support for chronically ill persons . 322 Table D.16.9 External support for orphans and vulnerable children. 322 Foreword | xix FOREWORD In the context of its desire to obtain a database designed to provide reliable indicators to monitor and assess the implementation of the country’s sector programs and policies, the Poverty Reduction Strategy, Vision 2020 and the commitments it has undertaken at the international level, in particular the Millennium Development Goals, the Government of Rwanda has just completed the Third Demographic and Health Survey (EDSR-III 2005). EDSR-III follows the surveys that were successfully conducted in 1992 and 2000, and is part of a broad, worldwide program of socio-demographic and health Surveys conducted in developing countries since the mid-1980’s. In addition to the indicators on fertility, family planning, and maternal and child health which the Survey normally provides, the main innovation of EDSR-III was the integration of a survey module on the seroprevalence of HIV and anemia as well as a module on domestic violence. As such, for the first time, the survey allowed us to determine the prevalence of HIV at the national level. Using this report, the reader will be better able to delineate the socio-demographic challenges the country faces and that it will have to meet, in particular: a maternal and infant mortality rate which remains high despite being in decline, poor utilization of childbirth and post-natal services, a continually high fertility rate, which places pressure on social costs and slows the pace of development, poor utilization of modern contraceptive methods, as well as an alarming nutritional status, above all among children under five years of age and their mothers. The reader could also be alerted to the fact that certain population groups are particularly impacted by a high prevalence of anemia or HIV. Most of these indicators can be improved by increased awareness and heightened responsibility within a couple or among individuals. Without this, the State’s investments would have limited impact. This Survey also draws attention to indicators of an appreciable level that will require strengthening of sustained efforts to maintain, if not to improve, trends. This is particularly the case with regard to the high level of breastfeeding, prenatal visits, vaccination rates of children under five years of age (except for the city of Kigali), and the use of iodized salt. The results of EDSR-III 2005 are thus extremely important because they allow us to assess the progress made in meeting the challenges mentioned above. The results also make it possible to readjust intermediate objectives, identify areas requiring priority attention, and even make projections of future socio-demographic development. The same results represent a daunting challenge to entities providing development financing and call for integrated financing approaches involving multiple sectors of socio- economic life. Accordingly, the Government of Rwanda and in particular the Ministry of Finance and Economic Planning is pleased to provide reliable results to policymakers, planners, and other users in both the public and private sector, based on the current context of the country. May this document be a source of valuable and useful information to all those individuals and organizations active in development who will use it to contribute to an improved quality of life for Rwanda’s population. Signed in Kigali on May 12, 2006 Monique Nsanzabaganwa Minister of State in Charge of Economic Planning at the Ministry of Finance and Economic Planning Acknowledgements | xxi ACKNOWLEDGMENTS This report would not have materialized without the participation of a large number of individuals and organizations. We would like to express our profound thanks to them. First, we extend our thanks to the men and women who generously agreed to respond to all of the questions submitted to them. There was a high response rate both from men (99.2%) and women (98.1%). We would like to express our sincere appreciation to the various Ministries for facilitating the implementation of the Survey. We offer our profound gratitude to the Ministry of Health for its cooperation during the preparation and completion of the survey. We also offer our sincere thanks to the Ministry of Local Government, Good Governance, Community Development and Social Affairs as well as to all of the provincial and district authorities for their assistance and their contribution to the smooth implementation of the Survey. Certainly, without the ongoing support of these various authorities, EDSR- III 2005 could not have been achieved. We also express our gratitude to the International Organizations for their vital financial assistance. Financial contributions from the United States Agency for International Development (USAID/Rwanda), the World Bank through the Support for the Multisectoral AIDS Project (MAP) and through the National AIDS Control Commission (CNLS), the Department For International Development (DFID), the United Nations Children’s Fund (UNICEF), the United Nations Population Fund (UNFPA), and the German Technical Cooperation enterprises (GTZ) to the EDSR-III budget were of immense significance to the effective accomplishment of the survey. We hereby express our profound gratitude to the team from ORC Macro, in particular Mr. Mohamed Ayad, responsible for drafting the project and technical coordination, Mrs. Fern Greenwell, ORC Macro Technical Advisor to EDSR-III 2005, Mr. Noah Bartlett, technical advisor for drafting the reports, and the other ORC Macro officers who contributed to the success of EDSR-III 2005 for their much appreciated technical assistance. The high quality of the analyses presented in this report is evidence of their support. We deeply appreciate the specific technical support of the CNLS, the Treatment and Research Aids Center (TRAC), and the National Reference Laboratory (LNR). Their active participation throughout the conduct of the survey demonstrated the effectiveness of the excellent collaboration between the country’s various institutions. The Third Demographic and Health Survey would not have been accomplished without the unfailing participation of the officers from the National Institute of Statistics who were relentlessly involved, in particular Mr. Philippe Gafishi Ngango, National Director of EDSR-III 2005, Mrs. Apolline Mukanyonga, Technical Director, and Mrs. Athanasie Kabagwira, Associate Technical Director, who, in cooperation with supervisors and administrative support personnel, supplied pertinent technical supervision and contributed to the analysis of the results. We warmly congratulate the cartographers, team leaders, monitors, and the men and women who conducted the surveys, as well as the drivers who were able to overcome the challenges and fatigue inherent in this type of operation. xxii | Acknowledgements We wish to reiterate our sincere thanks to all those, far and wide, who contributed to the completion of this Survey. Lastly, we offer our profound appreciation to the men and women who will use this document, as they have understood the ultimate aim of the production of this valuable report. Dr. Ir. Louis Munyakazi Managing Director of the National Institute of Statistics of Rwanda Abbreviations | xxiii ABBREVIATIONS AD Age at death AIDS Acquired Immunodeficiency Syndrome ANC Antenatal Care AQ Amodiaquine ARI Acute Respiratory Infection ASFR Age-specific Fertility Rate BCG Bacillus of Calmette and Guérin (vaccine against tuberculosis) BMI Body Mass Index CBR Crude Birth Rate CDC Centers for Disease Control and Prevention CNLS Commission Nationale de Lutte contre le Sida CSPro Census and Survey Processing CTS Conflict Tactics Scale DFID Department For International Development DHS Demographic and Health Surveys DPT Diphtheria-Pertussis-Tetanus vaccine EA Enumeration area EDSC Cameroon Demographic and Health Survey EDSBF Burkina Faso Demographic and Health Survey ENF Enquête Nationale sur la Fécondité (National Fertility Survey) EPI Expanded Program of Immunization ESD Enquête sociodémographique (Sociodemographic Survey) FP Family Planning FRw Rwandan Franc GAR Gross Attendance Ratio GDP Gross Domestic Product GFR General Fertility Rate GPI Gender Parity Index GTZ German Technical Cooperation HIV Human Immunodeficiency Virus IEC Information/Education/Communication INSR Institut National de la Statistique du Rwanda IPT Intermittent Preventive Treatment ITN Insecticide-Treated Mosquito Net IUD Intra Uterine Device xxiv | Abbreviations LAM Lactational Amenorrhea Method LNR National Reference Laboratory MAP Multi-country AIDS Program MDG Millennium Development Goals MMR Maternal Mortality Ratio NAR Net Attendance Ratio NCHS National Center for Health Statistics ORS Oral Rehydration Salts ORT Oral Rehydration Therapy OVC Orphaned and Vulnerable Children PNILP Programme National Intégré de Lutte contre le Paludisme (National Malaria Control Program) PRSP Poverty Reduction Strategy Papers PSU Primary Sampling Units RBM Roll Back Malaria RDHS-I First Rwanda Demographic and Health Survey, 1992 RDHS-II Second Rwanda Demographic and Health Survey, 2000 RDHS-III Third Rwanda Demographic and Health Survey, 2005 RGPH Recensement Général de la Population et de l’Habitat (General Population and Housing Census) RHF Recommended Home Fluids SDM Standard Days Method SP Sulfadoxine-Pyrimethamine STI Sexually Transmitted Infection TFR Total Fertility Rate TRAC Treatment and Research AIDS Center TWFR Total Wanted Fertility Rate UNFPA United Nations Population Fund UNDP United Nations Development Programme UNICEF United Nations Children’s Fund USAID United States Agency for International Development USD United States Dollars VCT Voluntary Counseling and Testing Center VIP Ventilation-Improved Pit Latrine WHO World Health Organization YSD Years since death Summary of Findings | xxv SUMMARY OF FINDINGS A total of 10,644 households were selected in the sample for the third Demographic and Health Survey in Rwanda (RDHS-III), and 10,307 of these were contacted at the time of the survey. The survey teams were able to interview individuals in 10,272 households, for a response rate of nearly 100 per- cent. In the 10,272 households surveyed, 11,539 women age 15 to 49 were considered eligible for the individual interview and 11,321 were successfully interviewed. Thus the response rate for women was 98 percent. The male survey was conducted in one out of every two households. A total of 4,959 men age 15 to 59 were identified in the subsample of households. Of the 4,959 men eligible for the indi- vidual interview, 4,820 were successfully inter- viewed, for a response rate of 97 percent among men. The survey results show that 44 percent of the women interviewed were age 15 to 24 at the time of the survey and 43 percent of men were in that age group. Two out of every five women and about one out of two men had never been married. These data indicate that Rwanda’s population is generally young, a fact that needs to be taken into considera- tion by policymakers when designing national de- velopment programs. The proportion of women with no formal education (23 percent) is higher than that of men (17 percent). Only 10 percent of women and 12 percent of men have at least a secondary level of education. The proportion of men and women who do not know how to read is 22 percent and 29 per- cent, respectively. At the national level, more than two out of five women (44 percent) and about one out of five men (19 percent) do not have access to any mass media. Only 8 percent of women and 10 percent of men read a newspaper at least once a week. Very few Rwandan households have electric- ity (5 percent). In rural areas less than 2 percent of households have electricity, compared with 25 per- cent in urban areas. Thirty-nine percent of urban households and 71 percent of rural households do not have reliably clean, potable water (tap water, boreholes, or protected wells). Two out of three households (67 percent) use uncovered latrines. A total of 5 percent have no toilet facilities. FERTILITY Analysis of the 2005 RDHS-III data indicates that the fertility rate of Rwandan women remains high. The total fertility rate (TFR) is 6.1 children for all women, 4.9 for urban women, and 6.3 for rural women. The main background variables for which dif- ferentials in fertility rates can be seen are urban-rural residence, marital status, education, and household wealth. Among the provinces, North and West prov- inces have the highest fertility rates while South province has the lowest. Fertility among adolescent women is negligi- ble, accounting for only 3 percent of national fertil- ity. Women older than age 40 account for 12 percent of fertility. The mean number of children ever born (CEB) among all women age 40 to 49 is 6.6 children per woman. Among urban residents in this age- group, the mean number of children ever born is 5.8, compared with 6.7 for rural residents. Regarding trends in fertility, the youngest and oldest age groups (15-19 and 40-49 years) show a decline from one survey to the next. Women age 20 to 39 account for the largest increase in fertility. A comparison of TFRs across past and current surveys indicates that the fertility rate in Rwanda stabilized at about 6 children per woman in 1992. FAMILY PLANNING Knowledge of Contraception. While almost all married women are aware of contraception, mod- ern methods in particular (98 percent in 2005, com- pared with 97 percent in 2000), relatively few women use them. xxvi | Summary of Findings Knowledge of contraception among men is also almost universal: 98 percent of male respon- dents reported that they knew of at least one modern contraceptive method and 77 percent said they knew of traditional methods. Contraceptive Prevalence. Contraceptive prevalence among currently married women is 17 percent, with 10 percent using modern methods. However, the proportion of married women using contraception has increased in the five years since the last DHS survey in Rwanda (RDHS-II), rising from 13 percent in 2000 to 17 percent in 2005 for all meth- ods and from 4 percent in 2000 to 10 percent in 2005 for modern methods. The modern methods most often used are injectables (5 percent) and pills (2 percent). The survey results show that contraceptive use is lowest among the youngest and oldest age groups: 7 percent for women age 15-24 and 10 percent for women age 45-49. MARRIAGE Forty-nine percent of women age 15-49 were in a union at the time of the survey. The proportion of never-married women decreases as age increases and it is rare to find a woman over 45 who has never been married (2 percent). Therefore, marriage, which remains practically the sole context of pro- creation in Rwanda, is very common. Twelve per- cent of Rwandan women live in polygamous house- holds. Rwandan women tend to marry late: only 19 percent of those age 25 to 49 married before age 18. The median age of first union for women is 20.7 years; the median age of first sex is 20.3 years. Men tend to marry at an older age than women. The median age for the first marriage for men is 25.0 years; the median age of first sex is 20.8 years. FERTILITY PREFERENCES Regarding fertility preferences, 42 percent of women reported that they did not wish to have any more children, while over half (52 percent) wanted a(nother) child. Among the latter group, 12 percent wanted to have the next child within two years, 39 percent wanted a child after two years, and 2 percent wished for another child without specifying the timeframe. The percentage of men (44 percent) who do not want any more children is similar to that of women. Forty percent said they wanted to wait two or more years for another child. The average ideal family size for all women, as well as for married women, is about 4 children. This ideal family size is less than the TFR of 6.1, a finding that partially explains the high percentage of women who do not want to have more children. MATERNAL AND CHILD HEALTH Antenatal Care. The vast majority of expec- tant mothers receive some antenatal care (94 per- cent). However, only 13 percent go for at least four antenatal care (ANC) visits, as recommended by the World Health Organization (WHO) and the Rwan- dan government. The first ANC visit tends to be late in the pregnancy: a median of 6.4 months into the pregnancy. During the ANC consultations, women were rarely informed of the signs of complications that may occur during pregnancy (6 percent). Most often, women were weighed (94 percent) and had their blood pressure taken (71 percent). About half the women (56 percent) said their height was measured. However, routine blood and urine tests were rare. A small percentage of women took iron supplements (28 percent) or antimalaria medication (6 percent). Delivery Care. A large proportion of Rwan- dan women give birth at home (70 percent). Six out of ten were not assisted by trained health providers; 43 percent were assisted by untrained traditional birth attendant, while 17 percent reported giving birth without any assistance. Vaccination Coverage. The objective of Rwanda’s Expanded Program on Immunization—to vaccinate all children within their first 12 months of life—has not yet been met. Only 75 percent of chil- dren age 12-23 months have received all the recom- mended vaccinations. Among these children, only 69 percent received all vaccinations before the age of one year. The dropout rate between the first and third doses of DPT was 10 percent, while for polio it was 13 percent. Childhood Illness. During the two weeks pre- ceding the survey, 17 percent of children under five had acute respiratory infection (ARI); 26 percent had fever, and 14 percent had diarrhea. Summary of Findings | xxvii Medical treatment or advice was sought for 27 percent of the children with ARI or a fever. For those who had diarrhea, only 14 percent received medical treatment. The great majority of mothers (87 percent) know about oral rehydration salts (ORS) for treat- ment for diarrhea. However, during the last episode of diarrhea, only 32 percent of children received either ORS, recommended home fluids, or increased fluids. A similar proportion of children were treated with traditional remedies. Thus, 33 percent of chil- dren with diarrhea received no treatment at all. NUTRITION Breastfeeding Practices. In Rwanda, breast- feeding is nearly universal and is of relatively long duration. Virtually all children under six months are breastfed and at 10-11 months, 97 percent are still being breastfed. The recommendation of exclusive breastfeeding for children up to six months of age is followed by nine out of ten mothers (88 percent). The median duration of breastfeeding is 24.9 months. It is very unusual to see other liquids or com- plementary food being introduced before the age of two months (5 percent). However, the recommended introduction of solid foods at six months is not gen- erally followed: only 69 percent of children age 6-9 months were receiving complementary foods Nutritional Status. More than four out of ten children under age five (45 percent) have chronic malnutrition (stunting) and nearly one out of five (19 percent) has the most severe form. Levels of stunting increase rapidly with age; the highest proportion is found among children age 12-23 months (55 per- cent), but remains fairly high (51 to 53 percent) among older children. The rate of stunting is highest in the North province (52 percent). Stunting tends to be lower among children of mothers with more edu- cation: 50 percent among those with no education, 44 percent among those with primary education, and 43 percent among those of at least secondary level. An estimated 4 percent of children have acute malnutrition (wasting), and 1 percent have the most severe form. The highest prevalence of wasting (9 percent) is found among children age 12-23 months. This corresponds to the period during which the child is most likely to be weaned and is, there- fore, vulnerable to illnesses (such as those linked to the introduction of foods that may have become con- taminated, or picked up as the child crawls around exploring the environment). Interestingly, rates of wasting in the City of Kigali (8 percent) are higher than in the other areas surveyed. Findings show that 22 percent of children in Rwanda are underweight and 4 percent are severely underweight. These figures indicate either chronic or acute malnutrition. Overall, the survey indicates that 56 percent of children age 6-59 months are anemic: 20 percent are mildly anemic, 27 percent are moderately anemic, and 9 percent are severely anemic. In general, chil- dren in urban and rural areas have similar levels of anemia, although the prevalence of severe anemia is higher in urban areas than in rural areas (13 percent versus 8 percent). Children in the City of Kigali experience higher levels of anemiaparticularly the severest form than other children. Women in Rwanda are less afflicted with anemia than the children. Nationally, 33 percent of women suffer from anemia: 19 percent are mildly anemic, 11 percent are moderately anemic, and 3 percent are severely anemic. As with children, ane- mia rates are similar in urban or rural areas; how- ever, women of the City of Kigali have a higher prevalence of moderate and severe anemia than women elsewhere. Vitamin Supplements. The survey results showed that 84 percent of last-born children age 0-3 years had received vitamin A supplements. How- ever, only 33 percent of mothers received vitamin A within the two months following delivery. Also, 71 percent of women did not receive any iron sup- plements during their pregnancy and 24 percent re- ceived supplements for no more than 3 months. Nearly nine out of ten women and children live in households that use adequately iodized salt (15+ ppm). MALARIA Possession of Mosquito Nets. In Rwanda, 18 percent of households own at least one mosquito net. Urban residents, especially in the City of Kigali, xxviii | Summary of Findings show a higher rate (40 percent) of households with at least one net than do rural residents. The percent- age is highest among the wealthiest households (45 percent versus 6 percent among the poorest). However, only 6 percent of the total of households own more than one mosquito net. Overall, almost all households with at least one mosquito net have a net that has been treated at some time. However, there is a difference between those with at least one net and those with an insecti- cide-treated mosquito net (ITN) (18 percent versus 15 percent). The same differential is observed among the households with more than one net (6 percent) and those with more than one ITN (4 percent). Mosquito Net Usage: Only 16 percent of children under the age of five slept under a mosquito net the night before the survey. Among pregnant women, 20 percent slept under a net the night before the survey. INFANT AND CHILD MORTALITY Childhood mortality remains high in Rwanda. Data for the five-year period preceding the survey indicate that for every 1,000 live births, 86 children die before their first birthday (37 between birth and 1 month and 49 between 1 and 12 months). For every 1,000 children who survive to age one, 72 do not reach their fifth birthday. Overall, the mortality risk between birth and exact age five is 152 per 1,000 live births. The RDHS-III results indicate a significant decline in infant and child mortality since the 2000 RDHS-II. However, comparison with the RDHS-I shows that the 2005 infant and under-five mortality rates have actually just returned to the 1992 levels. MATERNAL MORTALITY Maternal mortality continues to be high in Rwanda. According to the RDHS-III, the rate of ma- ternal mortality is about 750 deaths for every 100,000 live births. However, this level of mortality shows a considerable decline since the 2000 RDHS-II, which indicated a maternal mortality rate of 1,071 between 1995 and 1999. DOMESTIC VIOLENCE About one-third of women interviewed (31 percent) reported that they had been victims of physical violence at least once since they were age 15, and 19 percent reported experiencing violence during the past 12 months. Most often, it was the husband or partner who was responsible for the vio- lence. Whether physical or sexual, the violence resulted in serious consequences for the woman: in 22 percent of cases in the past 12 months, the women suffered bruises or wounds, and in 14 per- cent of cases, they experienced bone fractures. In 7 percent of cases, the women had to be treated by a doctor or were treated at a health care facility. STI AND HIV/AIDS-RELATED KNOWL- EDGE, ATTITUDES AND BEHAVIORS Almost all respondents reported that had heard of HIV/AIDS, but only 54 percent of women and 58 percent of men had a comprehensive knowledge of the disease. The level of knowledge about ways to avoid contracting HIV is insufficient: only 73 percent of women and 80 percent of men know it is possible to reduce the risk of getting the AIDS virus by using condoms and by limiting sex to one faithful and un- infected partner. Only 51 percent of men and 46 percent of women expressed positive attitudes towards people living with HIV/AIDS, indicating that the level of stigmatization and discrimination remains high in Rwanda. The survey shows that 8 percent of women and 14 percent of men reported having had higher- risk sex (intercourse with a nonmarital, noncohabit- ing partner). However, only 20 percent of these women and 41 percent of these men had used con- doms during the last higher-risk sex. Among pregnant women, only 22 percent reported receiving counseling on HIV/AIDS during their antenatal care visits or having been tested for HIV and received the results. Summary of Findings | xxix Among youth age 15-24, 51 percent of women and 54 percent of men had comprehensive knowl- edge of HIV/AIDS, and 12 percent of men and 7 percent of women used a condom during their first sexual intercourse. HIV PREVALENCE HIV Testing Rates. Overall, 97 percent of eligible respondents provided blood for HIV testing. The coverage rate was 94 percent in urban areas and 97 percent in rural areas. HIV Prevalence Rates. Survey results indi- cate that 3 percent of adults age 15-49 are infected with HIV. The prevalence rate is higher among women than among men; the ratio of women to men is 1.6. HIV prevalence is significantly higher in urban areas than in rural areas. Among all those age 15-49, the City of Kigali shows the highest level of HIV prevalence (6.7 percent). Among those age 15-24, HIV prevalence in Kigali is 3.4 percent. The lowest HIV prevalence is in the North province (2 percent). According to age and sex, the prevalence of HIV is highest among men age 40 to 44 (7.1 per- cent) and among women age 35 to 39 (6.9 percent). HIV and Associated Factors. HIV preva- lence is very high among respondents who declared having contracted a sexually transmitted infection in the 12 months prior to the survey (15.7 percent). Prevalence is also high among widowed women (15.9 percent) and divorced or separated women (10.9 percent). Fifty-six percent of men and 64 percent of women who tested positive for HIV at the time of the survey had never had an HIV test previously. CARE AND SUPPORT FOR VULNERABLE PERSONS Approximately one out of five children under the age of 18 is an orphan: 4 percent have lost both parents, 13 percent have lost their father, and 3 per- cent have lost their mother. Around 11 percent of children in Rwanda are considered to be vulnerable. Overall, 29 percent of children under age 18 can be classified as orphans or vulnerable children (OVC). The highest proportion of OVC is in the City of Kigali (35 percent) and the lowest is in the North province (25 percent). The RDHS-III results have shown that paren- tal survival status influences the school attendance of children age 10-14. When both parents are alive and the child lives with at least one parent, 91 per- cent attend school. School attendance drops to 75 percent when both parents are dead. In Rwanda, OVC do not seem to suffer more from malnutrition than other children, regardless of age or sex. A ratio of less than 1.0 (0.92) indicates that non-OVC are slightly more likely to be under- nourished than OVC. Early sexual relations seem to be slightly more frequent among OVC (6 percent among girls and 15 percent among boys) than among other children (5 percent among girls and 14 percent among boys). Very few Rwandan households have received assistance to care for sick family members. In only 12 percent of cases did households receive any assis- tance, whether medical, social, material or emo- tional. Less than 1 percent of households received all of these forms of assistance. In 87 percent of cases, households in Rwanda received no external support in caring for OVC. The external assistance that is provided tends to be for paying school fees (9 percent of households). Other types of support are virtually nonexistent. Millennium Development Goal Indicators | xxxi Millennium Development Goal Indicators, Rwanda 2005 Goal Indicator Value 1. Eradicate extreme poverty and hunger Prevalence of underweight children under five years of age Male: 22.9 % Female: 22.1 % Total: 22.5 % Net enrolment ratio in primary education1 Male: 73.8 % Female: 76.6 % Total: 75.2 % Percent of pupils starting grade 1 who reach grade 51 Male: 9.6 % Female: 10.3 % Total: 10.0 % 2. Achieve universal primary education Literacy rate of 15-24 year-olds2 Male: 67.8 % Female: 65.2 % Total: 66.0 % Ratio of girls to boys in primary and secondary education Primary: 1.03 Secondary: 0.81 Ratio of literate women to men, 15-24 years old2 0.96 3. Promote gender equality and empower women Share of women in wage employment in the non-agricultural sector3 8.8 % Under-five mortality rate (per 1,000 live births) 152 per 1,000 Infant mortality rate (per 1,000 live births) 86 per 1,000 4. Reduce child mortality Percent of 1 year-old children immunized against measles Male: 84.9 % Female: 86.4 % Total: 85.6 % Maternal mortality ratio (per 100,000 live births) 750 per 100,000 5. Improve maternal health Percent of births attended by skilled health personnel 38.6 % Condom use to overall modern contraceptive use among currently married women age 15-49 9.2 % Condom use at last higher-risk sex (population age 15-24)4 Male: 39.5 % Female: 26.0 % Percentage of population age 15-24 with comprehensive correct knowledge of HIV/AIDS5 Male: 53.6 % Female: 50.9 % Contraceptive prevalence rate (any modern method, currently married women age 15-49) 10.3 % 6. Combat HIV/AIDS, malaria and other diseases Ratio of school attendance of orphans to school attendance of non- orphans aged 10-14 years 0.82 Percent of population using solid fuels6 Urban: 98.3 % Rural: 99.8 % Total: 99.6 % Percent of population with sustainable access to an improved water source7, urban and rural Urban: 55.0 % Rural: 22.4 % Total: 27.4 % 7. Ensure environmental sustainability Percent of population with access to improved sanitation8, urban and rural Urban: 97.2 % Rural: 96.5 % Total: 96.6 % 1 Excludes children with parental status missing. 2 Refers to respondents who attended secondary school or higher and women who can read a whole sentence. 3 Wage employment includes respondents who receive wages in cash or in cash and kind. 4 Higher risk refers to sexual intercourse with a partner who neither was a spouse nor who lived with the respondent; time frame is 12 months preceding the survey. 5 A person is considered to have a comprehensive knowledge about AIDS when they say that use of condoms for every sexual intercourse and having just one uninfected and faithful partner can reduce the chance of getting the AIDS virus, that a healthy-looking person can have the AIDS virus, and when they reject the two most common local misconceptions. The most common misconceptions in Rwanda are that AIDS can be transmitted through mosquito bites and that a person can become infected with the AIDS virus by sharing food with someone who is infected. 6 Charcoal, firewood, or sawdust. 7 Improved water sources are: household connection (piped), public standpipe, borehole, or protected dug well. 8 Improved sanitation technologies are: flush toilet, traditional pit latrine, or ventilated improved pit latrine. xxxii | Map of Rwanda Country Profile and Survey Introduction | 1 COUNTRY PROFILE AND SURVEY INTRODUCTION 1 1.1 COUNTRY PROFILE 1.1.1 Geography The country of Rwanda is situated in central Africa immediately south of the equator between 1°4' and 2°51' south latitude and 28°63' and 30°54' east longitude. Its total area of 26,338 square kilometers is bordered by Uganda to the north, Tanzania to the east, the Democratic Republic of the Congo to the west, and Burundi to the south. Landlocked, Rwanda lies 1,200 kilometers from the Indian Ocean and 2,000 kilometers from the Atlantic Ocean. Rwanda forms part of the highlands of eastern and central Africa, with mountainous relief and an average elevation of 1,700 meters. However, there are three distinct geographical regions. Western and north-central Rwanda is made up of the mountains and foothills of the Congo-Nile Divide, the Virunga volcano range, and the northern highlands. This region is characterized by rugged mountains intercut by steep valleys, with elevations generally exceeding 2,000 meters. The Divide itself rises to 3,000 meters at its highest point but is dwarfed by the volcano range, whose highest peak, Kalisimbi, reaches 4,507 meters. The Congo-Nile Divide slopes westward to Lake Kivu, which lies 1,460 meters above sea level in the Rift Valley trough. In Rwanda’s center, mountainous terrain gives way to the rolling hills that give the country its nickname, “Land of a Thousand Hills.” Here the average elevation varies between 1,500 and 2,000 meters. This area is also referred to as the central plateau. Further east lies a vast region known as the “eastern plateaus,” where the hills level gradually into flat lowlands interspersed with a few hills and lake-filled valleys. The elevation of this region generally falls below 1,500 meters. Due to its elevation, Rwanda enjoys a temperate, sub-equatorial climate with average yearly temperatures of around 18.5°C. The average annual rainfall is 1,250 millimeters and occurs in two rainy seasons of differing lengths, alternating with one long and one short dry season. The climate varies somewhat from region to region, depending on the altitude, the volcano range and northern highlands being generally cooler and wetter, with average temperatures of 16°C, and average rainfall of above 1,300 millimeters. The maximum rainfall is 1,600 millimeters, above the Divide and the volcanic range. The hilly central region receives an average of between 1,000 and 1,300 millimeters of rain per year, while rainfall on the eastern plateau, whose climate is relatively warmer and drier, generally falls below 1,000 millimeters and can be as low as 800 millimeters. Although Rwanda enjoys more or less constant temperatures, the climate is known to vary from year to year, with extreme variations in rainfall sometimes resulting in flooding or, more often, drought. These extremes have a profound impact on agricultural production, which sometimes falls into recession. Rwanda has a dense network of rivers and streams, draining into the Congo River on the western slope of the Congo-Nile Divide, and into the Nile in the rest of the country via the Akagera River, which 2 | Country Profile and Survey Introduction receives all the streams of this watershed. Water resources also include several lakes surrounded by wetlands. Deforestation due primarily to land clearing for agricultural expansion has resulted in mostly anthropic vegetation with only a few small areas of natural forestland (representing 7 percent of the country) remaining on the Congo-Nile Divide and the slopes of the volcanic range. It should be noted that at the time the survey was conducted in 2005, the country was divided into 11 provinces and the City of Kigali, with the provinces being further subdivided into districts, sectors and cells. Since then, the country’s administrative structure and associated terminology have changed: there are now four geographically-based provinces (North, South, East, and West) and the City of Kigali, these being further subdivided into 30 districts, 415 sectors, cells and, finally, villages (Imidugudu). This report is based on the new administrative divisions (four provinces and the City of Kigali) but also includes the former names (11 provinces and the City of Kigali) for purposes of clarity in referring to the sample and to assist readers as yet unfamiliar with the new administrative entities. 1.1.2 Economy Although regular efforts have been made to develop the service sector and stimulate investment in the industrial sector, the Rwandan economy remains dominated by agriculture. According to the 2002 General Population and Housing Census (RGPH), more than 8 out of 10 people are employed in agriculture, including 81 percent of men and 93 percent of women. However, the agricultural sector is facing major problems: a production system dominated by small farming operations of less than one hectare, rudimentary techniques, and a low rate of investment. Agrarian reforms are being gradually introduced to address these problems, in particular through population resettlement and labor quality improvements focusing on specialized training mainly for women. Efforts are also underway to regionalize crops and fully expand the use of farm inputs. Agriculture accounts for the largest share of Rwanda’s Gross Domestic Product (GDP),1 roughly 45 percent in 2003, followed by services at 36 percent and industry at 19 percent at constant 1995 prices. Nevertheless, agricultural production declined by 4 percent in 2003 in relation to 2002, essentially due to poor rainfall during the two growing seasons. As a direct result, production dropped for grains (-3.4 percent), legumes (-1 percent), tubers (-10.7 percent), and bananas (-13.4 percent) over that of 2002. Among the export crops, coffee production alone dropped by 29 percent in relation to 2002. In 2003, industry value added grew by 7 percent, while mining value added declined significantly (-8.6 percent).2 At the same time, services value added increased by 4 percent in 2003 over the previous year. Financial institutions, transport and communications services, and hotels and restaurants were the main contributors to the increase in value added. The per capita GDP at constant 1995 prices was FRw 76,089 in 2003 compared with FRw 77,631 in 2002. The value added of final consumption expenditure dropped by 0.98 percent due to a decrease in private consumption expenditure, which in 2003 fell from FRw 558,293 million to FRw 537,746 million at constant 1995 francs, a decline of 3.78 percent over 2002. Government consumption expenditure increased by 10.6 percent in 2003 in relation to 2002 (Department of Statistics, 2004). 1 Republic of Rwanda, Ministery of Finances and Economic Planning, Department of Statistics: Rwanda Develop- ment Indicators 2004 2 Rwanda Development Indicators 2004 Country Profile and Survey Introduction | 3 Finally, because of the failure of most development strategies based on structural adjustment programs focusing on growth measured in terms of per capita GDP, the overwhelming majority of development partners are recognizing the need to incorporate social factors into development strategies. Therefore, new initiatives are geared toward pro-poor economic growth and poverty reduction to revive the economies of developing nations. Rwanda has also adopted this new orientation. 1.1.3 Population According to the 2002 Rwanda General Population and Housing Census (RGPH), the country’s population numbers 8,128,553 people. Although Rwanda suffered a major loss of human life (more than one million people) in the 1994 genocide, the population remains essentially the same because more than one million former refugees who had been living for years in exile returned at the end of the war and genocide. The population of Rwanda has increased steadily and rapidly from more than 2,000,000 in 1952, to 7,666,000 in 1996, to 8,128,553 in 2002. The increase is essentially due to rapid demographic growth. The 2002 RGPH estimated the natural growth rate at 2.6 percent and the fertility rate at 5.9. The rate of increase declined significantly to 1.2 percent between 1991 and 2002, compared with 3.1 percent between 1978 and 1991. Population density is high across the country and is increasing steadily: 321 inhabitants per square kilometer in 2002, compared with 283 in 1991 and 191 in 1978. The population is essentially young, with 67 percent of all Rwandans under the age of 20. In terms of gender, the 2002 RGPH shows females to be in the majority (52 percent) while males make up 48 percent of the population. The illiteracy rate remains fairly high: 36 percent of Rwandans age 15 years and older do not know how to read or write and only 4 percent of women are able to read. Sixty percent of the total population is considered literate. The education level of Rwandans age 6 years and above is also low. According to the 2002 RGPH, one in three people is completely uneducated (34 percent) and nearly 60 percent of all Rwandans have received no education beyond primary school. Only 5.8 percent have reached the secondary school level, while those receiving education beyond the secondary level make up less than 1 percent of the population. Under Article 33 of Rwanda’s current Constitution (adopted in 2003), “Freedom of thought, opinion, conscience, religion, worship, and the public manifestation thereof is guaranteed by the State in accordance with conditions determined by law.” Although numerous religions are practiced in Rwanda, Christianity is by far the dominant faith, practiced in some form by 93 percent of the resident population, the majority of whom are Catholic. In the 1991 Census, 90 percent of the resident population identified themselves as Christian. Their number has increased at the expense of those professing no religion, who have declined from 6.8 percent in 1991 to 3.6 percent in 2002. The number of Muslim adherents has risen slightly, from 1.2 percent of the population in the 1991 Census to 1.8 percent in 2002. Nearly all Rwandans speak the same language, Kinyarwanda (spoken by over 99 percent of the population), which is the country’s first official language, followed by French and English. Kiswahili, the third relatively common foreign language, is generally spoken in urban areas and in the provinces bordering on countries where this language is widely spoken (Democratic Republic of the Congo, Tanzania). 4 | Country Profile and Survey Introduction 1.1.4 Population Policy Out of concern for improving the country’s quality of life, the Rwandan government has developed various strategies over the years to ensure an acceptable balance between demographic growth and available resources, particularly since the 1980s. A family planning initiative developed in 1982 provided for training, improved access to family planning services and, in particular, the promotion of family planning through trained communicators known as Abakangurambaga (“Awakeners of the People”). A subsequent policy was adopted in 1990 aimed at curbing demographic growth and reducing fertility through family planning. To create an environment favorable to behavioral changes that result in lower fertility rates, other elements were included in the plan such as increased production, public health improvements, land use planning, training of communicators, the promotion of education and school attendance, and the employment and advancement of women. Following the 1994 genocide, population problems were seen in a new light with respect to both quality of life and population growth. A new national population policy was developed and issued to all development agents in 2003. This policy emphasizes quality of life by providing objectives and strategies used to affect both demographic (fertility, mortality) and socioeconomic factors. Concretely, it emphasizes: slowing demographic growth, managing natural resources sustainability, food safety, access to primary and secondary education for all children—with a focus on technical and vocational instruction and information technologies—good governance, equal opportunity, and participation in development by both men and women. 1.1.5 Public Health Policy The Ministry of Health, in collaboration with its partners, has just developed a policy aimed at the entire health sector. Special emphasis is placed on priority reproductive health issues such as making pregnancy safer, children’s health, family planning, sexually transmitted infections (STIs), HIV/AIDS, teenage reproductive health, prevention and response to sexual violence, and social changes aimed at increasing women’s decisionmaking power. Health indicators have shown clear improvement: the proportion of the population covered by health mutual schemes increased from 4 percent to 7 percent in one year, and the number of doctors and nurses rose by 10 percent and 7 percent, respectively. In addition, the Ministry of Health is developing incentives to encourage highly qualified medical personnel to serve in rural areas. HIV/AIDS is a major problem in Rwanda; for this reason, HIV/AIDS testing was included in the survey. HIV/AIDS affects all population strata, especially young women, sex workers, orphans, prisoners and truck drivers. The price of antiretroviral drugs continues to decline, and the prevention of mother-to- child transmission ((PMTCT) program launched in 2001 has been implemented in all provinces. Government budget allocations for health have increased substantially—by 185 percent between 2002 and 2004 (Finance Law of 2002 and 2004, Government of Rwanda). In 2004, the government allocated 6.1 percent of its budget to health (Department of Statistics, 2004). 1.2 OBJECTIVES AND METHODOLOGY OF THE SURVEY The Rwanda Demographic and Health Survey (RDHS-III, 2005) is the third of its kind, following surveys conducted in 1992 and 2000. Ordered by the Ministry of Finances and Economic Planning, it was carried out by the Department of Statistics (now known as the National Institute of Statistics of Rwanda) Country Profile and Survey Introduction | 5 with the technical assistance of ORC Macro, an American company that supervises the international Demographic and Health Surveys program through the MEASURE DHS project. Financial support for the survey was provided by the United States Agency for International Development (USAID/Rwanda), the United Nations Population Fund (UNFPA), the United Nations Children’s Fund (UNICEF), the Commission Nationale de Lutte contre le SIDA (CNLS) through the World Bank’s Multi-country AIDS Program (MAP), the British Department for International Development (DFID), and the German Technical Cooperation (GTZ). It was conducted on a representative sample of women between the ages of 15 and 49 and men between the ages of 15 and 59. 1.2.1 Objectives of the Survey The main objectives of the RDHS-III were: • At the national level, gather data to determine demographic rates, particularly fertility and infant and child mortality rates, and analyze the direct and indirect factors that determine fertility and child mortality rates and trends. • Evaluate the level of knowledge and use of contraceptives among women and men. • Gather data concerning family health: vaccinations; prevalence and treatment of diarrhea, acute respiratory infections (ARI), and fever in children under the age of five; antenatal care visits; and assistance during childbirth. • Gather data concerning the prevention and treatment of malaria, particularly the possession and use of mosquito nets, and the prevention of malaria in pregnant women. • Gather data concerning child feeding practices, including breastfeeding and, in half the households surveyed, collect anthropometric measurements to evaluate the nutritional status of women and children, and test for anemia in children under the age of five, women between the ages of 15 and 49, and men between the ages of 15 and 59. • Gather data concerning knowledge and attitudes of women and men about STIs and AIDS, and evaluate recent changes in behavior with respect to the use of condoms. • Gather data to determine adult mortality levels at the national level. • Gather quality data concerning domestic violence. • Gather data concerning the types of care and support received by those under the age of 60 who died in the 12 months preceding the survey. • Collect blood samples in half of the households surveyed to estimate the prevalence of HIV in the adult population of reproductive age—anonymous HIV testing of women age 15 to 49 and men age 15 to 59. 1.2.2 Questionnaires Three questionnaires were used in the RDHS-III: the Household Questionnaire, the Women’s Questionnaire, and the Men’s Questionnaire. The content of these questionnaires was based on model questionnaires developed by the MEASURE DHS project. Technical meetings between experts and representatives of the Rwandan government and national and international organizations were held beginning in June 2004 to discuss the content of the questionnaires. The inputs generated by these meetings were used to modify the model questionnaires to reflect the needs of users and the relevant population, family planning, HIV/AIDS, and other health issues in Rwanda. The final questionnaires were 6 | Country Profile and Survey Introduction then translated from French into English and Kinyarwanda. These questionnaires were further refined and then finalized in December 2004 after pretesting and training of field staff. The Household Questionnaire was used to list all of the usual members and visitors in the selected households. Some basic information was collected on the characteristics of each person listed, including age, sex, education, and relationship to the head of the household. The main purpose of the Household Questionnaire was to identify women and men who were eligible for the individual interview. The Household Questionnaire also collected information on characteristics of the household’s dwelling unit such as the main source of drinking water, type of toilet facilities, materials used for the floor of the house, the main energy source used for cooking, ownership of various durable goods, ownership and use of mosquito nets, and the type of salt used for cooking. In addition, questions were asked about the type of assistance or support received by vulnerable members of the population such as the very ill, and orphaned or otherwise vulnerable children. The questionnaire was also used to register people eligible for anthropometric (height and weight) measurements and the collection of samples for hemoglobin and HIV testing. The Women’s Questionnaire was used to collect information on all women of reproductive age (15-49 years) and covered a wide variety of topics, including: • Background characteristics • Reproductive history • Knowledge and use of contraceptive methods • Fertility preferences • Antenatal, childbirth, and postpartum care • Breastfeeding and child feeding practices • Vaccinations and childhood illnesses • Marriage and sexual activity • Women’s work and husband’s background characteristics • AIDS and other sexually transmitted infections • Adult mortality • Domestic violence The Men’s Questionnaire was administered to all men age 15-59 years living in every second household of the RDHS-III sample. The Men’s Questionnaire collected information similar to that of the Women’s Questionnaire, but was shorter because it did not contain questions on reproductive history, maternal and child health, or nutrition. All aspects of RDHS-III data collection, including anemia and HIV testing procedures, were pre- tested between 19 November and 15 December 2004. The 30 team members recruited received four weeks of training on the questionnaires and procedures for collecting blood samples for the anemia and HIV tests. Training in blood sample collection was provided by the Department of Statistics technical team and a representative of the National Reference Laboratory, with the assistance of ORC Macro. The training included a theory section and a practicum section both in the classroom and at health facilities in the city of Kigali. During the pilot survey, approximately 150 households were visited in urban and semi- urban clusters in the city of Kigali and Kigali Ngari. The blood sample collection acceptance rate was sufficiently high during the pretest (over 85 percent for women and men) to indicate the feasibility of conducting such samples during the survey itself. The lessons learned during this pretest were used to finalize the survey instruments and logistical arrangements. Country Profile and Survey Introduction | 7 1.2.3 Sample Design The sample for the RDHS-III survey covered the population residing in ordinary households across the country. A national sample of 10,644 households was selected. The sample was first stratified to provide adequate representation of urban and rural areas as well as all 12 provinces including the “City of Kigali,” the nation’s capital. Decentralization reforms were introduced after this sample was drawn, resulting in new geographically-based divisions that regroup the former districts into five new provinces. However, the sample used posed no obstacle to adequate representation of the new provinces and the data in this report present key indicators corresponding to the five recently created provinces (South, West, North, East, and the City of Kigali). The survey used a two-stage sample design. The first stage involved selecting primary sampling units (PSUs) based on the list of enumeration areas covered in 2002 General Population and Housing Census (RGPH) prepared by the National Census Bureau. These enumeration areas provided the master frame for the drawing of 462 clusters (351 rural and 111 urban), selected with a representative probability proportional to their size. A strictly proportional sample allocation would have resulted in a very low number of urban households in certain provinces such as Umutara. It was therefore necessary to slightly over-sample urban areas in order to survey a sufficient number of households to produce reliable estimates for urban areas. The second stage involved selecting a sample of households in these enumeration areas. In order to adequately guarantee the accuracy of the indicators, it was necessary to control the total size of the households drawn by setting the number of households to be surveyed at 20 in urban clusters and 24 in rural clusters. Because of the nonproportional distribution of the sample among the different strata and the fact that the number of households was set for each cluster, weighting was used to ensure the validity of the sample at both national and regional levels. All women age 15-49 years who were either usual residents of the selected household or visitors present in the household on the night before the survey were eligible to be interviewed (approximately 11,500 women). In addition, in a subsample of every second household selected for the survey, a sample of 5,000 men age 15-59 years was selected to be interviewed. In this subsample, all men and women eligible for the individual survey were also eligible for the HIV test. In addition, in this subsample of households, all women eligible for the survey and all children under the age of five were eligible for the anemia test. Finally, in this same subsample of households, all women eligible for the survey and all children under the age of five were eligible for height and weight measurements to determine their nutritional status. 1.2.4 Sample Coverage All of the 462 clusters selected for the sample were able to be surveyed for the RDHS-III. A total of 10,644 households were selected, of which 10,307 households were identified and occupied at the time of the survey. Among these households, 10,272 completed the Household Questionnaire, yielding a response rate of nearly 100 percent (Table 1.1). In the 10,272 households surveyed, 11,539 women age 15-49 years were identified as being eligible for the individual interview; interviews were completed with 11,321 of these women, yielding a response rate of 98 percent. Male interviews were conducted in every second household. A total of 4,959 men age 15-59 years were identified in the subsample of households. Of these 4,959 men, 4,820 completed the individual interviews, yielding a response rate of 97 percent. The response rates were slightly higher in rural areas for both men and women. 8 | Country Profile and Survey Introduction Table 1.1 Results of the household and individual interviews Number of households, number of interviews, and response rates, according to residence, Rwanda 2005 Residence Result Urban Rural Total WOMEN Household interviews Households selected 2,220 8,424 10,644 Households occupied 2,122 8,185 10,307 Households interviewed 2,107 8,165 10,272 Household 99.3 99.8, 99.7 Interviews with women Number of eligible women 2,689 8,850 11,539 Number of eligible women interviewed 2,616 8,705 11,321 Eligible woman response rate 97.3 98.4 98.1 MEN Household interviews Households selected 1,110 4,212 5,322 Households occupied 1,061 4,095 5,156 Households interviewed 1,053 4,083 5,136 Household response rate 99.2 99.7 99.6 Interviews with men Number of eligible men 1,183 3,776 4,959 Number of eligible men interviewed 1,130 3,690 4,820 Eligible man response rate 95.5 97.7 97.2 1.2.5 Hemoglobin and HIV Testing In every second household, women age 15-49 years, men age 15-59 years, and children under the age of five were eligible for the anemia test. These men and women were also eligible for the HIV test. The anemia and HIV test protocols were approved by the ORC Macro Internal Review Board in Calverton and the National Ethics Committee of Rwanda. Hemoglobin test Checking hemoglobin levels is the primary way of diagnosing anemia. This test is performed with the HemoCue system. An informed consent form is read to the eligible person or parent/responsible adult of the child or teenager between the ages of 15 and 17 years. This consent form explains the objectives of the test, informs the eligible individual (or parent/responsible adult) that the results will be communicated immediately after the test, and asks permission to conduct the test. Country Profile and Survey Introduction | 9 Before collecting the blood, the finger is cleaned with a swab dipped in alcohol and allowed to air dry. Then the tip of the finger (or heel, for children under 6 months, or under one year if very thin) is pricked with a sterile, single-use retractable blood lancet. One drop of blood was collected in a microcuvette and then introduced into the HemoCue photometer, which indicated the level of hemoglobin. These results were recorded on the Household Questionnaire and communicated to the person tested, or to the parent/responsible adult, with an explanation of their meaning. If the person presented severe anemia (hemoglobin below 7 g/dl, or 9 g/dl for pregnant women), the survey conductor provided a reference explaining how and where to seek treatment at a medical facility. HIV test The HIV test was given in the subsample of households selected for the men’s survey. Blood samples were collected from all eligible men and women who volunteered to be tested in these households. The HIV test protocol is based on the anonymous linked protocol developed by the DHS (Demographic and Health Surveys) program and approved by ORC Macro’s Internal Review Board. According to this protocol, names and other personal or geographic information that might identify an individual may not be linked to the blood sample. The anonymous linked protocol was also approved by the National Ethics Committee of Rwanda specifically for the RDHS-III. Because HIV tests are strictly anonymous, it was not possible and will not be possible to inform those surveyed of their test results. All persons eligible for the survey, whether or not they agreed to be HIV tested, received a card allowing them to obtain, if desired, counseling and free testing at a voluntary counseling and testing center (VCT). The card contained a list of 77 VCTs located throughout the country that offer free services to those who present the card. For the purposes of blood sample collection, two “survey technicians” were included on each field team to be specifically responsible for collecting blood samples. In addition to training in conducting the survey, these technicians received special training covering all aspects of the anemia and HIV test protocols. After explaining blood collection procedures, data confidentiality, and test anonymity, the technician sought to obtain the informed consent of each person eligible for the test. At that time, the eligible person was given a voucher for counseling and free testing at a VCT center. For men and women who consented to be tested, the technician collected drops of blood on a filter paper, observing all safety and hygienic precautions. In most cases, the drops of blood were collected from the same finger prick as for the anemia test. A barcode label was attached to each filter paper containing the blood sample. A duplicate label was attached to the Household Questionnaire on the line showing consent for that respondent and a third copy of the same barcode label was affixed to the Blood Sample Transmittal Form. The drops of blood on the filter paper were dried for a minimum of 24 hours in a drying box containing dessicants to absorb moisture. The next day, each dried sample was placed in a waterproof plastic Ziploc bag with a dessicant and moisture indicator for preservation. This kept the individual bags dry during transmittal from the field to the central office of the National Institute of Statistics in Kigali, where they were immediately verified and placed in a dry place prior to being logged and sent on to the National Reference Laboratory. Testing for the HIV antibody and compilation of results were performed by the National Reference Laboratory (LNR) in Kigali. The LNR undergoes rigorous internal quality audits on a regular basis as well as external quality audits. 1.2.6 Training and Data Collection Staff responsible for the survey at the National Institute of Statistics, in collaboration with the technical team, recruited approximately 95 people to participate in data collection during the main survey, 33 of whom were medically qualified to take blood samples. Four weeks of training were provided, from 10 | Country Profile and Survey Introduction 21 January to 21 February, followed by three days of practicum in urban and rural areas not selected for the main survey. After the training, the field agents were divided into 15 teams, each of which contained a team leader, a supervisor, three female interviewers and one male interviewer. One of the three female interviewers and the male interviewer also served as medical technicians. Data collection began on 28 February 2005 in the City of Kigali. This location made it possible to closely monitor the teams before they were dispatched to more distant areas. After two weeks, all teams, except for those remaining to complete the work in the City of Kigali, were deployed to their respective work zones. Data collection was completed on 13 July 2005. 1.2.7 Data Processing Data entry on personal computers began on 23 March 2005, three weeks after the survey was launched in the field. Data were entered by a team of eight data processing personnel recruited and trained for this task, assisted during these operations by 4 others. Completed questionnaires were periodically brought in from the field to the National Institute of Statistics in Kigali, where assigned agents checked them and coded the open-ended questions. Next, the questionnaires were sent to the data entry facility and the blood samples were sent to the National Reference Laboratory to be screened for HIV. Data were entered using CSPro, a program developed jointly by the United States Census Bureau, the ORC Macro MEASURE DHS+ program, and Serpro S.A. All questionnaires were entered twice to eliminate as many data entry errors as possible. In addition, a quality control program was used to detect some of the main data collection errors for each team. This information was shared with field teams during supervisory visits to improve data quality. The data entry and internal consistency verification phase of the survey was completed in October 2005. Household Characteristics | 11 HOUSEHOLD CHARACTERISTICS 2 This chapter presents information on the social, economic, and demographic characteristics of the households sampled, focusing on such background characteristics as age, sex, school attendance, and the educational attainment of the respondents, as well as the physical features of their dwellings and ownership of durable goods. The purpose of this chapter is to present a profile of the households and socioeconomic conditions in which the men, women, and children targeted by this survey live. Such descriptions are essential because socioeconomic and environmental factors are major determinants of the health status and overall living conditions of a country’s population. 2.1 HOUSEHOLD POPULATION BY AGE AND SEX Table 2.1 shows the distribution by age and sex of the household population surveyed, according to urban-rural residence. The household survey involved 46,490 respondents, of which 39,352, or 85 percent, live in rural areas and 7,139, or 15 percent, live in urban 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, Rwanda 2005 Urban Rural Total Age Male Female Total Male Female Total Male Female Total <5 17.3 15.7 16.4 19.1 16.3 17.6 18.8 16.2 17.4 5-9 14.6 13.7 14.1 17.0 14.4 15.6 16.6 14.3 15.4 10-14 12.9 11.9 12.4 14.2 13.3 13.7 14.0 13.1 13.5 15-19 10.4 12.4 11.5 11.6 10.4 11.0 11.4 10.7 11.0 20-24 11.8 11.5 11.6 8.5 9.3 8.9 9.0 9.6 9.4 25-29 8.6 8.4 8.5 5.9 6.9 6.4 6.3 7.1 6.7 30-34 6.5 6.3 6.4 4.4 5.8 5.2 4.8 5.9 5.4 35-39 4.8 5.2 5.0 3.8 4.5 4.2 4.0 4.6 4.3 40-44 4.1 4.3 4.2 3.9 4.6 4.3 3.9 4.6 4.3 45-49 3.0 2.8 2.9 3.5 3.9 3.7 3.4 3.7 3.6 50-54 1.9 2.7 2.3 2.5 3.2 2.9 2.4 3.1 2.8 55-59 1.3 1.4 1.3 1.6 1.7 1.7 1.5 1.7 1.6 60-64 0.9 1.4 1.2 1.1 1.7 1.4 1.0 1.6 1.3 65-69 0.7 1.0 0.9 0.9 1.5 1.2 0.8 1.4 1.1 70-74 0.4 0.6 0.5 0.9 0.9 0.9 0.8 0.9 0.9 75-79 0.3 0.4 0.4 0.5 0.7 0.6 0.5 0.7 0.6 80 + 0.4 0.4 0.4 0.6 0.8 0.7 0.6 0.7 0.7 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 3,316 3,822 7,139 18,446 20,906 39,352 21,762 24,727 46,490 Table 2.1 shows the distribution by age and sex of the household population as depicted by the age pyramid in Figure 2.1. The age pyramid is wide at the base, narrowing rapidly as it reaches the upper age limits, an indication of a population with high fertility and even higher mortality. In addition, there is a notable gender imbalance: 88 males for every 100 females in the total population. Further analysis reveals structural elements peculiar to the Rwandan population. First, the number of men drops off significantly in the 30-34 and 45-49 age groups, resulting in a shortage of males. The same trend occurs among females, but to a significantly lesser degree. The shortage of men and women may be attributed to the excess mortality caused by the 1994 genocide. Moreover, there is significant falloff in the 10-14 year 12 | Household Characteristics age group, a direct effect of the low birth rate during the years surrounding 1994. The higher proportion of children age 0-4 years reflects the return of fertility rates to their 1992 levels (6.2 compared with 6.1 in 2005). The overrepresentation of women overall is noted in both urban and rural areas. In rural areas, males predominate among those age 0 to 19 years. From age 20-24 on, however, the situation begins to reverse and the gap narrows. In urban areas, males age 0-4 and 10-14 outnumber females, but beginning at age 35-39, the proportion of females is slightly larger. 2.2 HOUSEHOLD SIZE AND COMPOSITION Table 2.2 shows that the mean size of a Rwandan household is 4.6 persons. This mean size varies somewhat: 4.5 in rural areas and 4.8 in urban areas. It is identical to the mean household size of 4.6 found in the previous survey, with variations of 4.5 in rural areas and 5.0 in urban areas. In addition, the results presented in Table 2.2 show that 66 percent of Rwandan households are headed by men. Female-headed households represent 34 percent of households in rural areas and nearly the same percentage in urban areas (33 percent). The percentage of female-headed households increased significantly from 21 percent to 36 percent between 1992 and 2000, but dropped again in 2005 (34 percent). Approximately half of all households contain 3 to 5 people. One-person households make up only 7 percent of the population. Only one in ten households (10 percent) contains 8 or more people. RDHS 2005 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 <5 Age 0246810 0 2 4 6 8 10 Male Female Percent Household Characteristics | 13 Table 2.2 Household composition Percent distribution of households by sex of head of household and household size, according to residence, Rwanda 2005 Residence Characteristic Urban Rural Total Sex of head of household Male 67.2 66.0 66.1 Female 32.8 34.0 33.9 Total 100.0 100.0 100.0 Number of usual members 1 8.8 6.2 6.5 2 10.1 11.3 11.1 3 13.1 17.4 16.7 4 17.7 18.4 18.3 5 14.6 16.1 15.9 6 13.5 12.8 12.9 7 9.4 8.1 8.3 8 5.3 5.3 5.3 9+ 7.6 4.4 4.9 Total 100.0 100.0 100.0 Number of households 1,510 8,762 10,272 Mean size 4.8 4.5 4.6 Note: Table is based on de jure members, i.e., usual residents. 2.3 SCHOOL ATTENDANCE AND EDUCATIONAL ATTAINMENT Tables 2.3.1 and 2.3.2 show the percent distribution of the male and female household population according to highest level of educationa attained, by age, residence, province, and household wealth quintile. Educational attainment is important; it contributes to improved living conditions not only for the individual household but for society as a whole. Reproductive behavior, the use of contraception, health habits, school attendance of household members, and habits relating to hygiene and nutrition are all influenced by educational attainment. The data in these two tables show that 29 percent of women and 22 percent of men have never attended school. The percentage of men and women who have completed primary school is nearly identical (8 percent for men, 7 percent for women). As educational attainment increases, the percentage of both men and women in these categories decreases: only 2 percent of men and 1 percent of women have completed secondary level education; less than 1 percent of men and women have attended any education beyond the secondary level. 14 | Household Characteristics Table 2.3.1 Educational attainment of household population: female Percent distribution of the de facto female household population age six and over by highest level of education attended or completed, according to background characteristics, Rwanda 2005 Background characteristic No education Primary Primary complete 1 Secondary incomplete Secondary complete2 Superior Total Number Age 6-9 35.7 63.3 0.0 0.0 0.0 0.0 100.0 2,746 10-14 6.3 92.2 0.6 0.5 0.0 0.0 100.0 3,232 15-19 9.3 75.5 8.5 6.1 0.3 0.0 100.0 2,647 20-24 17.7 57.4 14.1 6.6 3.5 0.7 100.0 2,382 25-29 18.0 53.1 16.5 6.9 3.5 1.6 100.0 1,759 30-34 24.9 53.8 8.9 8.9 2.4 0.7 100.0 1,464 35-39 36.2 44.6 7.0 9.1 2.2 0.7 100.0 1,141 40-44 42.7 38.9 11.0 5.6 1.5 0.2 100.0 1,136 45-49 48.9 37.4 8.7 3.8 0.6 0.3 100.0 921 50-54 65.9 26.6 3.5 2.8 0.8 0.4 100.0 762 55-59 70.7 22.3 3.1 2.3 0.3 0.0 100.0 417 60-64 77.6 18.6 1.8 0.8 0.7 0.0 100.0 403 65+ 87.4 9.8 0.7 0.1 0.0 0.0 100.0 914 Residence Urban 19.4 52.3 9.4 11.5 4.5 2.2 100.0 3,103 Rural 30.9 59.1 6.2 2.8 0.6 0.0 100.0 16,823 Province Kigali city 17.1 49.4 10.9 13.3 5.2 3.1 100.0 1,683 South 27.3 59.4 7.9 3.6 1.2 0.1 100.0 5,261 West 31.6 58.6 5.7 2.7 0.7 0.2 100.0 5,132 North 30.5 58.9 5.1 4.2 1.1 0.1 100.0 3,782 East 32.0 58.2 6.3 2.7 0.6 0.0 100.0 4,069 Wealth quintile Lowest 36.3 57.6 4.8 1.0 0.1 0.0 100.0 4,243 Second 30.0 61.4 6.6 1.6 0.1 0.0 100.0 3,878 Middle 31.8 59.9 5.5 2.0 0.3 0.0 100.0 3,932 Fourth 29.6 59.3 7.1 2.9 0.4 0.0 100.0 3,958 Highest 17.2 51.9 9.8 13.3 5.4 1.8 100.0 3,916 Total 29.1 58.0 6.7 4.1 1.2 0.4 100.0 19,927 1 Completed 6 grades at the primary level 2 Completed 6 grades at the secondary level A comparison of these proportions to those of the previous survey shows no significant improvement, although at the time of the previous survey, 35 percent of women and 28 percent of men had no education at all, compared with 29 percent and 22 percent, respectively, in the current survey. The percentage of men and women who have completed primary school has declined, from 12 percent to 8 percent for men, and from 10 percent to 7 percent for women. However, when compared with previous generations, the figures show significant gains: the proportion of females with no education at all has dropped from 87 percent for women 65 and over to 6 percent for girls between the ages of 10 and 14. The percentage for males in these age groups has dropped from 57 percent to 7 percent. In addition, the gap in educational attainment between the sexes seems to be narrowing in the younger age groups. The percentage of women having completed primary school is the same or close to that of men for all ages up to age 34: 9 percent of women between the ages of 15 and 19 said they had completed primary school, compared with 7 percent of men. This narrowing of the gap in educational attainment between the sexes is also seen at the secondary level: between the ages of 20 and 24, 3 percent of men and 4 percent of women have completed secondary school. This contrasts with the common situation of previous generations, when the proportion of women between the ages of 45 and 49 who had completed primary school was 9 percent, while that of men was 19 percent. Household Characteristics | 15 Table 2.3.2 Educational attainment of household population: male Percent distribution of the de facto male household population age six and over by highest level of education attended or completed, according to background characteristics, Rwanda 2005 Background characteristic No education Primary Primary complete 1 Secondary incomplete Secondary complete2 Superior Total Number Age 6-9 37.1 62.3 0.0 0.1 0.0 0.0 100.0 2,835 10-14 7.4 91.0 0.7 0.4 0.0 0.0 100.0 3,053 15-19 8.7 76.6 6.7 7.3 0.2 0.1 100.0 2,489 20-24 15.4 56.2 14.8 9.2 3.4 0.4 100.0 1,967 25-29 15.8 48.8 19.4 9.4 4.0 2.3 100.0 1,376 30-34 19.4 50.9 9.2 13.5 3.8 2.9 100.0 1,036 35-39 24.1 48.7 8.0 13.3 3.0 2.2 100.0 861 40-44 31.8 39.8 16.1 7.6 2.7 1.4 100.0 847 45-49 29.4 43.8 18.9 5.0 1.9 1.1 100.0 742 50-54 33.4 44.3 14.4 4.9 1.5 0.7 100.0 525 55-59 32.1 46.9 13.6 3.1 2.1 1.4 100.0 336 60-64 41.2 39.8 11.0 4.9 1.4 0.5 100.0 224 65+ 56.7 36.3 4.0 1.0 0.4 0.2 100.0 595 Residence Urban 15.4 52.0 10.0 13.2 4.6 3.6 100.0 2,660 Rural 22.6 64.3 7.7 4.0 0.9 0.2 100.0 14,231 Province Kigali city 13.6 47.1 11.7 16.0 5.3 4.7 100.0 1,536 South 20.8 65.0 8.1 4.4 1.0 0.4 100.0 4,436 West 21.6 63.7 7.8 4.3 1.5 0.4 100.0 4,185 North 21.8 64.3 6.5 5.2 1.2 0.4 100.0 3,137 East 25.2 62.2 8.0 3.6 0.6 0.1 100.0 3,596 Wealth quintile Lowest 27.6 64.9 4.9 2.2 0.1 0.0 100.0 3,226 Second 24.2 65.5 7.2 2.8 0.2 0.0 100.0 3,054 Middle 21.7 66.6 7.4 3.2 0.5 0.0 100.0 3,519 Fourth 22.4 62.5 8.9 4.9 0.7 0.1 100.0 3,477 Highest 12.6 53.1 11.2 13.2 5.5 3.3 100.0 3,614 Total 21.5 62.3 8.0 5.4 1.5 0.7 100.0 16,890 1 Completed 6 grades at the primary level 2 Completed 6 grades at the secondary level By residence, the data show significant gaps in educational attainment. In rural areas, 23 percent of men and 31 percent of women have no education at all, compared with 15 percent of men and 19 percent of women in urban areas. There are also variations between provinces. The City of Kigali has the lowest percentage of residents with no education (17 percent of women and 14 percent of men). Conversely, the East region has the highest percentage of men and women with no education (25 percent and 32 percent, respectively). As the level of educational attainment increases, the gaps between the provinces widen: in the City of Kigali, 5 percent have completed secondary school, compared with 2 percent, at most, for men and 1 percent for women in the other provinces. Results by wealth quintile show that the proportion of both men and women with no education decreases as the household standard of living increases. Conversely, the proportion of people having attained education at any given level increases with household wealth. The data also show that in households in the highest wealth quintile there is practically no gap in educational attainment between males and females, up to the secondary level. 16 | Household Characteristics The level of school attendance of school-age children is the primary indicator of a population’s access to education and, indirectly, the socioeconomic development of the area in which the population lives. The 2005 RDHS-III asked questions concerning school attendance of all respondents between the ages of 5 and 24. Table 2.4 shows net attendance ratios (NAR) and gross attendance ratios (GAR) by sex, residence, and province, by level of educational attainment. Table 2.4 School attendance ratios Net attendance ratios (NAR) and gross attendance ratios (GAR) for the de jure household population by level of schooling and sex, according to background characteristics, Rwanda 2005 Net attendance ratio1 Gross attendance ratio2 Background characteristic Male Female Total Male Female Total Gender parity index3 PRIMARY SCHOOL Residence Urban 79.4 81.6 80.5 134.3 133.6 133.9 0.99 Rural 72.9 75.7 74.3 132.7 138.1 135.4 1.04 Province Kigali city 80.5 82.0 81.2 134.7 133.3 134.0 0.99 South 73.1 75.6 74.4 129.3 132.9 131.0 1.03 West 74.2 76.2 75.2 138.8 142.3 140.6 1.03 North 75.9 78.9 77.4 125.6 127.6 126.6 1.02 East 69.8 74.1 71.9 136.8 148.6 142.5 1.09 Total 73.8 76.6 75.2 132.9 137.5 135.2 1.03 SECONDARY SCHOOL Residence Urban 11.1 12.3 11.8 20.6 20.8 20.7 1.01 Rural 3.3 2.2 2.7 5.6 3.5 4.5 0.62 Province Kigali city 12.8 14.4 13.7 23.7 26.1 25.0 1.10 South 2.9 3.5 3.2 4.8 5.2 5.0 1.08 West 3.4 2.4 2.9 7.3 4.1 5.6 0.56 North 5.3 3.6 4.4 8.4 4.2 6.2 0.50 East 3.7 2.0 2.8 5.5 4.0 4.8 0.72 Total 4.4 3.8 4.1 7.7 6.2 6.9 0.81 1 The NAR for primary school is the percentage of the primary-school-age (6-11 years) population that is attending primary school. The NAR for secondary school is the percentage of the secondary-school-age (12-18 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 over-age and underage students at a given level of schooling, the GAR can exceed 100 percent. 3 The Gender Parity Index for primary school is the ratio of the primary school GAR for females to the GAR for males. The Gender Parity Index for secondary school is the ratio of the secondary school GAR for females to the GAR for males. Net school attendance ratios (NAR) measure school attendance in children who have reached the official school age. For primary school, the NAR is the percentage of the primary-school-age population (age 7-12 in Rwanda) that is actually attending primary school. This table shows that the primary level NAR is 75 percent for Rwanda, which means that three-quarters of the population between the ages of 7 and 12 are attending primary school. The ratio is higher for urban areas than for rural areas (81 percent compared with 74 percent). In the provinces, the ratio ranges from a high of 81 percent in the City of Household Characteristics | 17 Kigali to a low of 72 percent in the East province. The NAR is also higher for women (77 percent) than for men (74 percent), regardless of residence and province. At the secondary level, which concerns the population between the ages of 13 and 19, the NAR is much lower (4 percent), which means that only 4 percent of the official secondary-school-age population are actually attending school. There is practically no gap between the sexes. However, it is much higher in urban areas than in rural areas (12 percent compared with 3 percent), which may explain the major gap between the City of Kigali, with an NAR of 14 percent, and the other provinces, whose NARs are between 3 percent and 4 percent. Table 2.4 also shows gross school attendance ratios (GAR). Unlike the NAR, the GAR measures school attendance in young people regardless of age. The GAR for primary school is the total number of students of any age attending primary school, expressed as a percentage of the official primary-school-age population, which is 7 to 12 years in Rwanda. Unless there are significant numbers of over-age and under-age students at a given level of schooling, the GAR is always higher than the NAR and can, in some cases, exceed 100 percent. In Rwanda, the GAR is 135 percent, which means that a significant proportion of children who do not fall into the official primary-school-age category are attending school at the primary level. These are likely to be children over the age of 12 who are still attending primary school; in fact, a program exists to reintegrate children who dropped out of primary school for any reason. In addition, the GAR is higher for girls than for boys (138 percent for girls compared with 133 percent for boys). Moreover, there is practically no difference by residence. At the secondary level, the NAR is very low. Only 4 percent of all children of official secondary school age are actually attending school. The ratio is nearly the same for girls and boys. However, it is higher in urban areas than in rural areas (12 percent compared with 3 percent). The GAR is also very low (7 percent), either because official secondary-school-age children are still in primary school, or because they have dropped out of secondary school or have never attended at all. Students who do not pass the national exam at the end of primary school are not allowed to attend state or state-subsidized secondary schools, which are less expensive than private schools. In addition, the GAR for boys is very similar to that of girls, but there is a pronounced difference by residence (21 percent for urban areas compared with 5 percent for rural areas). At 25 percent, the GAR for the City of Kigali stands out from the other provinces, where the GAR is very low (a maximum of 6 percent in the North province). The table includes a third school attendance indicator: the gender parity index (GPI), which is the ratio of the GAR for females to the GAR for males. The narrower the gap between the sexes, the closer the index is to 1. Table 2.4 shows a GPI for primary school of just above 1, which indicates an absence of disparity between the sexes. Curiously, only urban areas, in particular the City of Kigali, show a GPI of slightly below 1. The GPI for secondary school is below one (0.81); this indicates that girls are educationally disadvantaged at this level. The inequality is more pronounced in rural areas, which have a GPI of only 0.62, and in the West (0.56) and North (0.50) provinces. The City of Kigali has the highest GPI (1.10). Figure 2.2 shows that the rate of school attendance, which is low at ages 5 and 6, begins to increase at age 7, the official age for entering primary school. It reaches high levels between the ages of 8 and 13. This period corresponds to the primary school years for children in the normal primary cycle. After age 13, the curve declines steadily, reaching its lowest point at the age of 20. After the age of 13, school attendance rates approaching or exceeding 50 percent do not indicate high school attendance at the secondary level but, rather, that a majority of children are beginning primary school late. 18 | Household Characteristics It should also be noted that the proportion of women is higher between the ages of 6 and 12 everywhere, while the situation reverses itself after this up to age 23, although, paradoxically, the balance is restored at age 24. 2.4 LIVING CONDITIONS The household survey gathered information on certain housing characteristics (access to electricity, drinking water source, type of toilet facilities, roofing and flooring materials). Information was also sought concerning ownership of various modern durable goods (radio, television, refrigerator, bicycle, motorcycle/scooter, car/truck). These characteristics are used to evaluate the socioeconomic conditions of the household. Table 2.5 shows that, at the national level, very few households have access to electricity (5 percent). The situation has not changed much compared with 2000, when the proportion was 6 percent. The results show large disparities between urban and rural areas. In rural areas, only around 1 percent of households have electricity, compared with 25 percent in urban areas. With respect to drinking water, at the national level, almost 33 percent of households use spring water and one-quarter of households use a public tap; 14 percent of households use uncovered public wells as a source of drinking water and 22 percent consume water from a public tap. Overall, 19 percent of households use water that is considered unhealthy, leaving the population open to increased risk of contracting diseases related to unclean drinking water. The proportion of households with access to running water in their dwelling or courtyard remains low, approximately 3 percent. In rural areas, more than half of the households use unsafe drinking water because 55 percent draw their water from springs (35 percent), rivers/streams (12 percent), or ponds/lakes (8 percent). Figure 2.2 Age-specific Attendance Rates (Percentage of the population age 5-24 years attending school, by age and sex) 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Age 0 10 20 30 40 50 60 70 80 90 100 Percent Male Female RDHS 2005 Household Characteristics | 19 Table 2.5 Housing characteristics Percent distribution of households by housing characteristics, according to residence, Rwanda 2005 Residence Housing characteristic Urban Rural Total Electricity Yes 25.1 1.3 4.8 No 74.7 98.6 95.1 Total 100.0 100.0 100.0 Source of drinking water Piped into dwelling/compound/plot 14.1 0.5 2.5 Public tap 41.3 22.1 24.9 Open well in compound/plot 0.5 0.1 0.1 Open public well 12.0 13.8 13.5 Covered well in compound/plot 0.1 0.0 0.0 Covered public well 5.2 6.3 6.2 Spring 18.5 35.2 32.7 River, stream 4.9 12.2 11.2 Pond, lake 1.5 8.0 7.1 Dam 0.4 0.9 0.9 Rainwater 0.1 0.3 0.3 Tanker truck 0.1 0.0 0.0 Bottled water 0.1 0.0 0.0 Other 1.2 0.5 0.6 Total 100.0 100.0 100.0 Time to water source Percentage < 15 minutes 47.9 27.0 30.1 Median time to source (in minutes) 14.3 28.0 24.4 Sanitation facility Flush toilet 5.4 0.2 1.0 Traditional pit toilet 44.0 70.8 66.9 Ventilated improved pit (VIP) latrine 47.1 24.1 27.5 No facility, bush, field 3.4 4.8 4.6 Other 0.1 0.1 0.1 Total 100.0 100.0 100.0 Flooring material Earth, mud, sand 51.8 92.1 86.2 Dung 0.5 0.9 0.8 Parquet, polished wood 0.0 0.0 0.0 Vinyl, asphalt strips 0.0 0.0 0.0 Ceramic tiles 1.0 0.0 0.2 Cement 46.5 6.9 12.7 Carpet 0.2 0.0 0.0 Other 0.1 0.1 0.1 Total 100.0 100.0 100.0 Number of households 1,510 8,762 10,272 In urban areas, public taps constitute the main water source and are used by 41 percent of the households surveyed. 19 percent of urban households use spring water and 14 percent have running water in their dwellings or courtyards. Finally, 12 percent draw water from uncovered public wells. The situation has not improved since 2000. The proportion of households that have running water in their dwelling units has decreased by 3 percent. The number of households using water from a public tap has dropped by 4 percent. Table 2.5 shows that 30 percent of households are within 15 minutes of their water source. This proportion is lower in rural areas (27 percent) than in urban areas (48 percent). The median time to drinking water source is 24 minutes for the country as a whole, 28 minutes for rural areas and 14 minutes for urban areas. 20 | Household Characteristics Compared with 2000, the proportion of households less than 15 minutes from their water source has increased by 5 percent (from 25 percent to 30 percent). However, the change is insignificant in terms of the median time to drinking water source, which was 26 minutes in 2000 and is 24 minutes now. With respect to type of toilet facilities, Table 2.5 shows a high proportion of households with access only to open pits or uncovered latrines (67 percent); 28 percent of households use covered latrines. In the country as a whole, rural areas have more rudimentary latrines (71 percent) than ventilated improved pit (VIP) latrines (24 percent), while in urban areas the proportion of VIP latrines (47 percent) and rudimentary latrines (44 percent) are similar to one another. Very few households have flush toilets: 1 percent in the country as a whole, 5 percent in urban areas, and an insignificant percentage in rural areas. It should also be noted that 5 percent of households have no toilet facilities at all. Compared with the previous survey, the proportion of households using VIP latrines has increased significantly, from 7 percent to approximately 28 percent. The proportion of households with no facilities at all has remained the same. The type of material used for flooring is extremely important because some materials are a propagation factor for certain disease-causing germs and parasites. The great majority of Rwandan households use earth/sand/dung flooring (86 percent). The proportion is higher in rural areas (92 percent) than in urban areas (52 percent). It should also be noted that 13 percent of households have cement floors. However, this type of flooring occurs much more frequently in urban than in rural areas (47 percent compared with 7 percent). To evaluate households’ socioeconomic level, the survey gathered information about ownership of certain durable goods considered indicative of higher socioeconomic living standards. Table 2.6, shows that half of Rwandan households own none of the goods listed. The proportion is higher in rural areas than in urban areas (56 percent for rural, 32 percent for urban). However, it has declined in relation to 2000, when 63 percent of households owned none of the goods listed. Overall, the most frequently owned durable good is a radio (46 percent), which is found more often in urban households than in rural areas (65 percent compared with 43 percent). The proportion of house- holds owning radios has increased overall in relation to 2000, when only 35 percent of households owned a radio. Bicycles are used as a means of transportation in 11 percent of households in both rural and urban areas. In addition, in urban areas, 24 percent of households own a mobile telephone, and 14 percent own a television; in rural areas, these goods are more or less nonexistent. Table 2.7 shows the percent distribution of households by wealth quintile. The wealth index was developed on the basis of household goods data, using principal components analysis. The information on household goods comes from responses to questions about ownership of certain durable goods (television, radio, car, etc.) and questions about certain housing characteristics such as access to electricity, source of drinking water, type of toilet facilities, type of flooring material, number of rooms used for sleeping, type of cooking fuel, etc. The index was developed as follows: • Each durable good or housing characteristic is assigned a weight (score or coefficient) generated by principal components analysis. Table 2.6 Household durable goods Percentage of households possessing various durable consumer goods, by residence, Rwanda 2005 Residence Durable consumer good Urban Rural Total Radio 65.2 42.5 45.8 Television 14.0 0.3 2.3 Mobile telephone 24.1 1.3 4.6 Non-mobile telephone 4.8 0.1 0.8 Refrigerator 7.7 0.1 1.2 Bicycle 10.5 11.1 11.0 Motorcycle/scooter 1.8 0.3 0.5 Car, truck 4.2 0.1 0.7 None of the above 31.7 55.5 52.0 Number of households 1,510 8,762 10,272 Household Characteristics | 21 • The resulting scores for durable goods are standardized according to a normal distribution assuming a mean of 0 and a standard deviation of 1 (Gwatkin et al., 2000). • Each household is assigned a score for each durable good and these scores are added together to obtain a total for each household. • The households are classified in increasing order of total score and divided into 5 equal categories, quintiles. This yields a scale from 1 (poorest quintile) to 5 (richest quintile). • The score for each household is assigned to the individuals in that household. The individuals are thus distributed among the categories. The results show that in urban areas, 60 percent of households fall into the richest quintile, while in rural areas only 12 percent fall into this quintile. The proportion of rich households is highest in the City of Kigali (69 percent). Conversely, in urban areas, only 6 percent of households fall into the poorest quintile. In fact, the preceding tables showing ownership of durable goods, housing characteristics, and source of drinking water have already established that the population of Rwanda is generally poor. Table 2.7 only confirms the previous results and explains the relative lack of variation between provinces. Table 2.7 Wealth quintiles Percent distribution of households by wealth quintile, according to residence and province, Rwanda 2005 Wealth quintile Residence/ province Lowest Second Middle Fourth Highest Total Number Residence Urban 6.1 7.7 11.6 14.8 59.7 100.0 1,510 Rural 24.2 20.4 22.2 21.5 11.7 100.0 8,762 Province Kigali city 6.4 4.4 10.8 9.8 68.5 100.0 864 South 21.4 20.8 19.0 22.4 16.5 100.0 2,722 West 23.5 15.3 21.6 24.5 15.1 100.0 2,522 North 27.3 19.5 22.6 18.6 12.0 100.0 1,946 East 20.5 24.2 23.6 19.5 12.2 100.0 2,218 Total 21.6 18.6 20.6 20.5 18.7 100.0 10,272 2.5 BIRTH REGISTRATION WITH CIVIL AUTHORITIES Registering a child with civil authorities establishes the child’s legal family ties and his or her right to a name and nationality prior to the age of majority. It confers on the child the right to be recognized by his or her parents and the right to state protection if his or her rights are abused by the parents. It gives the child access to social assistance through the parents, including health insurance, and establishes family lineage. It is therefore an essential formality. 22 | Household Characteristics Registration of a child with civil authorities, if performed correctly, also provides a reliable source of sociodemographic statistics. For this reason, the survey asked whether children had been registered with the civil authorities. Table 2.8 shows that a majority of children have been registered with the civil authorities (82 percent); only 18 percent of children (less than one in five) have not been registered. Of those children declared with the civil authorities, 78 percent possess birth certificates. Children’s age and sex have little to do with whether or not they are registered with the civil authorities. Also, the level of wealth does not seem to influence the prevalence of birth registration. Children in the second and middle wealth quintiles showed the highest levels of registration (84 percent in both of these quintiles). There is some discrepancy by residence with, curiously, rural areas showing a higher percentage of birth registrations (83 percent compared with 79 percent in urban areas). Results by province show that households in the North and South provinces are the most likely to have declared their children with the civil authorities (85 percent and 89 percent, respectively). Table 2.8 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 characteristics, Rwanda 2005 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 79.6 2.8 82.4 3,411 2-4 76.6 5.7 82.3 4,711 Sex Male 77.3 4.5 81.8 4,103 Female 78.5 4.5 82.9 4,019 Residence Urban 74.0 4.6 78.6 1,170 Rural 78.5 4.5 83.0 6,952 Province Kigali city 74.1 5.2 79.2 596 South 81.7 3.2 84.9 2,013 West 73.8 5.0 78.8 2,166 North 83.8 5.2 89.0 1,622 East 74.4 4.5 78.8 1,725 Wealth quintile Lowest 76.9 4.9 81.8 1,687 Second 80.1 4.0 84.1 1,640 Middle 79.9 4.4 84.2 1,697 Fourth 76.0 4.7 80.7 1,623 Highest 76.3 4.5 80.9 1,475 Total 77.9 4.5 82.4 8,123 Characteristics of Household Respondents | 23 CHARACTERISTICS OF SURVEY RESPONDENTS 3 The purpose of this chapter is to provide a sociodemographic profile of the women age 15-49 and men age 15-59 who responded to this survey. This information is important for understanding the behavior of the population with respect to contraception, STIs, HIV/AIDS, and fertility preferences. Like the household questionnaire, the individual questionnaires gathered information concerning respondents’ age, place of residence, marital status, and educational attainment. This chapter will also analyze results with respect to literacy, exposure to mass media, and employment of the men and women surveyed. These characteristics will be used to interpret findings in the rest of the report. 3.1 BACKGROUND CHARACTERISTICS OF RESPONDENTS Given the importance of age in analyzing demographic phenomena, special attention was paid to making sure this statistic was accurately recorded in the survey. Prior to taking down any information, the interviewer asked respondents to gather all official documents providing information about themselves and other members of the household. If no official documents were available, the interviewer confirmed the age information provided by the respondent through reference to major life events (age at the time of marriage, age of first child, etc.) or well-known national or regional events. Table 3.1 shows no major disparities in the distribution of women age 15-49 and men age 15-59 grouped by five-year age increments. Proportions decline with increasing age. For women, the percentages range from 23 percent for the 15-19 age group, to 8 percent for the 45-49 age group. For men, the percentages range from 23 percent for ages 15-19, to 3 percent for ages 55-59. Table 3.1 Age of respondents Percent distribution of women and men by age, Rwanda 2005 Women Men Age Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number 15-19 22.8 2,585 2,595 22.9 1,102 1,079 20-24 20.8 2,354 2,356 19.6 946 951 25-29 15.4 1,738 1,745 13.1 632 647 30-34 12.9 1,466 1,460 10.6 509 515 35-39 10.0 1,134 1,133 9.2 442 435 40-44 10.0 1,135 1,127 8.4 404 408 45-49 8.0 910 905 7.8 378 378 50-54 na na na 5.4 260 261 55-59 na na na 3.1 147 146 Total 15-49 100.0 11,321 11,321 91.5 4,413 4,413 Total 15-59 na na na 100.0 4,820 4,820 na = Not applicable 24 | Characteristics of Household Respondents All men and women in the sample were asked their marital status. For the RDHS-III, all men and women were considered “married” if they were in union with a partner, whether the union was formal (legally married) or informal (“living together”). According to this definition, Table 3.2 shows that more than 4 in 10 women (44 percent) had never been married at the time of the survey, and more than half of the women (54 percent) were married. Nearly five in ten men were single (46 percent) and more than half of the men (52 percent) were married. In addition, 2 percent of the women were divorced, separated or widowed at the time of the survey, compared with 3 percent of the men. Table 3.2 Background characteristics of respondents Percent distribution of women and men by selected background characteristics, Rwanda 2005 Women Men Background characteristic Weighted percent Weighted number Unweighted number Weighted percent Weighted number Unweighted number Marital status Never married 44.0 4,983 4,961 45.6 2,196 2,214 Married 54.1 6,126 6,138 51.9 2,500 2,478 Divorced/separated 1.4 158 167 1.9 89 92 Widowed 0.5 54 55 0.7 35 36 Residence Urban 17.0 1,921 2,616 17.4 840 1,130 Rural 83.0 9,400 8,705 82.6 3,980 3,690 Old province Kigali 8.0 900 1,085 8.8 426 511 Kigali Ngali 9.9 1,118 945 9.3 449 387 Gitarama 10.8 1,219 930 10.8 522 400 Butare 9.6 1,090 945 9.4 452 381 Gikongoro 5.7 650 885 5.7 275 371 Cyangugu 7.5 852 1,010 8.0 386 461 Kibuye 5.7 649 921 5.1 244 344 Gisenyi 10.4 1,179 938 10.1 488 385 Ruhengeri 10.4 1,180 940 9.9 478 376 Byumba 7.7 873 893 8.2 395 398 Umutara 4.9 554 897 5.6 271 425 Kibungo 9.3 1,057 932 9.0 433 381 Province Kigali city 10.0 1,127 1,329 10.8 523 619 South 26.1 2,958 2,760 25.9 1,250 1,152 West 24.9 2,824 2,971 24.6 1,185 1,237 North 18.2 2,063 1,821 17.5 845 746 East 20.7 2,348 2,440 21.1 1,017 1,066 Education No education 23.4 2,646 2,603 17.4 839 819 Primary 67.1 7,591 7,497 70.3 3,389 3,357 Secondary 9.0 1,018 1,134 10.9 526 566 More than secondary 0.6 66 87 1.4 66 78 Wealth quintile Lowest 21.4 2,421 2,327 18.0 867 826 Second 20.5 2,325 2,195 18.3 884 819 Middle 18.5 2,099 1,988 20.3 978 927 Fourth 18.8 2,133 2,151 20.8 1,004 1,012 Highest 20.7 2,342 2,660 22.6 1,087 1,236 Religion Catholic 45.3 5,126 4,975 51.5 2,482 2,416 Protestant 37.5 4,247 4,382 31.9 1,539 1,586 Adventist 13.2 1,498 1,490 12.0 578 585 Muslim 1.8 207 234 2.0 96 112 Other 0.9 97 97 - 0 0 No religion 1.3 146 143 2.6 126 121 Total 100.0 11,321 11,321 100.0 4,820 4,820 Characteristics of Household Respondents | 25 The distribution of respondents by residence shows that the majority of the Rwandan population is living in rural areas (83 percent of women and men). Similarly, the data by province shows a relatively uniform distribution, with no significant disparities between men and women. The tabulation of respondents by religion indicates a majority of Catholic adherents (45 percent of women and 52 percent of men), with Protestant religions coming in second (38 percent of women and 32 percent of men). The Adventist faith is the next most common religion (13 percent of women and 12 percent of men), followed by the Muslim faith (2 percent of women and 2 percent of men). Table 3.2 also shows the distribution of men and women according to household wealth quintile. The development of this index is explained in Chapter 2. Table 3.2 provides educational attainment data for the respondents. The proportion of women with no education is significantly higher than that of men (23 percent for women, 17 percent for men). However, the gap between males and females is not very wide at the primary and secondary levels. 3.2 EDUCATIONAL ATTAINMENT Tables 3.3.1 and 3.3.2 show the distribution of respondents by highest level of education attained. The proportions of educated men are only slightly higher than those of women: 70 percent have completed primary school, compared with 67 percent of women. At the secondary level, the proportions are 11 percent for men and 9 percent for women. It should be noted that proportions for both men and women drop significantly from the primary to secondary and secondary to post-secondary levels. Table 3.3.1 Educational attainment by background characteristics: women Percent distribution of women by highest level of schooling attended or completed, according to background characteristics, Rwanda 2005 Highest level of schooling attended or completed Background characteristic No education Primary Secondary More than secondary Total Number of women Age 15-19 8.6 84.4 7.0 0.0 100.0 2,585 20-24 17.5 71.6 10.2 0.8 100.0 2,354 25-29 17.8 69.9 10.9 1.5 100.0 1,738 30-34 24.6 63.5 11.3 0.6 100.0 1,466 35-39 36.4 51.7 11.1 0.8 100.0 1,134 40-44 41.8 50.7 7.2 0.2 100.0 1,135 45-49 50.0 45.8 4.0 0.2 100.0 910 Residence Urban 13.5 58.9 24.3 3.4 100.0 1,921 Rural 25.4 68.7 5.9 0.0 100.0 9,400 Province Kigali city 11.3 58.6 26.0 4.1 100.0 1,127 South 20.3 71.4 8.0 0.3 100.0 2,958 West 28.1 65.7 5.9 0.3 100.0 2,824 North 25.4 65.2 9.3 0.1 100.0 2,063 East 25.5 68.9 5.5 0.1 100.0 2,348 Wealth quintile Lowest 32.3 65.7 2.0 0.0 100.0 2,421 Second 25.9 71.3 2.7 0.0 100.0 2,325 Middle 25.6 70.2 4.2 0.0 100.0 2,099 Fourth 22.5 71.2 6.3 0.0 100.0 2,133 Highest 10.4 57.6 29.2 2.8 100.0 2,342 Total 23.4 67.1 9.0 0.6 100.0 11,321 26 | Characteristics of Household Respondents Table 3.3.2 Educational attainment by background characteristics: men Percent distribution of men by highest level of schooling attended or completed, according to background characteristics, Rwanda 2005 Highest level of schooling attended or completed Background characteristic No education Primary Secondary More than secondary Total Number of men Age 15-19 6.8 86.2 7.0 0.0 100.0 1,102 20-24 12.8 74.5 12.2 0.6 100.0 946 25-29 15.9 68.2 13.4 2.5 100.0 632 30-34 20.3 60.6 16.3 2.9 100.0 509 35-39 22.9 57.9 15.7 3.5 100.0 442 40-44 29.7 56.7 11.6 1.9 100.0 404 45-49 25.9 67.1 5.9 1.1 100.0 378 50-54 29.6 62.9 6.6 0.9 100.0 260 55-59 29.5 62.9 6.6 1.0 100.0 147 Residence Urban 9.5 59.5 24.9 6.2 100.0 840 Rural 19.1 72.6 8.0 0.4 100.0 3,980 Province Kigali city 9.9 56.3 26.4 7.4 100.0 523 South 16.4 73.1 9.8 0.6 100.0 1,250 West 17.8 71.6 9.8 0.7 100.0 1,185 North 20.1 70.2 8.7 1.0 100.0 845 East 19.8 72.6 7.3 0.3 100.0 1,017 Wealth quintile Lowest 25.5 71.6 2.9 0.0 100.0 867 Second 22.3 72.3 5.4 0.0 100.0 884 Middle 18.1 76.0 5.9 0.0 100.0 978 Fourth 16.0 73.8 10.1 0.1 100.0 1,004 Highest 7.7 59.3 27.0 6.0 100.0 1,087 Total 17.4 70.3 10.9 1.4 100.0 4,820 The data by age show that the proportion of men and women with no education has decreased significantly from previous generations. For men, this proportion has dropped from 30 percent in the 55-59 age group to 7 percent in the 15-19 age group. For women, the proportions for these age groups are 50 percent and 9 percent, respectively. The gap between men and women in the previous generations has narrowed significantly: among men age 45 to 49 years, 26 percent have no education, compared with 50 percent for women in the same age group. For those age 15-19 years, the proportions are 7 percent for men and 9 percent for women. Similarly, in the 15-19 age group, the proportion of girls who have completed primary school is not significantly different from that of boys (84 percent for girls, 86 percent for boys), although the percentage of boys is still slightly higher. In addition, 11 percent of young men have completed secondary school, compared with 9 percent of young women. The gaps are due to early marriage and pregnancy, which often prevent girls from pursuing a regular course of education. The educational attainment of respondents varies by residence. The proportion of men and women with no education is higher in rural areas (19 percent for men, 25 percent for women) than in urban areas (10 percent for men, 14 percent for women). Urban areas also have the highest proportion of men and women at every level of education except primary. Results by province show a wide gap between the City of Kigali and the rest of the country. In the City of Kigali, 11 percent of women and 10 percent of men have no education; in the other provinces the Characteristics of Household Respondents | 27 proportions are nearly twice as high. The West province has the highest percentage of women with no education (28 percent); the North and East provinces have the highest proportion of uneducated men (20 percent each). The data in this table show a positive relationship between educational attainment and household wealth: the proportion of men and women with no education decrease as household wealth increases. 3.3 LITERACY For this survey, literacy was established by asking respondents who reported not having attended school and or having attended only primary school to read a sentence that was presented to them. Respondents were then classified into one of the following three levels: cannot read at all; can read part of a sentence; can read a whole sentence. The test was given only to men and women who had less than a secondary education; those with secondary or postsecondary educations (10 percent of women and 12 percent of men) were considered literate. Tables 3.4.1 and 3.4.2 show that a higher proportion of women than men cannot read (29 percent of women; 22 percent of men). Conversely, 78 percent of men and 70 percent of women are considered literate; that is, they have attended secondary school or, if they have attended only primary school, they are able to read all or part of a sentence. Table 3.4.1 Literacy: women Percent distribution of women by level of schooling attended and by level of literacy, and percent literate, according to background characteristics, Rwanda 2005 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 Total1 Number of women Percent literate2 Age 15-19 7.0 60.3 12.2 20.2 100.0 2,585 79.6 20-24 11.0 51.6 12.1 25.1 100.0 2,354 74.6 25-29 12.4 55.0 11.1 21.4 100.0 1,738 78.4 30-34 11.9 50.7 9.8 27.1 100.0 1,466 72.4 35-39 11.9 41.7 10.0 36.0 100.0 1,134 63.6 40-44 7.4 33.8 11.7 46.5 100.0 1,135 52.9 45-49 4.2 30.9 9.2 55.0 100.0 910 44.3 Residence Urban 27.7 48.3 7.8 15.6 100.0 1,921 83.8 Rural 5.9 49.8 11.9 32.1 100.0 9,400 67.6 Province Kigali city 30.1 47.5 7.4 14.7 100.0 1,127 85.0 South 8.3 55.9 8.6 26.8 100.0 2,958 72.8 West 6.2 47.0 12.6 33.8 100.0 2,824 65.8 North 9.4 45.7 14.4 30.3 100.0 2,063 69.5 East 5.5 49.0 11.7 33.3 100.0 2,348 66.2 Wealth quintile Lowest 2.0 44.6 13.3 39.9 100.0 2,421 59.8 Second 2.7 51.2 12.8 32.9 100.0 2,325 66.8 Middle 4.2 51.2 12.2 32.0 100.0 2,099 67.7 Fourth 6.3 54.1 10.6 28.7 100.0 2,133 71.0 Highest 32.0 47.5 7.0 13.1 100.0 2,342 86.5 Total 9.6 49.6 11.2 29.3 100.0 11,321 70.3 1 Includes those with missing information 2 Refers to women who attended secondary school or higher and women who can read a whole sentence or part of a sentence. 28 | Characteristics of Household Respondents Table 3.4.2 Literacy: men Percent distribution of men by level of schooling attended and by level of literacy, and percent literate, according to background characteristics, Rwanda 2005 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 Total1 Number of men Percent literate2 Age 15-19 7.0 60.6 13.0 19.0 100.0 1,102 80.7 20-24 12.7 54.7 11.3 20.9 100.0 946 78.8 25-29 15.9 57.4 9.5 17.2 100.0 632 82.8 30-34 19.2 50.1 9.4 21.0 100.0 509 78.7 35-39 19.2 51.1 9.3 20.4 100.0 442 79.6 40-44 13.5 49.9 7.8 28.4 100.0 404 71.2 45-49 7.0 51.3 9.9 30.7 100.0 378 68.1 50-54 7.5 51.8 9.8 29.7 100.0 260 69.1 55-59 7.6 49.2 13.8 28.0 100.0 147 70.6 Residence Urban 31.0 47.3 7.6 13.3 100.0 840 86.0 Rural 8.3 56.1 11.3 23.9 100.0 3,980 75.8 Province Kigali city 33.8 44.4 7.8 13.6 100.0 523 86.0 South 10.5 53.7 10.4 24.9 100.0 1,250 74.6 West 10.6 56.3 10.4 22.2 100.0 1,185 77.3 North 9.7 54.1 12.2 23.7 100.0 845 76.0 East 7.5 59.3 11.5 21.4 100.0 1,017 78.4 Wealth quintile Lowest 2.9 52.9 12.7 31.1 100.0 867 68.5 Second 5.4 53.6 12.5 27.9 100.0 884 71.5 Middle 5.9 57.7 12.4 23.6 100.0 978 76.1 Fourth 10.2 59.1 10.5 19.9 100.0 1,004 79.8 Highest 33.0 49.7 6.2 10.6 100.0 1,087 88.9 Total 12.3 54.6 10.7 22.0 100.0 4,820 77.5 1 Includes those with missing information 2 Refers to men who attended secondary school or higher and women who can read a whole sentence or part of a sentence. The level of illiteracy varies appreciably by residence. Illiteracy is higher in rural areas than in urban areas. The results by province show a gap between the City of Kigali and the rest of the country: in Kigali, 86 percent of men and 85 percent of women are literate, compared with a maximum of 78 percent of men in the East province and 73 percent of women in the South province. In addition, results by wealth quintile show that the level of illiteracy decreases considerably from the poorest to the richest quintile, dropping for women from 40 percent in the lowest quintile to 13 percent in the highest quintile and for men from 31 percent in the lowest quintile to 11 percent in the highest quintile. 3.4 EXPOSURE TO MASS MEDIA Data on the exposure of men and women to mass media are especially important to the development of education programs and the dissemination of all types information, particularly information about health and family planning. Tables 3.5.1 and 3.5.2 present data on the exposure of men and women to mass media (print or broadcast). It should be stated at the outset that it is not necessary for a household to own a radio or television or to buy a newspaper in order to have access to these media because many people listen to the radio or watch television at the homes of friends and neighbors. Characteristics of Household Respondents | 29 Table 3.5.1 Exposure to mass media: women Percentage of women who usually read a newspaper at least once a week, watch television at least once a week, and listen to the radio at least once a week, by background characteristics, Rwanda 2005 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 No media Number of women Age 15-19 11.2 6.9 56.7 1.5 39.7 2,585 20-24 8.6 6.7 57.0 1.9 40.4 2,354 25-29 8.8 5.2 56.0 1.8 42.3 1,738 30-34 7.5 4.2 53.7 1.2 44.4 1,466 35-39 6.8 4.2 50.0 1.3 47.5 1,134 40-44 6.7 2.6 48.6 0.7 49.9 1,135 45-49 3.6 2.6 48.2 0.8 50.7 910 Residence Urban 15.4 22.9 73.5 6.7 23.5 1,921 Rural 6.9 1.6 50.2 0.4 47.7 9,400 Province Kigali city 15.4 30.2 76.6 7.5 19.9 1,127 South 8.7 2.8 57.5 1.2 40.9 2,958 West 10.5 2.9 42.4 1.0 53.8 2,824 North 7.6 2.7 50.3 0.6 47.6 2,063 East 2.4 1.2 56.6 0.1 42.4 2,348 Education No education 0.8 1.1 37.7 0.1 61.5 2,646 Primary 8.1 3.3 55.6 0.5 41.7 7,591 Secondary or higher 28.2 28.2 83.5 11.3 12.7 1,084 Wealth quintile Lowest 4.1 0.4 22.6 0.0 75.1 2,421 Second 5.2 1.0 54.3 0.0 43.5 2,325 Middle 6.3 1.0 52.4 0.0 45.7 2,099 Fourth 7.1 1.6 63.4 0.2 34.8 2,133 Highest 18.7 21.3 79.6 6.7 17.0 2,342 Total 8.3 5.2 54.1 1.4 43.6 11,321 Table 3.5.1 shows that, at the national level, more than two in five women (44 percent) and approximately one in five men (19 percent) are not exposed to any media. However, there has been a significant improvement since the 2000 RDHS-II, which reported that 59 percent of women and 35 percent of men were not exposed to any media. Radio is the most common form of media exposure: more than half of the women (54 percent) and four out of five of the men (80 percent) reported listening to the radio at least once a week. One in twenty women (5 percent) and one in ten men (11 percent) watch television at least once a week. Men also reported reading a newspaper a little more frequently than women: only 8 percent of women, compared with 10 percent of men, reported reading a newspaper at least once a week. The proportions of men and women who are exposed to all three media are very low: only 1 percent of women and 4 percent of men. The data by age show that the younger generations are relatively more exposed to mass media than older people. In fact, the proportions of women who are not exposed to any media vary from 40 percent for women age 15-19 to 51 percent for women age 45-49. For men, the age differences are narrow and uneven. 30 | Characteristics of Household Respondents Table 3.5.2 Exposure to mass media: men Percentage of men who usually read a newspaper at least once a week, watch television at least once a week, and listen to the radio at least once a week, by background characteristics, Rwanda 2005 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 No media Number of men Age 15-19 6.5 11.1 77.5 2.3 21.3 1,102 20-24 10.7 14.3 84.0 5.4 15.0 946 25-29 14.4 13.2 81.4 6.5 16.7 632 30-34 12.5 14.1 79.2 7.8 19.8 509 35-39 12.6 8.7 80.1 5.1 18.8 442 40-44 9.1 6.2 77.3 2.7 21.8 404 45-49 9.3 6.3 76.4 2.5 22.3 378 50-54 7.9 6.4 77.4 3.0 21.0 260 55-59 6.9 3.7 78.4 2.2 21.6 147 Residence Urban 27.3 37.5 89.8 19.2 9.3 840 Rural 6.5 5.2 77.5 1.3 21.3 3,980 Province Kigali city 33.4 47.6 89.2 25.9 9.5 523 South 6.1 7.4 76.3 2.0 23.1 1,250 West 8.6 6.1 73.9 1.6 24.7 1,185 North 6.7 5.6 82.9 1.9 16.6 845 East 7.5 5.9 82.7 1.6 15.1 1,017 Education No education 0.5 3.0 67.1 0.1 32.4 839 Primary 7.4 7.9 80.2 2.1 18.5 3,389 Secondary or higher 38.7 38.7 94.4 23.8 4.6 592 Wealth quintile Lowest 3.1 2.4 62.3 0.3 36.7 867 Second 5.0 2.6 75.3 0.3 23.0 884 Middle 4.3 3.7 80.1 0.6 18.6 978 Fourth 7.4 6.0 85.9 1.1 13.5 1,004 Highest 27.4 35.1 90.8 17.4 7.8 1,087 Total 10.1 10.8 79.6 4.4 19.2 4,820 Results by residence reveal significant differentials. In rural areas, the percentage of women who are not exposed to any media is twice as high as in urban areas (48 percent compared with 24 percent). In rural areas, women often have no access to media at all even if media exists in the household. This is because housework takes up the majority of their time and the radio is often considered the property of the man, who may take it with him when he leaves the house. The differential is also wide for men: the proportion of men not exposed to any media varies from 21 percent in rural areas to 9 percent in urban areas. Results by province show significant differences between the City of Kigali and other provinces: indeed, in Kigali more than one-quarter of men are exposed to all three media, compared with approxi- mately 2 percent of men elsewhere. For women, the proportion varies from a maximum of 8 percent in Kigali to less than 1 percent in the North and East provinces. Educational attainment has a significant impact on the level of media exposure. For both men and women, those who have completed secondary or postsecondary levels are the most likely to be exposed to all three media: 11 percent of women and 24 percent of men, compared with only 1 percent of women and 2 percent of men who have completed only Characteristics of Household Respondents | 31 primary school. In addition, the results show that 62 percent of women with no education are not exposed to any media, compared with 13 percent of women with secondary or postsecondary educations. For men, 32 percent of those with no education are not exposed to any media, while only 5 percent of those with secondary or postsecondary educations are not exposed to any media. As in the case of educational attainment, there is a positive relationship between household wealth and media exposure. Men and women in the richest households have the highest levels of exposure to all three media: 7 percent of women and 17 percent of men, compared with less than 1 percent of men and 0 percent of women in the poorest households. 3.5 EMPLOYMENT The RDHS-III asked both men and women whether they were employed at the time of the survey. Respondents who reported having held a job, paid or unpaid, in any sector during the 12 months preceding the survey were considered employed. Table 3.6 shows that, at the national level, 9 percent of women were not working at the time of the survey even if they reported working in the preceding 12 months. More than three in five women (64 percent) were employed at the time of the survey. The percentage of women working at the time of the survey increases steadily with age, rising from 44 percent at age 15-19 to 76 percent at age 40-44. Women who were separated, divorced or widowed (74 percent) and married women (72 percent) were the most likely to be working at the time of the survey. The number of children also affects a woman’s level of employment. As the number of children increases, the proportion of women who work also increases, from 70 percent among women with only one or two children, to 74 percent among women with three children or more. Data by residence show that rural areas have the highest proportion of women working at the time of the survey (66 percent, compared with 54 percent in urban areas). The City of Kigali has the lowest percentage of women working (44 percent). In the provinces, the proportion of employed women ranges from 59 percent in the West province, to 61 percent in the North province, to a maximum of 72 percent in the South province and 74 percent in the East province. Results by educational attainment show that women with no education (70 percent) are proportionally more likely to be employed than women who have completed primary school (64 percent) and women who have completed secondary or postsecondary educations (53 percent). Finally, women in households in the two poorest wealth quintiles are more likely to be employed (73 percent and 77 percent) than women in the richest households (52 percent). The results for men show that 52 percent of men had some form of employment at the time of the survey. This is lower than for women (64 percent). As with women, the percentage of men working at the time of the survey increases with age, from 29 percent for those age 15 to 19, to 61 percent for those age 50 to 54. With respect to marital status, the results show married men and separated or divorced men being proportionally more likely to be working (59 percent for married men, 57 percent for separated, divorced, or widowed men) than other men. With respect to residence, urban areas had the highest proportion of men working at the time of the survey: 61 percent, compared with 50 percent in rural areas. With respect to educational attainment, the results show men with no education (59 percent) being proportionally more likely to be employed than men with primary educations (49 percent) and men with secondary or postsecondary education (54 percent). Finally, unlike women, the proportion of men working is lower in the poorest households than in the richest households (51 percent compared with 60 percent). 32 | Characteristics of Household Respondents Table 3.6 Employment status Percent distribution of women and men by employment status, according to background characteristics, Rwanda 2005 Women Men Employed in the 12 months preceding the survey Employed in the 12 months preceding the survey Background characteristic Currently employed Not currently employed Not employed in the 12 months preceding the survey Total Number Currently employed Not currently employed Not employed in the 12 months preceding the survey Total Number Age 15-19 43.5 6.2 50.0 100.0 2,585 29.4 3.5 66.3 100.0 1,102 20-24 62.8 10.0 27.2 100.0 2,354 57.0 5.2 37.4 100.0 946 25-29 70.2 8.4 21.4 100.0 1,738 56.9 7.5 35.3 100.0 632 30-34 72.0 10.4 17.6 100.0 1,466 58.8 6.7 34.6 100.0 509 35-39 73.8 8.7 17.5 100.0 1,134 59.9 6.7 33.3 100.0 442 40-44 75.9 10.6 13.5 100.0 1,135 58.2 3.8 37.8 100.0 404 45-49 75.1 10.4 14.5 100.0 910 58.3 5.6 35.9 100.0 378 50-54 na na na na na 61.0 8.0 30.6 100.0 260 55-59 na na na na na 60.0 3.6 36.3 100.0 147 Marital status Never married 50.4 7.8 41.7 100.0 4,263 43.4 5.0 51.1 100.0 2,196 Married 71.8 9.3 18.9 100.0 5,510 58.7 5.8 35.4 100.0 2,500 Divorced, separated, widowed 74.3 10.7 15.0 100.0 1,548 56.5 6.0 37.4 100.0 125 Number of living children 0 51.4 7.7 40.8 100.0 4,363 52.1 5.4 42.2 100.0 1,928 1-2 69.7 9.2 21.1 100.0 2,722 51.1 5.5 43.0 100.0 1,306 3-4 74.1 9.1 16.8 100.0 2,266 52.0 4.6 43.2 100.0 1,014 5 + 73.1 11.0 15.9 100.0 1,970 50.9 6.9 41.9 100.0 571 Residence Urban 53.5 8.9 37.4 100.0 1,921 60.9 8.7 29.6 100.0 840 Rural 66.3 8.9 24.8 100.0 9,400 49.7 4.7 45.3 100.0 3,980 Province Kigali city 43.5 10.6 45.7 100.0 1,127 55.7 10.8 33.1 100.0 523 South 71.7 11.4 16.7 100.0 2,958 43.0 6.2 50.6 100.0 1,250 West 58.8 7.8 33.5 100.0 2,824 51.5 7.5 40.7 100.0 1,185 North 60.6 9.5 29.9 100.0 2,063 52.8 2.9 43.7 100.0 845 East 73.9 5.8 20.3 100.0 2,348 59.4 1.4 38.9 100.0 1,017 Education No education 70.2 9.8 20.0 100.0 2,646 59.3 5.1 35.2 100.0 839 Primary 63.6 8.9 27.5 100.0 7,591 49.3 5.2 45.3 100.0 3,389 Secondary or higher 53.0 6.8 39.9 100.0 1,084 54.1 7.1 37.6 100.0 592 Wealth quintile Lowest 73.3 7.7 18.9 100.0 2,421 50.8 3.5 45.5 100.0 867 Second 76.9 8.4 14.7 100.0 2,325 52.4 3.6 44.1 100.0 884 Middle 62.9 9.4 27.7 100.0 2,099 50.6 6.7 42.4 100.0 978 Fourth 54.0 10.5 35.4 100.0 2,133 44.1 6.1 49.2 100.0 1,004 Highest 52.2 8.8 38.8 100.0 2,342 59.6 6.7 33.2 100.0 1,087 Total 64.1 8.9 26.9 100.0 11,321 51.7 5.4 42.6 100.0 4,820 na = Not applicable Characteristics of Household Respondents | 33 Table 3.7.1 shows women’s occupations. The majority of women who were employed at the time of the survey, or who had worked during the 12 months preceding it, were employed in agriculture (86 percent). Among those working in other occupations (13 percent), 5 percent worked in the sales and services; 4 percent performed unskilled manual labor, and 1 percent performed skilled manual labor. Only 3 percent reported working in a technical or administrative occupation. As expected, the data by residence show that the proportion of women working in agriculture is higher in rural areas (92 percent, compared with 44 percent in urban areas). It is much lower in the City of Kigali (27 percent). Outside the City of Kigali, the lowest proportion of women working in agriculture is 88 percent. With respect to educational attainment, 94 percent of women with no education work in agriculture compared with 39 percent of women with secondary or postsecondary education. Table 3.7.1 Occupation: women Percent distribution of women employed in the 12 months preceding the survey by occupation, according to background characteristics, Rwanda 2005 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agri- culture Missing Total Number of women Age 15-19 0.8 0.1 4.5 0.9 12.0 79.4 2.4 100.0 1,285 20-24 1.7 0.8 5.3 1.4 6.5 83.7 0.6 100.0 1,712 25-29 4.7 1.1 6.3 1.3 2.4 84.0 0.2 100.0 1,367 30-34 4.3 0.8 6.1 1.1 1.8 85.6 0.2 100.0 1,207 35-39 3.8 1.1 3.9 1.4 2.4 87.1 0.4 100.0 935 40-44 2.7 0.7 3.7 0.6 1.5 90.5 0.4 100.0 981 45-49 1.4 0.4 2.7 0.5 1.4 93.5 0.1 100.0 778 Marital status Never married 2.7 1.0 5.4 1.7 10.6 77.0 1.6 100.0 2,480 Married 2.7 0.7 4.6 0.7 1.4 89.8 0.2 100.0 4,470 Divorced, separated, Widowed 3.1 0.4 4.8 1.3 3.1 86.8 0.6 100.0 1,316 Residence Urban 9.9 3.7 17.6 3.7 19.2 44.4 1.5 100.0 1,199 Rural 1.6 0.2 2.7 0.6 1.9 92.4 0.5 100.0 7,067 Province Kigali city 13.9 5.4 18.9 5.2 27.0 26.9 2.8 100.0 609 South 1.9 0.5 2.2 0.8 2.7 91.2 0.7 100.0 2,460 West 1.7 0.2 6.6 0.6 2.9 87.3 0.5 100.0 1,879 North 2.7 0.4 4.3 1.2 3.0 88.1 0.3 100.0 1,446 East 1.4 0.2 2.4 0.5 2.0 93.1 0.3 100.0 1,872 Education No education 0.5 0.0 2.4 0.5 2.7 93.7 0.2 100.0 2,116 Primary 0.7 0.1 4.8 1.0 5.0 87.8 0.6 100.0 5,503 Secondary or higher 27.7 8.0 13.2 4.0 5.5 39.1 2.5 100.0 648 Wealth quintile Lowest 0.2 0.1 0.9 0.3 1.4 96.7 0.4 100.0 1,962 Second 0.2 0.1 1.4 0.5 0.6 96.9 0.3 100.0 1,984 Middle 1.0 0.1 4.5 0.9 2.4 90.5 0.6 100.0 1,517 Fourth 1.0 0.1 4.9 1.4 2.9 89.2 0.6 100.0 1,376 Highest 13.6 3.7 15.4 2.9 17.6 45.2 1.6 100.0 1,427 Total 2.8 0.7 4.8 1.1 4.4 85.5 0.7 100.0 8,266 34 | Characteristics of Household Respondents Table 3.7.2 shows men’s occupations. Like women, the majority of men work in agriculture (62 percent). One in seven men performs unskilled manual labor (14 percent), and 11 percent perform skilled manual labor. As expected, the proportion of men working in agriculture is higher in the rural areas (73 percent compared with 18 percent in urban areas). However, the proportion of men performing skilled or unskilled manual labor is significantly higher in urban areas than in rural areas (23 percent compared with 8 percent in rural areas for skilled, and 28 percent compared with 10 percent in rural areas for unskilled). With respect to educational attainment, the results show that, like women, the majority of men with no education work in agriculture (78 percent, compared with 22 percent of those with at least a secondary education). However, of those with the highest educational attainment levels, 37 percent work in managerial or technical occupations. Results by wealth quintile show that a majority of men in the poorest households work in agriculture (86 percent). Conversely, in the richest quintile, only 21 percent of men work in agriculture, and 18 percent work in managerial or technical occupations. Table 3.7.2 Occupation: men Percent distribution of men employed in the 12 months preceding the survey by occupation, according to background characteristics, Rwanda 2005 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agri- culture Missing Total Number of men Age 15-19 0.9 0.0 4.5 3.7 32.2 57.9 0.8 100.0 363 20-24 3.4 0.1 6.9 8.7 21.1 57.6 2.1 100.0 588 25-29 8.9 0.4 7.7 13.8 12.5 55.4 1.3 100.0 407 30-34 9.8 1.9 8.0 14.3 6.9 58.0 1.0 100.0 333 35-39 7.5 1.2 4.1 15.6 7.9 63.3 0.5 100.0 295 40-44 9.1 1.5 7.5 10.0 6.0 65.8 0.0 100.0 251 45-49 4.1 1.1 3.8 13.2 7.0 70.8 0.0 100.0 242 50-54 5.1 2.3 0.0 11.2 6.6 74.8 0.0 100.0 179 55-59 9.5 0.0 2.1 13.4 4.1 70.9 0.0 100.0 94 Marital status Never married 5.8 0.3 7.9 8.8 23.9 51.5 1.8 100.0 1,063 Married 6.2 1.1 4.5 12.5 7.8 67.4 0.4 100.0 1,611 Divorced, separated, widowed 4.3 2.9 0.0 11.9 7.5 73.3 0.0 100.0 78 Residence Urban 14.3 2.7 11.6 23.2 27.9 18.1 2.1 100.0 584 Rural 3.7 0.3 4.1 7.8 10.3 73.2 0.6 100.0 2,168 Province Kigali city 16.2 3.2 12.5 25.7 32.4 7.0 2.9 100.0 348 South 4.9 0.3 5.6 11.5 13.0 63.6 1.2 100.0 615 West 5.8 0.6 6.1 8.6 10.0 68.5 0.4 100.0 700 North 5.3 0.5 3.9 13.6 15.5 60.8 0.3 100.0 471 East 2.1 0.6 2.9 3.2 8.2 82.5 0.5 100.0 619 Education No education 0.3 0.0 2.1 5.3 13.7 78.2 0.4 100.0 540 Primary 1.5 0.2 6.4 11.3 15.6 64.2 0.8 100.0 1,849 Secondary or higher 37.2 5.3 7.6 18.4 6.6 22.4 2.5 100.0 362 Wealth quintile Lowest 0.5 0.0 1.9 7.1 4.7 85.7 0.2 100.0 471 Second 0.7 0.0 2.8 7.4 10.7 78.1 0.3 100.0 495 Middle 1.6 0.4 3.3 8.3 10.0 75.9 0.6 100.0 560 Fourth 4.1 0.0 5.1 10.6 15.2 64.4 0.4 100.0 505 Highest 17.9 2.9 12.5 18.5 24.8 20.9 2.5 100.0 721 Total 6.0 0.8 5.7 11.0 14.0 61.5 0.9 100.0 2,752 Characteristics of Household Respondents | 35 Table 3.8 shows the distribution of women employed during the 12 months preceding the survey by type of earnings, type of employer, and continuity of employment. Overall, 57 percent of women were not paid for their work, 16 percent were paid in cash and in kind, 15 percent were paid in cash only, and 12 percent were paid in kind only. Women in nonagricultural occupations were more likely to be paid in cash (82 percent) than those working in agriculture (4 percent). In the majority of cases (73 percent), women are self-employed, regardless of their occupations. Women who work in agriculture are more likely to work for a family member than women in nonagricultural occupations (17 percent compared with 3 percent). Finally, 75 percent of all women work all year, with the largest proportion working in agriculture (77 percent). Table 3.8 Type of employment Percent distribution of women 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 nonagricultural), Rwanda 2005 Employment characteristic Agricultural work Nonagricultural work Total Type de earnings Cash only 4.0 81.8 14.8 Cash and in-kind 17.4 5.7 15.7 In-kind only 14.4 0.7 12.4 Not paid 64.2 11.7 57.0 Total 100.0 100.0 100.0 Type of employer Employed by family member 16.8 3.4 14.8 Employed by non-family member 7.0 40.4 11.6 Self-employed 76.2 55.5 73.4 Total 100.0 100.0 100.0 Continuity of employment All year 76.5 67.8 75.2 Seasonal 15.3 14.3 15.2 Occasional 8.2 17.7 9.5 Total 100.0 100.0 100.0 Number of women 7,066 1,146 8,266 Note: Total includes women with missing information on type of employment. Table 3.9 shows the distribution of women employed in the 12 months preceding the survey by type of employer, according to background characteristics. Approximately three-quarters of women are self-employed (73 percent). This proportion increases with age, from 63 percent in the age group 20-24 years to 90 percent among those age 45-49 years. The proportion of self-employed women is highest in rural areas (76 percent) and among those having no education (82 percent). Moreover, nearly one in eight women (12 percent) is employed by a non-family member. This is especially true of women in urban areas (29 percent) and women with a secondary or postsecondary education (33 percent). In addition, more than one in eight women (15 percent) works for a family member, a situation affecting 43 percent of the youngest women, 16 percent of women in rural areas, and 19 percent of women with a primary education. 36 | Characteristics of Household Respondents Table 3.9 Type of employer Percent distribution of women employed in the 12 months preceding the survey by type of employer, according to background characteristics, Rwanda 2005 Type of employer Background characteristic Employed by family member Employed by non-family member Self- employed Total1 Number of women Age 15-19 43.0 17.0 39.5 100.0 1,285 20-24 23.6 13.2 63.0 100.0 1,712 25-29 9.2 10.7 79.8 100.0 1,367 30-34 4.6 10.2 85.1 100.0 1,207 35-39 4.1 10.3 85.7 100.0 935 40-44 2.9 8.5 88.5 100.0 981 45-49 2.2 8.1 89.5 100.0 778 Residence Urban 9.1 29.2 61.3 100.0 1,199 Rural 15.8 8.6 75.5 100.0 7,067 Education No education 7.0 10.9 81.9 100.0 2,116 Primary 18.7 9.2 71.8 100.0 5,503 Secondary or higher 6.9 33.3 59.5 100.0 648 Total 14.8 11.6 73.4 100.0 8,266 1 Includes those with missing information Fertility | 37 FERTILITY 4 For more than 20 years, Rwanda has been collecting sociodemographic data to evaluate the fertility levels and characteristics of its population. These efforts include the 1978 RGPH (General Population and Housing Census), the 1983 ENF (National Fertility Survey), the 1991 RGPH, the 1992 RDHS-I (Rwanda Demographic and Health Survey), the 1996 ESD (Socio-demographic Survey), the 2000 RDHS-II, the 2002 RGPH, and the current survey, the 2005 RDHS-III. Information on fertility obtained by the RDHS-III is used to estimate fertility levels, to determine the timing of births, and to describe the fertility characteristics of such variables as residence and educational attainment. It provides recent indicators of fertility rates and birth spacing not only at the national level, but also by province and residence. Fertility is one of the three principle components of population dynamics, the others being mortality and migration (United Nations, 1973). For this reason, the collection of data on fertility levels, trends, and differentials has been a prime objective of the Demographic and Health Surveys program since its inception. The continued collection of fertility data has been essential to recognizing the important role that fertility plays in Rwanda’s overall population growth equation. Rwanda has been conducting national fertility surveys since 1983, using them as the primary basis for developing its population policies. This chapter analyzes the fertility data gathered by the RDHS-III, which have been used to estimate fertility levels, trends, and differentials according to selected background characteristics. The chapter also presents results for age at first birth and birth intervals, and concludes with an analysis of teenage fertility, which has become critical to the issue of the fertility transition, and is a special emphasis of the National Reproductive Health Policy Declaration. Fertility data were obtained by posing a series of questions to all eligible women respondents. During the interview, interviewers recorded the total number of children to whom the woman had given birth, the gender of each child, the number of children currently living with the mother, the number of children living elsewhere, the number of children who had died, and the number still living. A complete birth history was compiled, from the earliest to the most recent birth. In addition, the following information was gathered for each birth: type of birth (single or multiple), sex of child, date of birth, and survival status. For living children, respondents were asked the current age of the child and whether the child was living with its mother or elsewhere. For children who had died, respondents were asked the age at the time of death. At the end of the interview, the interviewer verified that the number of children reported by the mother initially (for each category: living and dead) was consistent with the number of children obtained from the birth history. Because this is a retrospective survey, the data can be used to estimate not only current fertility levels, but also fertility trends over the past 20 years. Despite the organization and controls established to ensure the achievement of survey objectives (including training, instructions to field and data processing personnel, and quality controls at all levels), the data obtained may be subject to various types of errors, primarily errors inherent in all retrospective surveys, including: 38 | Fertility • Underreporting of births, in particular, the omission of children living elsewhere and children who died very young (a few days or hours after birth), which can result in under- estimation of fertility levels. • Misreporting of date of birth and/or age, in particular, the tendency to round off ages or year of birth, which can result in under- or overestimation of fertility at certain ages and/or for certain periods. • Selective survival bias or selectivity effect: the women surveyed are those who have survived. Assuming that the fertility of women who died prior to the survey differs from the fertility of the survivors, the fertility levels obtained by the survey may be slightly biased. Finally, for the men’s survey, as for the women’s survey, information was gathered concerning total fertility by asking men a series of questions, including: the number of children they had, the gender of each child, the number of children living with them, the number living elsewhere, the number of children who had died, and the number still living. However, the men were not asked to provide a complete birth history. 4.1 FERTILITY LEVELS AND DIFFERENTIALS Current fertility levels are measured in terms of age-specific fertility rates (ASFRs) and the total fertility rate (TFR). ASFRs are calculated by dividing the number of births in each age group into the total number of women for that age group. The TFR, a common measurement of current fertility, is the average of all of the ASFRs. It corresponds to the number of children the average woman would bear in her lifetime if fertility rates were to remain constant at the level prevailing during the period under consideration, in this case, the three years preceding the survey. Table 4.1, illustrated by Figure 4.1, indicates that, at the national level, general fertility rates (GFRs) by age group follow the classic pattern of high fertility countries. In Rwanda, this pattern is characterized by relatively high early fertility (42 births per 1,000 for women age 15-19), followed by a rapid increase to very high levels for women age 20-24 (235 per 1,000), 25-29 (305 per 1,000), and 30-34 (273 per 1,000). This high fertility is sustained over a very long period (211 per 1,000 at age 35-39), before declining precipitously at the very end of the childbearing years (32 per 1,000 at age 45-49). These data show that the fertility of Rwandan women remains very high: at the end of her childbearing years, a Rwandan woman has an average of 6.1 children, nearly identical to the TFR of the 1992 RDHS-I (6.2), and even slightly higher than the TFR of the 2000 RDHS-II (5.8). However, the 1994 genocide seems to have had the effect of slowing the significant decline in fertility observed since the National Fertility Survey of 1983 (TFR of 8.5). The data in Table 4.1 show clear differentials in fertility by residence: women in urban areas have lower fertility than those in rural areas. The TFR, estimated at 6.1 children per woman for the country as a whole, ranges from 4.9 in urban areas to 6.3 in rural areas. This means that, if Table 4.1 Current fertility Age-specific and cumulative fertility rates, the general fertility rate, and the crude birth rate for the three years preceding the survey, by urban-rural residence, Rwanda 2005 Residence Age Urban Rural Total 15-19 35 43 42 20-24 172 249 235 25-29 269 313 305 30-34 228 283 273 35-39 170 218 211 40-44 90 121 117 45-49 17 34 32 TFR 4.9 6.3 6.1 GFR 152 198 190 CBR 39.8 43.8 43.2 Note: Rates for age group 45-49 may be slightly biased because of truncation. TFR: Total fertility rate for ages 15-49, expressed per woman GFR: General fertility rate (births divided by the number of women age 15-44), expressed per 1,000 women CBR: Crude birth rate, expressed per 1,000 popu- lation Fertility | 39 current fertility levels were to remain constant, by the end of her childbearing years a woman living in a rural area would have an average of 1.4 children more than a woman living in an urban area. This differential in fertility levels is seen at all ages (Figure 4.1). Childbearing begins very early for women in rural areas: 43 per 1,000 for women age 15 to 19, compared with 35 per 1,000 for this age group in urban areas. At ages 20 to 24, 1,000 women in rural areas give birth to an average of 249 children, compared with 172 for women in urban areas. However, women reach their peak fertility between the ages of 25 and 29 in both rural (313 births per 1,000) and urban (269 births per 1,000) areas. Table 4.1 shows the crude birth rate (CBR), or average number of live births annually in the total population, estimated at 43 per 1,000 for the country as a whole, and the general fertility rate (GFR), that is, the average number of live births per 1,000 women of reproductive age (15-44), estimated here at 190 per 1,000. Like the TFR, these two indicators vary significantly by residence. Rural areas have a GFR of 198 per 1,000, which means that 1,000 women in rural areas are giving birth to an average of 46 more children annually than their urban counterparts (GFR of 152 per 1,000). Similarly, the CBR for rural areas (44 per 1,000) is 4 points higher than the CBR for urban areas (40 per 1,000). Table 4.2 presents fertility rates by background characteristic. The TFR varies considerably by province, ranging from a high of 6.6 children per woman in the West province to a low of 4.3 children per woman in the City of Kigali. In other words, women in the West province have an average of 2.3 more children than women in the City of Kigali. The TFR is strongly correlated with level of educational attainment, varying from a low of 4.3 children for women with secondary educations or higher, to 6.9 for women with no education. This means that a woman with no education (6.9) has an average of 0.8 more children than a woman who has attended primary school (6.1), and an average of 2.6 more children than a woman who has attended secondary school or higher (4.3). Figure 4.1 Age-Specific Fertility Rates, by Residence RDHS 2005 , , , , , , , + + + + + + + * * * * * * * 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 Births per 1,000 women Urban Rural Total* + , 40 | Fertility Table 4.2 and Figure 4.2 show the mean number of live births for women age 40 to 49. This figure is an indicator of completed, or cumulative fertility. Unlike the TFR, which measures the current or recent fertility of women age 15 to 49, cumulative fertility shows the past fertility of women surveyed at the end of their childbearing years. In a population whose fer- tility does not change, the cumulative fertility rate more or less coincides with the TFR. But TFRs that are lower than the mean number of children ever born to women at the end of their childbearing years indicate a downward trend in fertility. In Rwanda, the total cumulative fertility rate is estimated at 6.6 children. This is slightly higher than the TFR (6.1). The difference, though small, suggests a slight decline in fer- tility. In the 1992 RDHS-I, the difference be- tween the two was 1.5 children; in the 2000 RDHS-II, it was 1 child. As stated previously, the significant downward trend observed be- tween 1983 (ENF – National Fertility Survey) and 1992 (RDHS-I) did not continue. Table 4.2 Fertility by background characteristics Total fertility rate for the three years preceding the survey, percentage of women 15-49 currently pregnant, and mean number of children ever born to women age 40-49 years, by background characteristics, Rwanda 2005 Background characteristic Total fertility rate1 Percentage currently pregnant 1 Mean number of children ever born to women age 40-49 Residence Urban 4.9 6.3 5.8 Rural 6.3 8.3 6.7 Province Kigali city 4.3 6.9 5.9 South 5.6 7.6 6.1 West 6.6 7.9 7.1 North 6.4 7.9 6.7 East 6.5 9.0 6.6 Education No education 6.9 8.4 6.8 Primary 6.1 8.1 6.5 Secondary or higher 4.3 6.2 4.9 Wealth quintile Lowest 6.1 8.4 6.8 Second 6.3 7.7 6.6 Middle 6.7 8.2 6.5 Fourth 6.4 9.8 6.8 Highest 5.0 5.9 6.1 Total 6.1 8.0 6.6 1 Women age 15-49 years Figure 4.2 Total Fertility Rate and Mean Number of Children Ever Born to Women Age 40-49 6.1 4.9 6.3 6.9 6.1 4.3 6.1 6.3 6.7 6.4 5.0 6.6 5.8 6.7 6.8 6.5 4.9 6.8 6.6 6.5 6.8 6.1 RWANDA RESIDENCE Urban Rural EDUCATION No education Primary Secondary or higher WEALTH QUINTILE Lowest Second Middle Fourth Highest 0.0 2.0 4.0 6.0 8.0 Number of children per woman TFR Mean number of children ever born RDHS 2005 Fertility | 41 The fertility results by background characteristic show cumulative fertility rates above the TFR for all categories except women with no education, indicating that fertility is declining for all women, regardless of residence or province. However, the difference between cumulative fertility (number of children ever born) and the TFR is greater in the City of Kigali (1.6 children) and in the wealthiest households (1.1 children) than anywhere else. Table 4.2 shows the percentage of women who reported being pregnant at the time of the survey. At the national level, 8 percent of women reported being pregnant. This is likely to be an underestimate because women in the early stages of pregnancy may be unaware or unsure of their pregnancy status. Age, residence, culture, and/or beliefs may also affect a woman’s willingness to report her condition. In Rwanda, women generally declare their pregnancies only when their condition they becomes visible. For these reasons, the differentials in pregnancy rates shown here must be interpreted with a great deal of caution. It should be noted, however, that the findings are generally consistent with current fertility levels. In fact, the lowest pregnancy rates are observed for women living in the wealthiest households (6 percent), women with a secondary education or higher (6 percent), and women living in the City of Kigali (7 percent), which are the groups that also have the lowest fertility levels. 4.2 FERTILITY TRENDS Two national demographic data collection efforts are conducted regularly in Rwanda: the General Population and Housing Census and the Demographic and Health Survey (DHS). The Censuses of 1978, 1991, and 2002 gathered information on natural population dynamics and were used to estimate fertility levels for those years by asking questions about births that occurred in the 12 months preceding the survey. However, this method generally results in underestimates of fertility levels. The DHS surveys employ a more accurate method (women’s birth histories), which yield more reliable results. Yet the various RDHS surveys (1992, 2000, and 2005) and the Censuses of 1991 and 2002 have produced more or less similar results with respect to the TFR, which fluctuates around 6. This means that fertility has remained relatively stable in Rwanda since the RDHS-I. Table 4.3 shows age-specific fertility rates (ASFRs) for the three DHS surveys. Figure 4.3 examines past fertility trends based on the results of the 1992 RDHS-I, the 2000 RDHS-II, and the 2005 RDHS-III. The three ASFR curves follow a similar pattern: they increase rapidly with age and reach their peak between the ages of 25 and 29, then taper off steadily as they move toward the age group 45 to 49. It should be emphasized that the decline slows with this age group, demonstrating high levels of late fertility. However, the curve for the current survey (2005 RDHS-III) drops lower after age 40 than the other two curves, indicating a trend toward declining fertility in women of these generations. Table 4.3 Trends in fertility Age-specific fertility rates (per 1,000 women) and total fertility rates, 1992 RDHS, 2000 RDHS, and 2005 RDHS Age group 1992 RDHS 2000 RDHS 2005 RDHS 15-19 60 52 42 20-24 227 240 235 25-29 294 272 305 30-34 270 257 273 35-39 214 190 211 40-44 135 123 117 45-49 46 33 32 TFR 6.2 5.8 6.1 Note: Age-specific fertility rates are per 1,000 women. 42 | Fertility The data collected in the RDHS-III were used to track fertility trends for five-year periods preceding the survey based on women’s ASFRs (Table 4.4 and Figure 4.4). Fertility rates declined fairly steadily between the earliest period (15-19 years prior to the survey) and the most recent period, except for the youngest age groups (15-19 and 20-24) in the period 5-9 years preceding the survey (1996- 2000). The slight increase in the TFR in 2005 seems to be the result of an increase in fertility among women age 20 to 35. In other words, fertility among teenagers (age 15-19) has been declining steadily from one survey to the next, as has the fertility of women age 40 and over, especially in the recent periods. The ASFRs for the RDHS-III were used to calculate the TFR for women age 15 to 34—when the greatest number of births occur—for each five-year period. These data, presented in Figure 4.5, were positioned in relation to the central year of each period for which fertility was calculated. This figure also includes similar data from the RDHS-I (by four-year periods) and the RDHS-II (by five-year periods). The data reveal no general trends and no significant changes in fertility levels in Rwanda. In addition, there have been no changes in the factors that generally affect fertility. On the contrary, use of contraception, although it has risen since 2000, has not reached 1992 levels. There are no changes in median age at first birth, or in the other determinants influencing exposure to the risk of pregnancy: age at first marriage has remained relatively stable since 1992, and age at first sexual intercourse has risen only slightly. Table 4.4 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, Rwanda 2005 Number of years preceding survey Age group 0-4 5-9 10-14 15-19 15-19 44 71 53 62 20-24 232 264 236 257 25-29 292 310 321 338 30-34 261 283 289 [334] 35-39 207 232 [259] - 40-44 118 [166] - - 45-49 [34] - - - Note: Age-specific fertility rates are per 1,000 women. Estimates in brackets are truncated. Figure 4.3 Trends in Age-Specific Fertility Rates, Rwanda 1992, 2000, and 2005 , , , , , , , ( ( ( ( ( ( ( # # # # # # # 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 Births per 1,000 women 1992 RDHS 2000 RDHS 2005 RDHS# ( , Fertility | 43 Figure 4.4 Age-Specific Fertility Rates for Five-Year Periods Preceding the Survey + + + + $ $ $ $ $ ! ! ! ! ! ! # # # # # # # 15 20 25 30 35 40 45 50 Woman's age 0 50 100 150 200 250 300 350 400 Births per 1,000 women Years preceding survey 0-4 years 5-9 years 10-14 years 15-19 years# ! $ + RDHS 2005 Figure 4.5 Trends in the Total Fertility Rate among Women Age 15-34, Rwanda 1992, 2000, and 2005 ( ( ( ( # # # # & & & & 79 83 87 88 91 93 94 98 03 Central year of estimates 0 1 2 3 4 5 6 Children per woman 1992 RDHS 2000 RDHS 2005 RDHS& # ( 2005 RDHS 1992 RDHS 2000 RDHS 44 | Fertility 4.3 PARITY AND PRIMARY INFERTILITY Women’s average parity by age group is calculated on the basis of the total number of children ever born in their lifetime. Table 4.5.1 presents these parities for all women and for currently married women. For all women, parity increases steadily and rapidly with age: from an average of 0.04 children at age 15 to 19, parity increases to 0.8 children at age 20 to 24, and to 7.0 children at age 45 to 49, the end of the childbearing years. In addition, the distribution of women by number of children ever born shows relatively late childbearing. Only 3.3 percent of women under the age of 20 have given birth to at least one child. Even at ages 20 to 24, less than half the women (46 percent) have given birth to at least one child, and only 16 percent of the women in this age group have given birth twice. The fertility level accelerates between age 25 and 29: more than one-fifth (23 percent) of women in this age group have given birth to at least 3 children. However, nearly one-quarter of the women in the 30-34 age group have had at least 4 births. Finally, at age 45 to 49, the end of the reproductive period, 16 percent of women have given birth to at least 10 children. Table 4.5.1 Children ever born and living : women Percent distribution of all women and currently married women by number of children ever born, and mean number of children ever born and mean number of living children, according to age group, Rwanda 2005 Number of children ever born Age group 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 96.7 3.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 2,585 0.04 0.03 20-24 54.2 22.8 16.1 5.5 1.2 0.1 0.0 0.0 0.0 0.0 0.0 100.0 2,354 0.77 0.67 25-29 16.6 15.5 21.1 22.9 15.8 5.7 1.6 0.8 0.0 0.0 0.0 100.0 1,738 2.33 1.96 30-34 5.6 8.5 12.1 18.2 23.4 17.1 8.5 4.6 1.3 0.4 0.2 100.0 1,466 3.66 3.01 35-39 3.1 3.9 9.3 10.1 13.9 16.0 17.8 13.4 7.8 2.7 2.0 100.0 1,134 4.96 3.92 40-44 2.9 2.0 3.9 6.6 9.9 11.2 14.4 17.6 12.4 9.3 9.9 100.0 1,135 6.20 4.89 45-49 2.2 1.6 2.0 4.5 5.1 11.0 12.8 16.3 14.0 14.6 15.9 100.0 910 7.02 5.24 Total 37.4 9.6 9.7 9.1 8.5 6.7 5.6 5.1 3.3 2.4 2.5 100.0 11,321 2.68 2.14 CURRENTLY MARRIED WOMEN 15-19 42.9 50.5 4.2 2.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 65 0.66 0.57 20-24 14.7 37.4 32.7 12.2 2.6 0.3 0.1 0.0 0.0 0.0 0.0 100.0 980 1.52 1.32 25-29 4.8 13.7 23.7 27.1 19.8 7.7 2.2 1.1 0.0 0.0 0.0 100.0 1,254 2.80 2.36 30-34 1.2 5.2 10.3 19.0 26.1 19.8 10.3 5.8 1.7 0.4 0.3 100.0 1,112 4.08 3.38 35-39 1.0 1.8 6.0 7.1 14.2 16.3 20.8 16.6 10.1 3.4 2.8 100.0 807 5.55 4.41 40-44 1.4 1.7 1.9 5.0 6.5 8.4 14.5 18.7 15.5 12.5 13.9 100.0 739 6.87 5.46 45-49 1.3 1.4 0.6 3.5 3.3 7.8 12.4 13.8 16.5 18.6 20.8 100.0 554 7.63 5.74 Total 4.9 12.1 14.5 14.3 13.5 10.1 8.8 7.7 5.6 4.1 4.4 100.0 5,510 4.24 3.42 The results for married women do not vary significantly from the results for all women, except for the younger age groups. More than half of married women between the ages of 15 and 19 (57 percent) have had at least one child, compared with 3.3 percent of all women. However, as for all women, the fertility of young married women age 15 to 19 remains relatively low: only 7 percent of the women in this age group have had two or more children. Fertility | 45 Even at age 20-24, when more than half of women are still never-married (53 percent: see Marital Status in Chapter 6), there is still a wide gap between the proportion of married women who have had at least one child (85 percent) and the proportion of all women in this age group who have had at least one child (46 percent). Women who remain childless voluntarily are relatively rare in Rwanda, where the population is still very pro-natal (see Chapter 7, Fertility Preferences). For this reason, zero parity among married women age 35 to 49 would be an indicator of total or primary infertility. In Rwanda, only 1.2 percent of married women age 35 to 49 years (when the arrival of a first child is unlikely) have never had a child and can be considered infertile. This shows that the level of primary infertility has remained stable at low levels since 1983, when it was calculated at 1.5 percent. The percentage was as low as 0.7 percent in the RDHS-I, and was 1.2 percent in the RDHS-II. It should be noted that the level of primary infertility observed in Rwanda in 2005 is lower than the level found in some sub-Saharan countries such as Cameroon (3.6 percent in the 2004 EDSC), but is similar to that of other countries in this region such as Burkina Faso (1 percent in the 2003 EDSBF). The average number of children by age group for men is calculated on the basis of the total number of children ever born to men in their lifetime. Table 4.5.2 shows the number of children ever born for all men and for married men. Table 4.5.2 Children ever born and living : men Percent distribution of all men and currently married men by number of children ever born, and mean number of children ever born and mean number of living children, according to age group, Rwanda 2005 Number of children ever born Age group 0 1 2 3 4 5 6 7 8 9 10+ Total Number of men Mean number of children ever born Mean number of living children ALL MEN 15-19 99.8 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 1,102 0.00 0.00 20-24 83.5 9.9 4.8 1.4 0.3 0.0 0.0 0.1 0.0 0.0 0.0 100.0 946 0.26 0.23 25-29 39.9 17.8 17.9 16.0 5.2 2.0 0.8 0.4 0.1 0.0 0.0 100.0 632 1.40 1.22 30-34 15.7 11.2 15.5 19.5 18.7 12.1 3.2 2.7 0.7 0.7 0.0 100.0 509 2.86 2.38 35-39 8.2 6.1 9.1 14.6 16.8 14.6 13.3 9.4 3.5 3.0 1.5 100.0 442 4.26 3.50 40-44 2.9 2.3 4.7 8.5 11.8 13.1 13.6 16.8 11.1 7.6 7.5 100.0 404 5.90 4.72 45-49 1.7 1.1 3.7 5.7 3.8 8.0 15.4 13.1 12.4 10.8 24.4 100.0 378 7.43 5.79 50-54 1.3 0.8 1.6 2.1 1.6 5.9 7.6 12.0 15.8 13.0 38.4 100.0 260 8.63 6.50 55-59 0.8 1.2 0.7 0.0 3.8 3.8 3.3 14.0 13.3 15.8 43.3 100.0 147 9.28 6.59 Total 47.3 6.4 6.6 7.0 5.8 5.0 4.5 4.7 3.6 3.0 6.1 100.0 4,820 2.76 2.18 CURRENTLY MARRIED MEN 15-19 * * * * * * * * * * * * 2 * * 20-24 20.2 44.3 25.6 7.4 1.9 0.0 0.0 0.5 0.0 0.0 0.0 100.0 173 1.29 1.15 25-29 7.6 25.5 27.9 25.5 8.3 3.1 1.2 0.7 0.1 0.0 0.0 100.0 394 2.20 1.92 30-34 3.7 11.2 17.7 22.8 22.0 14.1 3.5 3.3 0.8 0.8 0.0 100.0 429 3.32 2.75 35-39 3.4 5.5 9.3 14.4 17.5 16.1 14.7 10.4 3.6 3.4 1.7 100.0 400 4.56 3.77 40-44 1.0 1.7 4.8 8.1 11.2 13.9 13.6 17.8 11.8 8.1 8.0 100.0 381 6.11 4.90 45-49 0.2 0.6 3.1 4.9 3.3 8.2 14.7 13.7 12.7 11.8 26.6 100.0 346 7.77 6.07 50-54 0.3 0.5 1.0 2.3 1.4 4.8 7.0 12.2 15.6 13.8 41.1 100.0 235 8.90 6.79 55-59 0.5 0.8 0.7 0.0 4.0 3.4 3.5 13.5 13.2 15.9 44.5 100.0 139 9.40 6.69 Total 4.0 10.4 12.0 12.9 10.5 9.4 8.1 8.9 6.5 5.7 11.5 100.0 2,500 5.10 4.06 Note : An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed 46 | Fertility For all men, the average number of children ever born and living increases steadily and rapidly with age, from an average of 0.3 children at age 20 to 24, to 1.4 children at age 25 to 29, to 2.9 children at age 30 to 34, and to 9.3 children at age 55 to 59. For married men, the average number of children is higher in the younger age groups. A comparison of these results with those for married women shows that the average number of children increases more rapidly for married women than for married men. However, in the older age groups (45 to 49), the average number of children ever born is much higher for married men than married women. 4.4 BIRTH INTERVALS Examination of birth intervals, defined as the length of time between two successive live births, is important not only for their impact on the health status of both mother and child, but for their role in fertility analysis and the design of reproductive health programs. Currently, short birth intervals (less than 24 months) are considered harmful to the health and nutritional status of children, increasing their risk of death. In addition, short birth intervals diminish a woman’s physiological capacity, exposing her to a greater risk of complications during and after pregnancy (miscarriage, eclampsia, etc.), and are also associated with high cumulative fertility. Table 4.6 shows the distribution of non-first births in the five years preceding the survey by number of months since the preceding birth, according to background characteristics. Table 4.6 shows that 8 percent of births occur less than 18 months after the preceding birth and that 15 percent of children are born between 18 and 24 months after the birth of their immediately older sibling. In 23 percent of all cases, the birth interval is less than two years. However, a large proportion of births (41 percent) occur between 2 and 3 years after the preceding birth, and more than one-third of all children (36 percent) are born three years or more after the birth of their immediately older sibling. The mean duration of the birth interval is slightly more than two and a half years (31.3 months), which means that half of all births take place 31.3 months after the preceding birth. With respect to age, birth intervals are shorter for younger women; that is, the younger the woman, the shorter the birth interval: the mean duration increases from 27.7 months at age 20 to 29, to 37.1 months at age 40 to 49. The results also show a significant increase in the length of birth intervals associated with birth order, from 29.9 months for birth orders 2-3, to 32.7 months for birth orders 7 and higher. However, differentials by gender are not significant (31.6 months for boys; 30.9 months for girls). Survival of the preceding child is an important factor. When the preceding child has died, the birth interval between that birth and the next birth is a median of 26.4 months. When the preceding child is living, the birth interval is a median of 32.1 months, or approximately six months later than the birth following the death of the preceding sibling. The median interval between births is lowest in rural areas (31.4 months, compared with 29.9 in urban areas). In 2005, the differential between rural and urban areas was 1.5 months; in 2000, it was 3.2 months. This is not a very significant difference; it is to be expected that women in urban areas, who have greater access to family planning services, will have much longer birth intervals than women in rural areas. With respect to provinces, the birth interval varies from a low of 30.1 months in the City of Kigali to a high of 32.6 months in the South province. Fertility | 47 Table 4.6 Birth intervals Percent distribution of non-first births in the five years preceding the survey by number of months since preceding birth, according to background characteristics, Rwanda 2005 Months since preceding birth Background characteristic 7-17 18-23 24-35 36-47 48+ Total Number of non-first births Median number of months since preceding birth Mother’s age 15-19 * * * * * * 7 * 20-29 12.2 20.1 43.4 14.6 9.7 100.0 2,578 27.7 30-39 6.2 13.7 41.0 21.3 17.7 100.0 3,243 32.5 40-49 4.2 8.6 34.1 24.2 29.0 100.0 1,245 37.1 Birth order 2-3 10.6 17.2 38.9 16.3 17.0 100.0 2,880 29.9 4-6 6.3 14.0 42.3 20.7 16.7 100.0 2,753 31.9 7+ 6.3 13.3 41.4 22.7 16.4 100.0 1,441 32.7 Sex of preceding birth Male 7.9 15.0 40.4 19.5 17.2 100.0 3,611 31.6 Female 8.2 15.4 41.0 19.2 16.2 100.0 3,463 30.9 Survival of preceding birth Living 5.4 14.2 42.7 20.5 17.1 100.0 5,834 32.1 Dead 20.4 19.5 31.1 14.1 15.0 100.0 1,240 26.4 Residence Urban 11.5 16.4 36.3 15.7 20.1 100.0 968 29.9 Rural 7.5 15.0 41.4 19.9 16.2 100.0 6,106 31.4 Province Kigali city 12.7 16.3 32.6 16.6 21.9 100.0 486 30.1 South 6.9 14.9 39.1 21.1 18.0 100.0 1,708 32.6 West 7.9 16.6 41.7 19.2 14.6 100.0 1,874 30.2 North 6.8 13.4 46.3 18.7 14.8 100.0 1,445 31.8 East 9.1 15.1 38.6 19.1 18.1 100.0 1,562 31.3 Education No education 7.3 15.0 37.1 21.1 19.5 100.0 2,128 32.7 Primary 8.3 14.9 43.0 18.9 15.0 100.0 4,368 30.8 Secondary or higher 9.4 17.6 37.1 15.7 20.2 100.0 578 30.1 Wealth quintile Lowest 6.2 13.0 40.4 20.9 19.6 100.0 1,513 32.8 Second 6.8 13.0 42.8 20.4 17.0 100.0 1,474 32.1 Middle 8.4 16.1 40.5 20.0 15.0 100.0 1,465 30.9 Fourth 8.4 14.7 41.9 19.4 15.7 100.0 1,395 31.0 Highest 11.1 19.9 37.3 15.4 16.3 100.0 1,226 28.8 Total 8.1 15.2 40.7 19.3 16.7 100.0 7,074 31.3 Note: First-order births are excluded. The interval for multiple births is the number of months since the preceding pregnancy that ended in a live birth. An asterisk indicates that a figure is based on fewer than 25 unweighted cases and has been suppressed. Regarding mother’s level of educational attainment, birth intervals for women with no education are longer (32.7 months) than birth intervals for women with a secondary education or higher (30.1). This is explained, among other things, by the fact that the median duration of breastfeeding is longer for women with no education than for women with some education (see Table 10.3). Household wealth data follow a similar pattern: 50 percent of children in the poorest quintile are born 32.8 months after the birth of their immediately older sibling, or 4 months later than children in the richest quintile (28.8). 48 | Fertility 4.5 AGE AT FIRST BIRTH The age at which childbearing begins is an important demographic indicator because it has a direct bearing on a women’s cumulative fertility, particularly when there is little or no contraceptive use. The younger a woman is when she begins childbearing, the greater her likelihood of having many children. At the same time, having children at too young an age can have negative repercussions on the mother’s health and can put her children at risk of dying. Table 4.7 shows the distribution of women by age at first birth and median age at first birth by age at the time of the survey. The results show that median age at first birth has remained practically unchanged from one generation to the next (from a low of 21.7 to a high of 22.2) and no trends indicate a rise or fall in this median age. Table 4.7 Age at first birth Among all women, the percentage who first gave birth by exact age, and median age at first birth, by current age, Rwanda 2005 Percentage who gave birth by exact age: Age group 15 18 20 22 25 Percentage who have never given birth Number of women Median age at first birth 15-19 0.2 na na na na 96.7 2,585 a 20-24 0.6 7.8 22.3 na na 54.2 2,354 a 25-29 0.6 10.2 31.1 53.4 75.4 16.6 1,738 21.7 30-34 1.3 8.7 24.1 48.3 77.8 5.6 1,466 22.1 35-39 1.3 10.0 25.8 48.2 73.9 3.1 1,134 22.2 40-44 0.9 8.8 28.8 51.1 77.5 2.9 1,135 21.9 45-49 1.1 9.0 25.8 50.6 80.3 2.2 910 22.0 na = Not applicable a = Omitted because less than 50 percent of women had a birth before reaching the beginning of the age group. Table 4.8 shows a median age at first birth of 22.0 years for women age 25 to 49; this is identical to the median age observed for women the same age in the 2000 RDHS-II. Table 4.8 shows median age at first birth according to various socioeconomic characteristics. The first child arrives at a younger age for women in rural areas (21.9 years) than for those in urban areas (22.4 years). The South province has the highest median age at first birth (22.9 years), followed by the City of Kigali (22.4 years). In the other provinces, median age at first birth varies from a low of 21.2 years in the East province to a high of 21.7 years in the West province. Women’s level of educational attainment affects the median age at first birth: women with no education (21.4 years) and women with primary education (22.0 years) have a lower median age at first birth than women with secondary or higher education (23.9 years). Results by household wealth show that the first birth occurs later among women in the richest quintile (22.7 years) than among those in the poorest quintile (21.8 years). Fertility | 49 Table 4.8 Median age at first birth by background characteristics Median age at first birth among women age 25-49 years, by current age and background characteristics, Rwanda 2005 Current age Background characteristic 25-29 30-34 35-39 40-44 45-49 Women age 25-49 Residence Urban 22.2 22.5 23.0 22.1 21.6 22.4 Rural 21.6 22.1 22.0 21.9 22.0 21.9 Province Kigali city 22.4 23.0 23.7 21.2 20.7 22.4 South 22.9 23.1 23.4 22.8 22.5 22.9 West 21.2 22.1 21.4 21.6 22.1 21.7 North 21.2 21.4 22.2 21.7 22.1 21.6 East 21.0 21.3 21.3 21.6 21.1 21.2 Education No education 20.8 21.3 21.2 21.3 21.8 21.4 Primary 21.6 22.2 22.3 22.1 22.1 22.0 Secondary or higher 23.9 23.6 24.5 24.0 22.9 23.9 Wealth quintile Lowest 21.4 21.9 22.2 21.5 21.8 21.8 Second 21.0 22.0 21.2 21.9 22.2 21.7 Middle 21.9 21.9 22.2 21.8 21.9 21.9 Fourth 21.7 22.1 21.9 22.1 21.9 21.9 Highest 22.3 22.9 23.2 22.6 21.9 22.7 Total 21.7 22.1 22.2 21.9 22.0 22.0 4.6 TEENAGE FERTILITY Teenage fertility is an important demographic factor for many reasons. First, children born to very young mothers run a greater risk of illness and death. Second, teenage mothers are more likely to suffer complications during pregnancy and less likely to treat them, exposing them to greater risk of complications during delivery and greater risk of dying for reasons related to childbearing. Third, early childbearing seriously affects a woman’s ability to pursue an education, thereby limiting her job opportunities. In Rwanda, teenagers make up 23 percent of all women of childbearing age, but only 3 percent have had at least one birth. Table 4.9 shows the proportion of teenagers who have already had one or more children, as well as those currently in their first pregnancy. Together these two subgroups make up the proportion of teenagers who have already begun childbearing: 4 percent of young women between the ages of 15 and 19 (3 percent are already mothers; 0.8 percent are pregnant for the first time). At age 15, 0.3 percent of women have begun childbearing, but the percentage increases steadily and rapidly with age: at age 17, 2 percent of women have already had at lea

View the publication