Ghana - Demographic and Health Survey - 2004

Publication date: 2004

Ghana Demographic and Health Survey 2003 G h an a 2003 D em ographic and H ealth Survey The 2003 Ghana Demographic and Health Survey (GDHS) is part of the worldwide MEASURE DHS project. The survey and this publication were made possible through support provided by the U.S. Agency for International Development, under the terms of contract No. HRN-C-00-97-0019-00. The opinions expressed herein are those of the authors and do not necessarily reflect the views of the U.S. Agency for International Development. Ghana Demographic and Health Survey 2003 Ghana Statistical Service Accra, Ghana Noguchi Memorial Institute for Medical Research Legon, Ghana ORC Macro Calverton, Maryland, USA September 2004 Ghana Statistical Service Noguchi Memorial Institute for Medical Research MEASURE DHS+ This report highlights the findings of the 2003 Ghana Demographic and Health Survey (GDHS), a nationally representative survey of 5,691 women age 15-49 and 5,015 men age 15-59. The primary pur- pose of the GDHS is to generate recent and reliable information on fertility, family planning, infant and child mortality, maternal and child health, and nutrition. In addition, the survey collected information on malaria treatment and prevention, anaemia and HIV prevalence. This information is essential for making informed policy decisions, planning, monitoring, and evaluating programmes on health in general and reproductive health in particular, at both the national and regional levels. This survey is the fourth in a series of population and health surveys conducted in Ghana as part of the global Demographic and Health Surveys (DHS) programme. The 2003 GDHS was implemented by the Ghana Statistical Service (GSS) in collaboration with the Noguchi Memorial Institute for Medical Research (NMIMR) and the Ghana Health Service. Technical assistance was provided by ORC Macro through the MEASURE DHS programme. Financial support for the survey was provided by the U.S. Agency for International Development (USAID) and the Govern- ment of Ghana. Additional information about the GDHS may be obtained from the Ghana Statistical Service, P.O. Box 1098, Accra, Ghana (Telephone: (233-21) 671-732 and Fax: (233-21) 671-731). Information about the DHS project may be obtained from ORC Macro, 11785 Beltsville Drive, Calverton, MD (Telephone: 301-572-0200; Fax: 301-572-0999; E-mail: reports@orcmacro.com; Internet: http://www.measuredhs.com). Recommended citation: Ghana Statistical Service (GSS), Noguchi Memorial Institute for Medical Research (NMIMR), and ORC Macro. 2004. Ghana Demographic and Health Survey 2003. Calverton, Maryland: GSS, NMIMR, and ORC Macro. Contents | iii CONTENTS Tables and Figures . ix Foreword. xvii Contributors . xix Summary of Findings . xxi Map of Ghana .xxviii CHAPTER 1 INTRODUCTION .1 1.1 Geography, History, and Economy . 1 1.1.1 Geography.1 1.1.2 History.1 1.1.3 Economy .2 1.2 Demographic Profile. 2 1.3 Population Policy and Reproductive Health Programmes. 3 1.4 Objectives and Organisation of the Survey. 4 1.5 Sample Design. 4 1.6 Questionnaires . 5 1.7 Haemoglobin and HIV Testing. 5 1.7.1 Haemoglobin Testing.6 1.7.2 HIV/AIDS Testing.6 1.8 Pretest, Training, and Fieldwork. 7 1.8.1 Pretest .7 1.8.2 Training and Fieldwork .8 1.9 Data Processing . 8 1.10 Response Rates. 8 CHAPTER 2 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS .11 2.1 Household Population by Age and Sex . 11 2.2 Household Composition . 13 2.3 Educational Attainment of Household Members . 14 2.4 Housing Characteristics. 19 2.5 Household Durable Goods . 22 CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS.23 3.1 Background Characteristics of Respondents . 23 iv | Contents 3.2 Educational Attainment and Literacy. 25 3.3 Access to Mass Media . 29 3.4 Employment . 32 3.4.1 Employment Status .32 3.4.2 Occupation .35 3.4.3 Type of Employer, Form of Earnings, and Continuity of Employment .38 3.4.4 Control Over Earnings and Women’s Contribution to Household Expenditures.40 3.5 Women’s Empowerment . 43 3.5.1 Women’s Participation in Decisionmaking.43 3.5.2 Attitudes toward Wife-beating .46 3.5.3 Attitudes toward Refusing Sex.49 CHAPTER 4 FERTILITY .53 4.1 Fertility Levels and Trends. 53 4.1.1 Fertility Levels.53 4.1.2 Differentials in Current and Completed Fertility .54 4.1.3 Trends in Fertility.56 4.2 Children Ever Born and Children Surviving . 58 4.3 Birth Intervals . 59 4.4 Age at First Birth . 61 4.5 Median Age at First Birth by Background Characteristics . 61 4.6 Teenage Fertility . 62 CHAPTER 5 FAMILY PLANNING .65 5.1 Knowledge of Contraceptive Methods . 65 5.2 Ever Use of Contraception . 68 5.3 Current Use of Contraceptive Methods. 70 5.4 Trends in the Use of Family Planning . 74 5.5 Current Use of Contraception by Women’s Status . 75 5.6 Number of Children at First Use of Contraception . 77 5.7 Use of Social Marketing Brands. 77 5.8 Knowledge of Fertile Period. 79 5.9 Source of Supply . 80 5.10 Informed Choice . 81 5.11 Future Use of Contraception. 84 5.12 Reasons for Not Intending to Use Contraception . 84 5.13 Preferred Method of Contraception for Future Use . 85 5.14 Exposure to Family Planning Messages. 86 5.15 Exposure to Specific Radio Messages on Family Planning . 89 Contents | v 5.16 Contact of Non-users with Family Planning Providers. 91 5.17 Discussion about Family Planning with Husband . 92 5.18 Attitudes of Men towards Family Planning . 93 5.19 Attitudes of Couples towards Family Planning . 94 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY .99 6.1 Current Marital Status . 99 6.2 Polygyny. 100 6.3 Age at First Marriage . 102 6.4 Age at First Sexual Intercourse . 104 6.5 Recent Sexual Activity. 107 6.6 Postpartum Amenorrhoea, Abstinence, and Insusceptibility . 110 6.7 Menopause . 112 CHAPTER 7 FERTILITY PREFERENCES. 113 7.1 Desire For More Children . 113 7.2 Need For Family Planning Services . 116 7.3 Ideal Family Size. 118 7.4 Fertility Planning . 121 7.5 Ideal Family Size And Unmet Need By Women’s Status . 122 CHAPTER 8 INFANT AND CHILD MORTALITY . 125 8.1 Definition, Data Quality and Methodology . 125 8.2 Levels and Trends in Infant and Child Mortality . 127 8.3 Socio-economic Diffferentials in Mortality . 129 8.4 Demographic Characteristics and Child Mortality. 131 8.5 Women’s Status and Child Mortality . 134 8.6 Perinatal Mortality . 135 8.7 High-Risk Fertility Behaviour. 136 CHAPTER 9 MATERNAL AND CHILD HEALTH. 137 9.1 Maternity Care . 137 9.1.1 Antenatal Care. 137 9.1.2 Delivery Care . 144 9.1.3 Postnatal Care . 148 9.2 Reproductive Health Care and Women’s Status. 150 9.3 Child health. 151 9.3.1 Vaccination of Children . 151 9.3.2 Acute Respiratory Infections . 155 9.3.3 Diarrhoeal Diseases . 157 vi | Contents 9.4 Child Health Care and Women’s Status . 163 9.5 Women’s Perceptions of Problems in Obtaining Health Care. 164 9.6 Use of Smoking Tobacco . 165 CHAPTER 10 NUTRITION . 169 10.1 Breastfeeding. 169 10.1.1 Initiation of Breastfeeding . 169 10.1.2 Age Pattern of Breastfeeding . 171 10.2 Complementary Feeding. 175 10.2.1 Types of Complementary Foods. 175 10.2.2 Frequency of Foods Consumed by Children . 176 10.3 Micronutrients . 179 10.3.1 Iodisation of Household Salt . 179 10.3.2 Micronutrient Intake among Children . 179 10.3.3 Micronutrient Intake Among Mothers . 182 10.3.4 Prevalence of Anaemia in Children. 184 10.3.5 Prevalence of Anaemia in Women. 184 10.3.6 Prevalence of Anaemia in Children by Anaemia Status of Mother . 187 10.4 Nutritional Status of Children under Age Five. 187 10.4.1 Measures of Nutritional Status in Childhood . 187 10.4.2 Trends in Children’s Nutritional Status . 191 10.5 Nutritional Status of Women . 191 CHAPTER 11 MALARIA . 195 11.1 Mosquito Nets . 195 11.1.1 Ownership of Mosquito Nets . 195 11.1.2 Use of Mosquito Nets by Children . 196 11.1.3 Use of Mosquito Nets by Pregnant Women . 198 11.2 Exposure to Media Messages on Malaria . 199 11.3 Malaria Diagnosis, Case Management, and Treatment . 200 11.3.1 Malaria Prophylaxis during Pregnancy. 200 11.3.2 Prevalence and Management of Childhood Malaria . 202 Contents | vii CHAPTER 12 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR . 207 12.1 HIV/AIDS–Related Knowledge and Attitudes. 207 12.2 Knowledge of Prevention Methods . 208 12.3 Beliefs about AIDS. 210 12.4 Stigma and Discrimination Associated with HIV/AIDS . 213 12.5 Knowledge of Prevention of Mother-to-Child Transmission . 216 12.6 HIV Testing. 219 12.7 Counselling and Testing Pregnant Women. 221 12.8 Attitudes towards Negotiating Safer Sex . 222 12.9 Higher-Risk Sex and Condom Use . 224 12.10 Paid Sex . 227 12.11 Self-Reporting of Sexually Transmitted Infections and Symptoms . 227 12.12 STI Treatment-Seeking Behaviour . 229 12.13 Sexual Behaviour among Young Women and Men . 229 12.14 Orphanhood and Children’s Living Arrangements. 235 CHAPTER 13 HIV PREVALENCE AND ASSOCIATED FACTORS . 239 13.1 Coverage of HIV Testing . 240 13.2 HIV Prevalence. 243 13.2.1 HIV Prevalence by Socioeconomic Characteristics . 243 13.2.2 HIV Prevalence by Other Socio-demographic Characteristics. 246 13.2.3 HIV Prevalence by Sexual Risk Behaviour . 247 13.2.4 HIV Prevalence by Other Characteristics Related to HIV Risk. 249 13.2.5 HIV Prevalence and Male Circumcision . 250 13.2.6 Prevalence among Couples. 251 13.3 Distribution of the HIV Burden in Ghana. 253 REFERENCES . 255 APPENDIX A SAMPLE IMPLEMENTATION . 257 APPENDIX B ESTIMATES OF SAMPLING ERRORS . 263 APPENDIX C DATA QUALITY TABLES . 281 APPENDIX D PERSONS INVOLVED IN THE 2003 GHANA DEMOGRAPHIC AND HEALTH SURVEY . 287 APPENDIX E QUESTIONNAIRES . 291 Tables and Figures | ix TABLES AND FIGURES CHAPTER 1 INTRODUCTION Table 1.1 Basic demographic indicators.2 Table 1.2 Results of the household and individual interviews .9 CHAPTER 2 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS Table 2.1 Household population by age, sex, and residence .12 Table 2.2 Household composition.13 Table 2.3.1 Educational attainment of household population: women .15 Table 2.3.2 Educational attainment of household population: men .16 Table 2.4 School attendance ratios.18 Table 2.5 Household characteristics .20 Table 2.6 Household durable goods.22 Figure 2.1 Population Pyramid .12 Figure 2.2 Age-Specific Attendance Rates.19 CHAPTER 3 CHARACTERISTICS OF SURVEY RESPONDENTS Table 3.1 Background characteristics of respondents .24 Table 3.2.1 Educational attainment by background characteristics: women.25 Table 3.2.2 Educational attainment by background characteristics: men.26 Table 3.3.1 Literacy: women.28 Table 3.3.2 Literacy: men .29 Table 3.4.1 Exposure to mass media: women.30 Table 3.4.2 Exposure to mass media: men .31 Table 3.5.1 Employment status: women.33 Table 3.5.2 Employment status: men .34 Table 3.6.1 Occupation: women.36 Table 3.6.2 Occupation: men .37 Table 3.7.1 Type of employment: women.39 Table 3.7.2 Type of employment: men .39 Table 3.8 Decision on use of earnings and contribution of earnings to household expenditures.41 Table 3.9 Women's control over earnings .42 Table 3.10 Women's participation in decisionmaking .43 Table 3.11 Women's participation in decisionmaking by background characteristics.45 Table 3.12.1 Women's attitude toward wife-beating.47 Table 3.12.2 Men's attitude toward wife-beating .48 Table 3.13.1 Women's attitude toward wives refusing sex with husbands .50 Table 3.13.2 Men's attitude toward wives refusing sex with husbands.51 Table 3.14 Men's attitudes towards justifiable actions if wife refuses sex .52 x | Tables and Figures Figure 3.1 Occupation of Women Age 15-49 and Men Age 15-59.38 Figure 3.2 Type of Earnings of Employed Women Age 15-49 and Men Age 15-59 .40 Figure 3.3 Women’s Participation in Decisionmaking: Number of Decisions in Which Women Participate in the Final Say, Based on Five Household Decisions .44 CHAPTER 4 FERTILITY Table 4.1 Current fertility .54 Table 4.2 Fertility by background characteristics.55 Table 4.3 Trends in age-specific fertility rates .56 Table 4.4 Trends in fertility.57 Table 4.5 Children ever born and living.58 Table 4.6 Birth intervals.60 Table 4.7 Age at first birth .61 Table 4.8 Median age at first birth by background characteristics.62 Table 4.9 Teenage pregnancy and motherhood .63 Figure 4.1 Total Fertility Rates, Ghana and Selected Sub-Saharan Countries .54 Figure 4.2 Total Fertility Rate by Background Characteristics .56 Figure 4.3 Trends in Total Fertility Rate, Ghana 1988-2003.57 CHAPTER 5 FAMILY PLANNING Table 5.1.1 Knowledge of contraceptive methods: women .66 Table 5.1.2 Knowledge of contraceptive methods: men .67 Table 5.2.1 Ever use of contraception: women.69 Table 5.2.2 Ever use of male method of contraception: men.70 Table 5.3 Current use of contraception .71 Table 5.4 Current use of contraception by background characteristics .72 Table 5.5 Trends in the use of family planning.74 Table 5.6 Current use of contraception by women's status.76 Table 5.7 Number of children at first use of contraception .77 Table 5.8 Pill brand and cost .78 Table 5.9 Condom brand and cost .79 Table 5.10 Knowledge of fertile period.80 Table 5.11 Source of contraception.81 Table 5.12 Informed choice .83 Table 5.13 Future use of contraception .84 Table 5.14 Reason for not intending to use contraception .85 Table 5.15 Preferred method of contraception for future use .86 Table 5.16.1 Exposure to family planning messages: women .87 Table 5.16.2 Exposure to family planning messages: men.88 Table 5.17 Exposure to specific radio shows on family planning .90 Table 5.18 Contact of non-users with family planning providers .91 Table 5.19 Discussion of family planning with husband.92 Table 5.20 Men's attitudes towards contraception .93 Tables and Figures | xi Table 5.21.1 Approval of family planning: women .95 Table 5.21.2 Approval of family planning: men .96 Table 5.22 Wife's perception of husband's attitude toward family planning .97 Figure 5.1 Current Use of Family Planning Among Currently Married Women Age 15-49 .74 Figure 5.2 Trends in Current Use of Contraceptive Methods, Ghana, 1988-2003.75 Figure 5.3 Trends in Source of Modern Contraceptive Methods, Ghana 1988-2003.83 Figure 5.4 Percentage of Women and Men Exposed to Family Planning Messages in the Media.90 CHAPTER 6 OTHER PROXIMATE DETERMINANTS OF FERTILITY Table 6.1 Current marital status.99 Table 6.2 Polygyny. 101 Table 6.3 Age at first marriage . 103 Table 6.4 Median age at first marriage. 104 Table 6.5 Age at first sexual intercourse. 105 Table 6.6 Median age at first intercourse . 106 Table 6.7.1 Recent sexual activity: women. 108 Table 6.7.2 Recent sexual activity: men . 109 Table 6.8 Postpartum amenorrhea, abstinence and insusceptibility . 110 Table 6.9 Median duration of postpartum insusceptibility by background characteristics . 111 Table 6.10 Menopause . 112 Figure 6.1 Percentage of Married Men with Two or More Wives, by Region. 102 CHAPTER 7 FERTILITY PREFERENCES Table 7.1 Fertility preferences by number of living children. 114 Table 7.2 Desire to limit childbearing. 115 Table 7.3 Need for family planning . 117 Table 7.4 Ideal number of children . 119 Table 7.5 Mean ideal number of children by background characteristics . 120 Table 7.6 Fertility planning status . 121 Table 7.7 Wanted fertility rates . 122 Table 7.8 Ideal number of children and unmet need by women's status . 123 CHAPTER 8 INFANT AND CHILD MORTALITY Table 8.1 Early childhood mortality rates. 127 Table 8.2 Trends in early childhood mortality rates . 128 Table 8.3 Early childhood mortality rates by socio-economic characteristics . 129 Table 8.4 Early childhood mortality rates by demographic characteristics . 131 Table 8.5 Early childhood mortality rates by women's status . 132 Table 8.6 Perinatal mortality . 134 Table 8.7 High-risk fertility behaviour. 135 xii | Tables and Figures Figure 8.1 Trends in Infant and Under-five Mortality Rates, Ghana 1988-2003 . 128 Figure 8.2 Under-Five Mortality by Background Characteristics . 130 Figure 8.3 Under-Five Mortality by Socio-Economic Characteristics. 131 CHAPTER 9 MATERNAL AND CHILD HEALTH Table 9.1 Antenatal care . 138 Table 9.2 Number of antenatal care visits and timing of first visit . 140 Table 9.3 Components of antenatal care . 142 Table 9.4 Tetanus toxoid injections . 143 Table 9.5 Place of delivery . 145 Table 9.6 Assistance during delivery . 146 Table 9.7 Delivery characteristics . 147 Table 9.8 Postnatal care by background characteristics. 149 Table 9.9 Reproductive health care by women's status . 150 Table 9.10 Vaccinations by source of information . 152 Table 9.11 Vaccinations by background characteristics. 154 Table 9.12 Vaccinations in first year of life. 155 Table 9.13 Prevalence and treatment of symptoms of ARI 156 Table 9.14 Hand-washing materials in household. 158 Table 9.15 Disposal of children's stools. 159 Table 9.16 Prevalence of diarrhoea . 160 Table 9.17 Knowledge of ORS packets . 161 Table 9.18 Diarrhoea treatment . 162 Table 9.19 Feeding practices during diarrhoea . 163 Table 9.20 Children’s health care by women's status . 164 Table 9.21 Problems in accessing health care . 166 Table 9.22 Use of smoking tobacco. 167 Figure 9.1 Trends in Maternity Care Indicators, Ghana 1988-2003. 139 Figure 9.2 Number of Antenatal Care Visits . 140 Figure 9.3 Percentage of Children Age 12-23 Months with Specific Vaccinations. 152 Figure 9.4 Trends in Vaccination Coverage, Ghana 1988-2003 . 153 CHAPTER 10 NUTRITION Table 10.1 Initial breastfeeding . 170 Table 10.2 Breastfeeding status by age . 172 Table 10.3 Median duration and frequency of breastfeeding. 174 Table 10.4 Foods consumed by children in the day or night preceding the interview . 176 Table 10.5 Frequency of foods consumed by children in the day or night preceding the interview . 177 Table 10.6 Frequency of foods consumed by children in preceding seven days . 178 Table 10.7 Iodisation of household salt . 180 Table 10.8 Micronutrient intake among children . 181 Table 10.9 Micronutrient intake among mothers . 183 Table 10.10 Prevalence of anaemia in children . 185 Tables and Figures | xiii Table 10.11 Prevalence of anaemia in women . 186 Table 10.12 Prevalence of anaemia in children by anaemia status of mother. 187 Table 10.13 Nutritional status of children. 190 Table 10.14 Nutritional status of women by background characteristics. 193 Figure 10.1 Breastfeeding Practices by Age, Ghana 2003 . 173 Figure 10.2 Frequency of Meals Consumed by Children under 36 Months of Age Living with Their Mother, by Breastfeeding Status, Ghana 2003. 178 Figure 10.3 Stunting, Wasting, and Underweight by Age, Ghana . 191 CHAPTER 11 MALARIA Table 11.1 Ownership of mosquito nets. 196 Table 11.2 Use of mosquito nets by children. 197 Table 11.3 Use of mosquito nets by pregnant women . 198 Table 11.4. Exposure to messages on malaria . 199 Table 11.5 Use of Intermittent Preventive Treatment (IPT) by pregnant women. 201 Table 11.6 Use of Fansidar for Intermittent Preventive Treatment (IPT) . 202 Table 11.7 Prevalence and prompt treatment of fever . 203 Table 11.8 Type and timing of anti-malarial drugs . 205 CHAPTER 12 HIV/AIDS-RELATED KNOWLEDGE, ATTITUDES, AND BEHAVIOR Table 12.1 Knowledge of AIDS. 208 Table 12.2 Knowledge of HIV prevention methods . 209 Table 12.3.1 Beliefs about AIDS: women . 211 Table 12.3.2 Beliefs about AIDS: men. 212 Table 12.4.1 Accepting attitudes towards those living with HIV: women . 214 Table 12.4.2 Accepting attitudes towards those living with HIV: men . 215 Table 12.5.1 Knowledge of prevention of mother to child transmission of HIV: women. . 217 Table 12.5.2 Knowledge of prevention of mother to child transmission of HIV: men . 218 Table 12.6 Women and men who had an HIV test and received test results. 220 Table 12.7 Pregnant women counselled and tested for HIV. 221 Table 12.8 Attitudes towards negotiating safer sex with husband . 223 Table 12.9 Higher-risk sex and condom use at last higher-risk sex among women and men age 15-49 . 225 Table 12.10 Multiple sex partners among women and men. 226 Table 12.11 Paid sex in past year . 227 Table 12.12 Self-reportiing of sexually transmitted infection (STI) and STI symptoms . 228 Table 12.13 Women and men seeking treatment for STIs. 229 Table 12.14 Age at first sex among young women and men . 230 Table 12.15 Knowledge of a source for condoms among young women and men . 231 Table 12.16 Condom use at first sex among young women and men . 232 Table 12.17 Premarital sex and use of condom among young women and men. 233 Table 12.18 Higher-risk sex and condom use at last higher-risk sex . 234 Table 12.19 Age discontinuity in sexual relationships . 236 Table 12.20 Children's living arrangements and orphanhood. 237 xiv | Tables and Figures Figure 12.1 Reason for Getting HIV Test among Women and Men Age 15-49 Who Have Ever Been Tested. 222 Figure 12.2 Abstinence, Being Faithful, and Using Condoms Among Women and Men Age 15-24 . 235 CHAPTER 13 HIV PREVALENCE AND ASSOCIATED FACTORS Table 13.1 Coverage of HIV testing . 240 Table 13.2 Coverage of HIV testing, by background characteristics. 242 Table 13.3 HIV prevalence by age. 243 Table 13.4 HIV prevalence by background characteristics . 245 Table 13.5 HIV prevalence by selected socio-demographic characteristics. 246 Table 13.6 HIV prevalence by sexual behaviour characteristics . 248 Table 13.7 HIV prevalence by other indicators . 249 Table 13.8 HIV prevalence by prior HIV testing. 250 Table 13.9 HIV prevalence among men by circumcision status. 251 Table 13.10 HIV prevalence among couples . 252 Figure 13.1 HIV Prevalence by Age Group and Sex . 244 APPENDIX A SAMPLE IMPLEMENTATION Table A.1 Sample implementation: women . 257 Table A.2 Sample implementation: men. 258 Table A.3 Coverage of HIV testing among interviewed women by socio-demographic characteristics . 259 Table A.4 Coverage of HIV testing among interviewed men by socio-demographic characteristics . 260 Table A.5 Coverage of HIV testing by sexual behaviour characteristics: women . 261 Table A.6 Coverage of HIV testing by sexual behaviour characteristics: men. 262 APPENDIX B ESTIMATES OF SAMPLING ERRORS Table B.1 List of selected variables for sampling errors, Ghana 2003. 266 Table B.2 Sampling errors for total sample, Ghana 2003. 267 Table B.3 Sampling errors for urban sample, Ghana 2003 . 268 Table B.4 Sampling errors for rural sample, Ghana 2003 . 269 Table B.5 Sampling errors for Western sample, Ghana 2003 . 270 Table B.6 Sampling errors for Central sample, Ghana 2003 . 271 Table B.7 Sampling errors for Greater Accra sample, Ghana 2003 . 272 Table B.8 Sampling errors for Volta sample, Ghana 2003 . 273 Table B.9 Sampling errors for Eastern sample, Ghana 2003 . 274 Table B.10 Sampling errors for Ashanti sample, Ghana 2003 . 275 Table B.11 Sampling errors for Brong Ahafo sample, Ghana 2003 . 276 Table B.12 Sampling errors for Northern sample, Ghana 2003 . 277 Table B.13 Sampling errors for Upper East sample, Ghana 2003 . 278 Tables and Figures | xv Table B.14 Sampling errors for Upper West sample, Ghana 2003. 279 APPENDIX C DATA QUALITY TABLES Table C.1 Household age distribution. 281 Table C.2 Age distribution of eligible and interviewed women and men . 282 Table C.3 Completeness of reporting . 283 Table C.4 Births by calendar years . 284 Table C.5 Reporting of age at death in days . 285 Table C.6 Reporting of age at death in months . 286 Foreword | xvii FOREWORD The 2003 Ghana Demographic and Health Survey (GDHS) is a nationwide sample survey carried out to provide information on population, family planning, maternal and child health, nutrition, childhood mortality, and AIDS and sexually transmitted infections (STIs). This is the fourth survey of its kind to be undertaken in Ghana, others being in 1988, 1993, and 1998. This latest GDHS included, for the first time, testing of blood samples to provide national rates of anaemia and HIV. All four demographic and health surveys have been implemented by the Ghana Statistical Service, in close collaboration with other stake- holders. The Statistical Service of Ghana acknowledges the invaluable assistance given by a number of organisations and individuals both local and international towards the successful implementation of the 2003 GDHS. The Service is grateful to the Ministry of Health and the Ghana Health Service for releasing nurses for the survey fieldwork and personnel for training the interviewers. Our appreciation also goes to the Ghana AIDS Commission for their support at the time of training and in helping to set up VCT cen- tres as a follow-on to the HIV testing in the survey. We sincerely thank the Ghana Registered Midwives Association (GRMA) and the National Popu- lation Council (NPC) for providing vehicles for the data collection. We are very grateful to all members of the National Steering Committee, the Ethics Committee and the project personnel for their immense support and contribution during the different phases of the survey. We appreciate the work done by the Noguchi Memorial Institute for Medical Research on HIV testing, and especially commend the laboratory team assigned to work on the blood samples for their tire- less efforts in getting the testing done successfully. The Service is particularly thankful to the United States Agency for International Development (USAID) for funding the survey through its mission in Ghana, and to ORC Macro for providing technical assistance. We thank the authors of this report. They are mentioned specifically by name at the beginning of this report. We owe an immense gratitude to the field coordinators, interviewers, nurses, laboratory person- nel, supervisors, field editors, regional statisticians, and drivers for their hard work and dedication. We have printed the names of all survey personnel in Appendix D as a sign of our appreciation for their valu- able assistance. Most of all, we truly appreciate the co-operation of all survey respondents in making the 2003 GDHS a success. Dr. Grace Bediako, Government Statistician, Ghana Statistical Service Contributors | xix CONTRIBUTORS This report was authored by the following persons: Dr. Clement Ahiadeke, Institute for Social, Statistical and Economic Research Mrs. Faustina Ainguah, Ghana Statistical Service Dr. William Ampofo, Noguchi Memorial Institute for Medical Research Mr. Jacob Arthur-Quarm, Noguchi Memorial Institute for Medical Research Dr. Amanua Chinbuah, Ghana Health Service, Health Research Unit Dr. Agnes Dzokoto, National AIDS Control Programme Dr. Pav Govindasamy, ORC Macro Mr. Steve Grey, National Population Council Dr. Steven O. Kwankye, Regional Institute for Population Studies Mr. Emmanual Larbi, Ghana AIDS Commission Mrs. Edith Mote, Ghana Statistical Service Dr. Gloria Quansah-Asare, Ghana Health Service, Maternal and Child Health/Family Planning Unit Dr. Isabella Sagoe-Moses, Ghana Health Service, Nutrition Unit Mrs. Jasbir Sangha, ORC Macro Dr. Tesfay Teklu, Regional Institute for Population Studies Summary of Findings | xxi SUMMARY OF FINDINGS The 2003 Ghana Demographic and Health Survey (2003 GDHS) is a nationally representa- tive survey of 5,691 women age 15-49 and 5,015 men age 15-59 from 6,251 households covering 412 sample points (clusters) throughout Ghana. This survey is the fourth in a series of national- level population and health survey conducted as part of the global Demographic and Health Sur- veys (DHS) program and is designed to provide data to monitor the population and health situa- tion in Ghana as a follow-up of the 1988, 1993 and 1998 GDHS surveys. The survey utilised a two-stage sample based on the 2000 Population and Housing Census and was designed to pro- duce separate estimates for key indicators for each of the ten regions in Ghana. Data collection took place over a three-month period, from late July to late October 2003. The survey obtained detailed information on fertility levels, marriage, sexual activity, fer- tility preferences, awareness and use of family planning methods, breastfeeding practices, nutri- tional status of women and young children, childhood mortality, maternal and child health, awareness and behaviour regarding HIV/AIDS, and other sexually transmitted infections (STIs). In addition, the 2003 GDHS collected informa- tion on malaria and use of mosquito nets, and carried out anaemia testing in children and women and HIV testing in adults. The 2003 GDHS was implemented by the Ghana Statistical Service (GSS) in collaboration with the Noguchi Memorial Institute for Medical Research (NMIMR) and the Ghana Health Ser- vice. Technical assistance was provided by ORC Macro through the MEASURE DHS pro- gramme. Financial support for the survey was provided by the U.S. Agency for International Development (USAID) and the Government of Ghana. FERTILITY Fertility Levels and Trends. Comparison of the data from the 2003 GDHS with the three earlier DHS surveys indicates that the dramatic decline in fertility experienced in the eighties and nineties appears to have slowed down. The TFR, estimated for the three years preceding each survey, declined dramatically from 6.4 children per woman in 1988 to 5.2 children per woman in 1993, and to 4.4 children in 1998, a nearly 2-child drop in fertility over the decade. However, the demographic transition experi- enced in Ghana seems to have stalled in the last three years even though contraceptive use has continued to rise. Nevertheless, with a current TFR of 4.4, Ghana’s fertility rate is one of the lowest in sub-Saharan Africa. Fertility Differentials. Differentials by background characteristics are marked. Rural women have nearly twice as many children (5.6 children per woman) as urban women (3.1 chil- dren per woman). The total fertility rate is high- est in the Northern Region (7.0 children per woman) and lowest in Greater Accra (2.9 chil- dren per woman). As expected, women’s educa- tion is strongly associated with lower fertility, decreasing from 6.0 children per woman among those with no education to 2.5 children per woman among those with at least secondary education. Similar differentials are observed by wealth quintile, with TFR decreasing from 6.4 children per woman among women in the lowest wealth quintile to 2.8 children per woman among those in the highest wealth quintile. Unplanned Fertility. Despite a steady rise in the level of contraceptive use over the last fifteen years, the 2003 GDHS data indicate that unplanned pregnancies are common in Ghana. Overall, 16 percent of births in Ghana are un- wanted, while 24 percent are mistimed (wanted later). The proportion of unplanned births de- clined slightly from 42 percent in 1993 to 36 percent in 1998 but rose again to 40 percent in 2003. What is more troubling, however, is the fact that the proportion of births that are un- wanted has increased rather dramatically from the 1993 level of 9 percent to 16 percent in 2003. xxii | Summary of Findings Fertility Preferences. There is consider- able desire among currently married Ghanaians to control the timing and number of births. Thirty-eight percent of currently married women would like to wait for two years or more for the next birth, and 36 percent do not want to have another child. About a fifth (18 percent) would like to have a child soon (within two years). A comparison of the data over the four DHS sur- veys show that the desire to space births among currently married women has declined in the last 15 years, from 45 percent in 1988 to 38 percent in 2003. On the other hand, the desire to limit has increased from 23 percent in 1988 to 34 per- cent in 2003. However, this change has been minimal in the last ten years. There has been a decline in ideal family size among currently married women over time, from a mean of 5.5 children in 1988 to 4.8 chil- dren in 2003. There has been little change in the ideal number of children over the last 10 years. FAMILY PLANNING Knowledge of Contraception. Knowledge of family planning is nearly universal, with 98 percent of all women age 15 to 49 and 99 per- cent of all men age 15 to 59 knowing at least one modern method of family planning. Among all women, the most widely known methods of family planning are the male condom (95 per- cent), injectables (89 percent), the pill (88 per- cent) and female condom (83 percent). Seventy percent of all women have heard of female ster- ilisation, while 61-65 percent have heard of the IUD, implants, and periodic abstinence. There has been an increase in levels of awareness of contraceptive methods over time. Among all women, the proportions who know any method has risen since 1988 for all methods (from 76 percent in 1988, 91 percent in 1993, 93 percent in 1998 and 98 percent in 2003). The proportions who know of implants has risen steeply since 1993 (from 4 percent in 1993, 21 percent in 1998 and 62 percent in 2003). A simi- lar trend is seen among men with remarkable increases in knowledge of IUD, male sterilisa- tion and LAM. Use of Contraception. The contraceptive prevalence rate among married women is 25 percent. The most commonly used modern method among married women is the pill (6 per- cent), followed closely by injectables (5 per- cent). Male condoms and female sterilisation are used by 3 percent and 2 percent of married women, respectively, while implants and IUD are used by 1 percent each. The most commonly used traditional method is periodic abstinence, used by 5 percent of married women. Trends in Contraceptive Use. Current use of contraception by married women has in- creased from 13 percent in 1988, 20 percent in 1993, 22 percent in 1998 and 25 percent in 2003. There has been a steady increase in the use of modern methods from 5 percent in 1988 to 19 percent in 2003. However, while there was an increase in the use of traditional methods from 8 percent in 1988 to 10 percent in 1993, use of these methods have since decreased to 9 percent in 1998 and to 7 percent in 2003. Use of male condoms, pills, injectables and implants has in- creased. Differentials in Contraceptive Use. Women in urban areas are more likely to use contraceptive methods (31 percent) than their rural counterparts (21 percent). Male condoms, IUD, and female sterilisation use in urban areas is two to three times higher than in rural Ghana. The more urbanised regions such as Greater Ac- cra and Brong-Ahafo have contraceptive preva- lence rates above 30 percent. Two of the three northern regions (Upper East and Northern) re- port low levels of contraceptive use (12 percent each). Women with at least some secondary education are more than twice as likely to use contraception as women with no education. The proportion currently using contraception gener- ally increases with increasing number of chil- dren. Fourteen percent of women without chil- dren are currently using contraceptive methods, compared with 26 percent of women with five or more children. Wealth and current use of contra- ception is positively related, increasing from 14 percent among currently married women in the lowest quintile to 35 percent in the highest quin- tile. Summary of Findings | xxiii Source of Modern Methods. In Ghana, both the public and private sectors are important sources of supply for users of modern methods (41 percent and 54 percent, respectively). The most common public sector source are govern- ment hospitals and polyclinics, which provide most of the services (26 percent), while govern- ment health centres and family planning clinics provide 11 percent and 4 percent of users, re- spectively. In the last five years, there has been a shift in the source of modern contraceptive methods from the public to the private sector. The propor- tion of current users relying on private medical sources has increased from 45 percent in 1998 to 54 percent in 2003, while the reliance on public sources for all modern methods decreased from 47 percent in 1998 to 41 percent in 2003. Unmet Need for Family Planning. Thirty- four percent of married women have an unmet need for family planning. Unmet need for spac- ing is higher than unmet need for limiting chil- dren (22 percent and 12 percent, respectively), unchanged since 1998. Only 43 percent of the demand for family planning is currently being met, implying that the needs of more than one in two Ghanaian women are currently not being met. MATERNAL HEALTH Antenatal Care. A relatively high percent- age of women received antenatal care from a trained health professional (21 percent from a doctor and 71 percent from a nurse/midwife). One percent of mothers received antenatal care from a traditional birth attendant (TBA) and 6 percent received no antenatal care. A compari- son of the 2003 GDHS data with data from the three earlier DHS surveys show that there has been an improvement in the utilization of ante- natal services in the last fifteen years from 82 percent of mothers receiving care for their most recent birth in the five-year period preceding the survey in 1988, to 92 percent in 2003. Half of women received at least two doses of tetanus toxoid for their most recent birth in the five years preceding the survey, a third of women received only one tetanus toxoid injec- tion and 14 percent received none. The data show that there has been an improvement in tetanus toxoid coverage, for the most recent birth in the five years preceding the survey, over the last fifteen years, from 70 percent in 1988 to 83 percent in 2003. With regard to anti-malarial indicators, the data show that 10 percent of pregnant women slept under a net, 4 percent slept under an ever- treated net, and 3 percent slept under an insecti- cide treated net (ITN), the night before the inter- view with no difference in the use of nets be- tween pregnant and non-pregnant women. The data show that 58 percent of mothers reported that they received anti-malarial drugs for the prevention of malaria during pregnancy. It also shows that chloroquine is more frequently (12 percent) taken than SP/Fansidar (1 percent), pre- sumably because the old programme was still in force during the fielding of the survey. The 1 percent of women who used SP/Fansidar re- ceived the drug during an antenatal visit. Delivery Care. Nationally, 46 percent of births in the last five years are delivered in health facilities, with 36 percent in public health facilities and 9 percent in private health facili- ties. About half of births (53 percent) occur at home. The data also show that medically trained providers assisted with 47 percent of deliveries, TBAs assisted with 31 percent of deliveries and relatives or friends attended 19 percent of deliv- eries. Medically assisted deliveries continue to be low in Ghana, with less than fifty percent benefiting from professional delivery assistance over the last fifteen years. Postnatal Care. One in four women who had a non-institutional live birth in the five years preceding the survey received postnatal care within two days of delivery, one in ten women received postnatal care 3-6 days after delivery and one in eight received postnatal care 7-41 days after delivery. More than half of women who had a non-institutional birth in the five years preceding the survey did not receive post- natal care at all. xxiv | Summary of Findings CHILD HEALTH Childhood Mortality. Data from the 2003 GDHS show that there has been a slowing down in the mortality decline over the last five years. Data for the most recent five-year period sug- gests that one in every nine Ghanaian children dies before reaching age five. Nearly three in five of these deaths occur in the first year of life―infant mortality is 64 deaths per 1,000 live births and child mortality is 50 deaths per 1,000 children age one. Neonatal mortality is 43 deaths per 1,000 live births in the most recent five-year period, while postneonatal mortality is 21 deaths per 1,000 live births. Neonatal deaths account for two-thirds of the deaths in infancy. Childhood Vaccination Coverage. Sixty- nine percent of Ghanaian children age 12-23 months are fully immunised, while 5 percent received no vaccinations at all. Nine in ten chil- dren have received the BCG and first dose of DPT and polio vaccines. While the coverage for the first dose of DPT and polio is high, coverage declines for subsequent doses of DPT and polio, with only about 80 percent of children receiving the recommended three doses of these vaccines. Eighty-three percent of children received the measles vaccine and 77 percent have been vac- cinated against yellow fever. The percentage of children age 12-23 months who have been fully vaccinated has increased over the last twenty years, from 47 percent in 1988 to 69 percent in 2003. Child Illness and Treatment. Among chil- dren under five years of age, 10 percent were reported to have had symptoms of acute respira- tory illness in the two weeks preceding the sur- vey. Of these, 44 percent were taken to a health facility or provider for treatment. Fifteen percent of children under five years had diarrhoea in the two weeks preceding the survey. Twenty-six percent of children with diarrhoea were taken to a health provider. Just over a third of children with diarrhoea (39 percent) were given a solu- tion made from oral rehydration salts (ORS), 11 percent received recommended home fluids (RHF) and 40 percent were given increased flu- ids. Overall, 63 percent received ORS, RHF, or increased fluids. Twenty-one percent of children under five years had a fever in the two weeks preceding the survey. Of these, 63 percent took an anti- malarial drug. Forty-four percent of children took the anti-malarial drug on the same day or the next after the onset of the illness. Chloro- quine is by far the most common anti-malarial drug taken for fever (59 percent), followed by Amodiaquine and Quinine (2 percent each) and SP/Fansidar (less than 1 percent). NUTRITION Breastfeeding Practices. The data indicate that almost all (97 percent) Ghanaian children are breastfed for some period of time. Forty-six percent of infants were put to the breast within one hour of birth, and 75 percent started breast- feeding within the first day. The data from 2003 can be compared with similar data collected five years ago. The data show that over the last five years, there was little difference in the percent of children ever breastfed. The 2003 GDHS data indicate that supple- mentary feeding of children begins early. For example, among newborns less than two months of age, 38 percent are receiving supplementary foods or liquids other than water. The median duration of breastfeeding in Ghana is 23 months. Twelve percent of children under six months are given a feeding bottle with a nipple. Bottle-feeding reaches its peak (15 percent) at age 4-5 months. The percentage of young chil- dren bottle-fed has declined markedly over the last five years. Iodisation of household salt. Ninety per- cent of the households interviewed in the 2003 GDHS had their salt tested for iodine, while 9 percent had no salt available in the household. Fifty-nine percent of households are consuming salt that is not iodised, 13 percent of households are consuming inadequately iodised salt (<15 ppm) and only 28 percent are consuming adequately iodised salt (15+ ppm). Intake of Vitamin A. Ensuring that chil- dren between six months and 59 months receive enough vitamin A may be the single most effec- Summary of Findings | xxv tive child survival intervention, since deficien- cies in this micronutrient can cause blindness and can increase the severity of infections, such as measles and diarrhoea. Seventy-eight percent of children 6-59 months are reported to have received a vitamin A supplement in the 6 months preceding the survey. Forty-one percent of children under three who live with their mothers consume fruits and vegetables rich in vitamin A. Forty-three percent of mothers with a birth in the last five years reported receiving a vitamin A dose postpartum. Eight percent of interviewed women reported night blindness during preg- nancy. When adjusted for blindness not attrib- uted to vitamin A deficiency during pregnancy, the data show only two percent of women re- ported night blindness during their last preg- nancy. Prevalence of anaemia. Iron-deficiency anaemia is a major threat to maternal health and child health. Overall, more than three-quarters of Ghanaian children 6-59 months old have some level of anaemia, including 23 percent of children who are mildly anaemic, 47 percent who are moderately anaemic and 6 percent who are severely anaemic. The prevalence of anaemia is less pro- nounced among women than among children. Forty-five percent of Ghanaian women age 15- 49 are anaemic, with 35 percent mildly anaemic, 9 percent moderately anaemic, and less than 1 percent severely anaemic. Nutritional Status of Children. Accord- ing to the 2003 GDHS, 30 percent of children under five are stunted and 11 percent severely stunted. Seven percent of children under five are wasted and 1 percent severely wasted. Weight-for-age results show that 22 percent of children under five are underweight, with 5 per- cent severely underweight. Children whose bio- logical mothers were not in the household are more likely to be malnourished (34 percent stunted, and 25 percent underweight) than chil- dren whose mothers were interviewed. The proportion of children under five who are stunted has increased from 26 percent in 1998 to 30 percent in 2003. The proportion un- derweight decreased from 10 percent in 1998 to 7 percent in 2003. The proportion of children who are wasted also decreased from 25 percent in 1998 to 22 percent in 2003. Nutritional Status of Women. The mean height of Ghanaian women is 159 centimetres, which is above the critical height of 145 centi- metres. Only 1 percent are below 145 centime- tres. Nine percent of women were found to be chronically malnourished (BMI less than 18.5), while 25 percent are overweight or obese. There has been little change in the percentage of moth- ers whose height is below 145 centimetres and in the mean BMI over the last ten years from 1993 to 2003. HIV/AIDS Awareness of AIDS. Almost all (98 per- cent) women and men (99 percent) have heard of AIDS indicating that awareness of AIDS in Ghana is universal. Thirty-seven percent of women and 38 percent of men age 15-49, know someone personally who has the virus that causes AIDS or who has died of AIDS. Seventy- three percent of women and 82 percent of men know that condom use is a major prevention method. Eighty-six percent and 90 percent of women and men, respectively, know that limit- ing sex to only one uninfected partner is vital to the prevention of HIV. Sixty-nine percent of women and 78 percent of men know that these two preventive measures in combination can reduce the risk of HIV infection. In addition, 79 percent of women and 83 percent of men know that abstinence can prevent HIV infection. About four in five women and men cor- rectly know that a healthy looking person can have the AIDS virus. Fifty-five percent of women and sixty percent of men know that AIDS cannot be transmitted through mosquito bites. Less than half of women and three-fifths of men know that AIDS cannot be transmitted by supernatural means. More than 70 percent of women and men know that a person cannot be- xxvi | Summary of Findings come infected with HIV/AIDS by sharing food with someone who has AIDS. General knowledge on HIV transmission during pregnancy, delivery and breastfeeding is relatively high and ranges between 69 and 75 percent among women and 74 to 82 percent among men. However, few women and men (16 percent each) know that the risk of MTCT can be reduced if a mother takes special drugs dur- ing her pregnancy. Attitudes Towards People Living with HIV/AIDS. It is encouraging to see that more than two-thirds of women and men age 15-49 are willing to care for a family member with HIV in their own household, and that three-fifths of women and two-thirds of men do not believe that the HIV positive status of a family member should be kept a secret. Two-fifths of women and half of men also believe that an HIV posi- tive female teacher should be allowed to con- tinue teaching. However, only one in four women and one in three men say that they would buy fresh vegetables from a vendor with AIDS. HIV-Related Behavioural Indicators. One of the strategies for reducing the risk of contracting an STI is for young persons to delay the age at which they become sexually active. Seven percent of women and 4 percent of men had sex by exact age 15. Forty-six percent of women and 27 percent of men first had sex by exact age 18. Sexual intercourse with a non-marital or non-cohabiting partner is associated with an in- creased risk of contracting sexually transmitted diseases. One in five women and two in five men age 15-49 reported engaging in higher-risk sexual behaviour. Even more disturbing is the fact that half of women age 15-24 and more than four-fifths of men in the same age cohort engage in risky sexual behaviour. Sexual intercourse with more than one partner is also associated with a high risk of ex- posure to sexually transmitted diseases. One percent of women and 10 percent of men age 15- 49 report having had sexual intercourse with more than one partner in the twelve months prior to the survey. Promoting the use of condoms is an impor- tant strategy in the fight against HIV/AIDS transmission. Overall, only 28 percent of women and 45 percent of men age 15-49 used a condom during their last episode of higher-risk sex. HIV Prevalence. HIV tests were conducted for 89 percent of the 5,949 eligible women and 80 percent of the 5,345 eligible men. Results from the 2003 GDHS indicate that 2 percent of Ghanaian adults are HIV positive. HIV preva- lence in women age 15-49 is nearly 3 percent, while for men 15-59, it is under 2 percent. This female-to-male ratio of 1.8 to 1 is higher than that found in most population-based studies in Africa and implies that young women are par- ticularly vulnerable to HIV infection compared with young men. Prevalence among females is consistently higher than among males at all age groups except at ages 40-44, where male preva- lence is higher. The female-male gap is particu- larly large among women and men age 25-29, where women are nearly three and a half times as likely to be HIV positive as men. The peak prevalence among women is at age 35-39 (5 percent), while prevalence rises gradually with age among men to peak at age 40-44 (4 percent). Patterns of HIV Prevalence. Urban resi- dents have only a slightly higher risk of being HIV positive than rural residents with the urban- rural difference among women slightly higher than among men. Overall prevalence is highest in the Eastern Region (4 percent), followed by the Western and Brong Ahafo regions (3 percent each). Prevalence is lowest in the Northern, Cen- tral and Volta regions (1 percent each). Gender differences are apparent in all the regions. Those who have completed primary and mid- dle/JSS education are more likely to be HIV positive than those with either no education or at least secondary education. Work status is related to HIV prevalence among both women and men, with prevalence twice as high among those cur- rently working than those not currently working. Prevalence is highest among both women and men in the middle wealth quintile. Summary of Findings | xxvii Prevalence is significantly higher among widowed women (7 percent), followed closely by divorced or separated women (6 percent). Among men, prevalence is markedly higher among divorced or separated men (6 percent). Results from the 2003 GDHS indicate that, for the vast majority (96 percent) of cohabiting couples, both partners are HIV negative, while only in 1 percent of couples, are both partners HIV positive. There is discordance in the HIV positive status in under 2 percent of couples, where one partner is infected and the other is not. xxviii | Map of Ghana GHANA Introduction | 1 INTRODUCTION 1 1.1 GEOGRAPHY, HISTORY, AND ECONOMY 1.1.1 Geography The Republic of Ghana is centrally located in West Africa and has a total land area of 238,537 square kilometres. It is bordered by French-speaking countries, on the east by the Republic of Togo, on the north and northwest by Burkina Faso, and on the west by Côte d’Ivoire. The Gulf of Guinea lies to the south and stretches across the 560 kilometres of the country’s coastline. Ghana is a lowland country, except for a range of hills that lie on the eastern border and Mt. Afadjato, the highest point of about 884 metres above sea level, which is to the west of the Volta River. Ghana can be divided into three distinguishable ecological zones: the sandy coastline backed by a coastal plain that is crossed by several rivers and streams; the middle belt and western parts of the country, heavily forested with many streams and rivers; and an undulating savannah to the north that is drained mainly by the Black and White Volta Rivers. The Volta Lake, created as a result of a hydroelectric dam in the east, is one of the largest artificial lakes in the world. The climate of Ghana is tropical, but temperatures and rainfall vary by distance from the coast and elevation. The average annual temperature is about 26ºC (79º F). There are two distinct rainy seasons, April to June and September to November. In the north, however, the rainy season begins in March and lasts until September. Annual rainfall ranges from about 1,015 millimetres (40 inches) in the north to about 2,030 millimetres (80 inches) in the southwest. The harmattan, a dry desert wind, blows from the northeast between December and March, lowering the humidity and creating very warm days and cool nights in the north. In the south, the effects of the harmattan are felt mainly in January. 1.1.2 History Ghana gained its independence from British rule on 6 March 1957, and on 1 July 1960 became a sovereign state in the British Commonwealth of Nations. The administrative and political capital of the country is Accra, with a population of 1.7 million (GSS, 2002). Ghana is a constitutional democracy and currently operates a multi-party democratic presidential system of government following the promulgation of the 1992 fourth Republic Constitution of Ghana. The country has an Executive Presidency elected for four years with a maximum of two terms. There is a parliament elected every four years, an independent judiciary, and a vibrant media. The population is made up of several ethnic groups. The Akans constitute the largest ethnic group (49 percent) followed by the Mole-Dagbon (17 percent), Ewe (13 percent), and Ga/Dangme (8 percent). Various smaller ethnic groups can also be found in many parts of the country (GSS, 2002). Ghana is divided into 10 administrative regions, Western, Central, Greater Accra, Volta, Eastern, Ashanti, Brong Ahafo, Northern, Upper East, and Upper West. The regions are further divided into 138 districts to ensure efficient and effective administration at the local levels. 2 | Introduction 1.1.3 Economy The structure of the economy has not changed much over the past two decades. Agriculture, mining, logging, and retail trade are still the most important areas of economic activity. Agriculture is the main sector and employs about 50 percent of the population (GSS, 2002). High proportions of the working population in Ghana are concentrated in the informal sector, made up largely of self-employed persons. The leading exports of the country are cocoa, gold, and timber. In recent times, the economy has diversified and includes exports of non-traditional commodities such as pineapples, bananas, yams, and cashew nuts. Tourism is fast gaining prominence as a foreign exchange earner. The overriding objective of the Government of Ghana’s (GoG) economic development programme is poverty reduction and general improvement in the welfare of all Ghanaians. In 1995, the GoG developed the Vision 2020 strategy for poverty reduction, which emphasises economic growth, integrated rural development, expansion of employment opportunities, and improved access, especially by the rural and urban poor, to basic public services such as education, health care, water and sanitation, and family planning services (World Bank, 2003). Under this strategy, it is envisaged that national income will grow by at least 8 percent from the current 4-5 percent. 1.2 DEMOGRAPHIC PROFILE Ghana has undertaken four censuses since independence in 1957. The first was conducted in 1960, recording a population of 6.7 million. The 1970 Census reported Ghana’s population as 8.6 million with an intercensal growth rate of 2.4 percent. The 1984 and 2000 censuses put the population at 12.3 million and 18.9 million, respectively, with an average growth rate of 2.7 percent between the two census periods (Table 1.1). The population density per square kilometre has more than doubled from 36 persons in 1970 to 79 persons in 2000. The proportion urban increased significantly from 29 percent in 1970 to 44 percent in 2000. The sex ratio over the last 30 years has fallen slightly from 98.5 males per 100 females in 1970 to 97.9 in 2000. The proportion of the population under 15 years however, has decreased from 47 percent in 1970 to 41 percent in 2000, while the proportion 65 years and older increased from less than 4 percent to a little more than 5 percent over the same period. The changes observed in the age structure may be attributed to Table 1.1 Basic demographic indicators Selected demographic indicators for Ghana, 1970, 1984, 2000 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Indicator 1970 1984 2000 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Population (millions) 8.6 12.3 18.9 Intercensal growth rate (percent) 2.4 2.6 2.7 Density (pop./km2) 36.0 52.0 79.3 Percent urban 28.9 32.0 43.8 Sex ratio 98.5 97.3 97.9 Proportion age 0-14 years 46.9 45.0 41.3 Proportion age 65+ 3.6 4.0 5.3 Life expectancy (years) Male u 50.3 55.4 Female u 53.8 59.6 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– u=Unknown (Not available) Source: GSO, 1979; GSS, 1985; GSS, 2002 Introduction | 3 declining fertility and improvements in the health conditions of the people. Life expectancy at birth has increased from 50 years among males in 1984 to 55 years in 2000 and among females from 54 years to 60 years over the same period. 1.3 POPULATION POLICY AND REPRODUCTIVE HEALTH PROGRAMMES Ghana’s population policy was formulated and adopted in 1969 in recognition of the high population growth and fertility rates observed at the time. After 25 years of implementation, however, the 1969 population policy made only a modest impact. It was therefore revised in 1994 to take into account emerging issues such as HIV/AIDS, population and the environment, concerns about the elderly and children, and also to develop new strategies that would ensure the achievement of the revised policy objectives. This meant ensuring the systematic integration of population issues in all areas of development planning. Major targets aimed at achieving these objectives include the following: the reduction of the total fertility rate from 5.5 in 1993 to 5.0 by the year 2000; the achievement of a contraceptive prevalence rate of 15 percent for modern methods by the year 2000, and 50 percent by the year 2020; and the reduction in the annual population growth rate from about 3 percent per annum to 2 percent by the year 2020 (World Bank, 2003). The attainment of these policy goals is recognised as integral components of the national strategy to accelerate the pace of economic development, eradicate poverty, and enhance the quality of life of all citizens as outlined in the Vision 2020 Plan of Action. It is expected that these goals would propel Ghana into middle- income country status by the year 2020. The National Population Council and its Secretariat were established in 1992 as the highest statutory body to advise the government on population related issues as well as to facilitate, monitor, coordinate, and evaluate the implementation of population programmes of other organisations both public and private within the country. Ghana, in collaboration with the United Nations Fund for Population Activities (UNFPA), the United States Agency for International Development (USAID), the World Bank, and other development partners, has implemented several projects aimed at reducing reproductive health problems among the population. The support from these agencies is geared towards strategies on policy coordination and implementation as well as service delivery. The government is committed to improving access and equity of access to essential health care, and ensuring that the health sector plays a key role in the Ghana Poverty Reduction Strategy (GPRS). The priority health intervention areas identified include addressing the problems of HIV/AIDS and other sexually transmitted infections (STIs), malaria, tuberculosis, guinea worm disease, poliomyelitis, reproductive health, maternal and child health, accidents and emergencies, non-communicable diseases, oral health and eye care, and specialised services. The spread of HIV/AIDS is currently receiving considerable attention from the government and its development partners, with the immediate challenges including ensuring implementation of the Ghana HIV/AIDS Strategic Framework: 2001-2005 (World Bank, 2003). The objectives of the framework include reducing new HIV infections among the 15-49 age group and other vulnerable groups, and especially among the youth by the year 2005, improving service delivery and mitigating the impact of HIV/AIDS on individuals, the family, and the community by the year 2005, reducing individual and societal vulnerability and susceptibility to HIV/AIDS through the creation of an enabling environment for the implementation of the national response, and establishing a well managed multi-sectoral and multi-disciplinary institutional framework for the coordination and implementation of HIV/AIDS programmes in the country. 4 | Introduction The government recognises that a critical constraint to poverty reduction is the limited choice of contraceptives especially to poor families. A two-pronged strategy will be employed to address this issue. The first is to decentralise service delivery and expand the sale of contraceptives through community agents, including maternity homes and field workers. The second, a national campaign on fertility regulation, will be instituted and will include a comprehensive, systematic, and culturally sensitive information, education, and communication programme to promote the use of family planning (World Bank, 2003). In addition, the GPRS emphasises cost-effective interventions on immunisations and supervised delivery; a high impact and rapid delivery programme to reduce under-five and maternal mortality and malnutrition, especially in the three northern and central regions of the country; prevention and effective treatment of malaria and the availability and use of insecticide-treated bed nets; and the eradication of guinea worm (World Bank, 2003). 1.4 OBJECTIVES AND ORGANISATION OF THE SURVEY The principal objective of the 2003 Ghana Demographic and Health Survey (GDHS) is to provide data to monitor the population and health situation in the country. This is the fourth round in a series of national-level population and health surveys conducted in Ghana under the worldwide Demographic and Health Surveys Program. The primary objective is to provide current and reliable data on fertility and family planning behaviour, infant and child mortality, breastfeeding, antenatal care, children’s immunisations and childhood diseases, nutritional status of mothers and children, use of maternal and child health services, and awareness and behaviour regarding AIDS and other STIs. New features of the 2003 GDHS include the collection of information on female and male circumcision, information on malaria and ownership and use of insecticide-treated bed nets, and haemoglobin and HIV testing. The long-term objective of the survey includes strengthening the technical capacity of major government institutions, including the Ghana Statistical Service (GSS). The 2003 GDHS also provides comparable data for long-term trend analyses in Ghana, since the surveys were implemented by the same organisation, using similar data collection procedures. It also contributes to the ever-growing international database on demographic and health-related information. The 2003 GDHS was conducted by the Ghana Statistical Service in collaboration with the Noguchi Memorial Institute for Medical Research (NMIMR) and the Ghana Health Service. ORC Macro provided technical support for the survey through the MEASURE DHS+ programme. Funding for the survey came from the U.S. Agency for International Development (USAID), through its office in Ghana, and the Government of Ghana. 1.5 SAMPLE DESIGN The sample for the 2003 GDHS covered the population residing in private households in the country. A representative probability sample of about 6,600 households was selected nationwide. The list of enumeration areas (EAs) from the 2000 Ghana Population and Housing Census was used as a frame for the sample. The frame was first stratified into the 10 administrative regions in the country, then into rural and urban EAs. The sample was selected in such a manner as to allow for separate estimates for key indicators for the country as a whole, for each of the 10 regions in Ghana, as well as for urban and rural areas separately. The 2003 GDHS used a two-stage stratified sample design. At the first stage of sampling, 412 sample points or EAs were selected, each with probability proportional to size, based on the number of households. A complete household listing exercise was carried out between May and June 2003 within all the selected EAs (clusters). The second stage of selection involved systematic sampling of households from this list. The sample selected per EA varied by region depending on the population size. Fifteen households per EA were Introduction | 5 selected in all the regions except in Brong Ahafo, Upper East, and Upper West regions, where 20 households per EA were selected, and in the Northern region, where 16 households per EA were selected. The objective of this exercise was to ensure adequate numbers of complete interviews to provide estimates for important population characteristics with acceptable statistical precision. Due to the disproportional number of EAs and different sample sizes selected per EA among regions, the household sample for the 2003 GDHS is not self- weighted at the national level. The sample design is discussed in detail in Appendix A, and the methodology used in estimating sampling errors together with a list of sampling errors for key variables are provided in Appendix B. 1.6 QUESTIONNAIRES All women age 15-49 and all men age 15-59 who were either usual residents of the households in the GDHS sample or visitors present in the household the night before the survey were eligible to be interviewed in the survey. Three questionnaires were used for the 2003 GDHS: the Household Questionnaire, the Women’s Questionnaire, and the Men’s Questionnaire. The contents of these questionnaires were based on the model questionnaires developed by the MEASURE DHS+ programme and were designed to provide information needed by health and family planning programme managers and policymakers. The questionnaires were adapted to the Ghanaian situation and a number of questions pertaining to ongoing health, HIV, and family planning programmes were added. These questionnaires were translated from English into the five major languages (Akan, Nzema, Ewe, Ga, and Dagbani). The questionnaires are attached in Appendix E. The Household Questionnaire was used to list all the usual members and visitors in the selected households. Information was collected on the characteristics of each person listed, including the age, sex, education, and relationship to the head of household. The main purpose of the Household Questionnaire was to identify eligible women and men for the individual interview. The Household Questionnaire collected information on characteristics of the household’s dwelling unit, such as the source of drinking water, type of toilet facilities, flooring materials, ownership of various consumer goods, and ownership and use of mosquito nets. It was also used to record height and weight measurements of women 15-49 and children under the age of 5, and to record the respondents’ consent to the haemoglobin and HIV testing. The Women’s Questionnaire was used to collect information from all women age 15-49. These women were asked questions on the following topics: respondent’s background characteristics, such as education, residential history, media exposure, knowledge and use of family planning methods, fertility preferences, antenatal and delivery care, breastfeeding and infant and child feeding practices, vaccinations and childhood illnesses, childhood mortality, marriage and sexual activity, woman’s work and husband’s background characteristics, and awareness and behaviour regarding AIDS and other STIs. The Men’s Questionnaire was administered to all men age 15-59 in every household in the GDHS sample. The Men’s Questionnaire collected much of the same information found in the Women’s Questionnaire, but was shorter because it did not contain a reproductive history or questions on maternal and child health and nutrition. 1.7 HAEMOGLOBIN AND HIV TESTING In all households selected for the 2003 GDHS, women age 15-49 and children under age 5 were tested for anaemia. In addition, all eligible women and men were tested for HIV. Anaemia and HIV testing were only carried out if consent was provided by the respondents and in the case of a minor, by the parent or guardian. The protocol for haemoglobin and HIV testing was approved by the Ghana Health Service Ethical Review Committee in Accra and the ORC Macro Institutional Review Board in Calverton, Maryland, USA. 6 | Introduction 1.7.1 Haemoglobin Testing Haemoglobin testing is the primary method of anaemia diagnosis. In the GDHS, testing was done using the HemoCue system. A consent statement was read to the eligible woman and to the parent or responsible adult for young children and women age 15-17. This statement explained the purpose of the test, informed prospective subjects tested and/or their caretakers that the results would be made available as soon as the test was completed, and also requested permission for the test to be carried out, as well as the consent to report their names to health personnel in the local health facility if their haemoglobin level was low (severe). Before the blood was taken, the finger was wiped with an alcohol prep swab and allowed to air-dry. Then the palm side of the end of a finger (in case of adults and children six months of age and older) was punctured with a sterile, non-reusable, self-retractable lancet and a drop of blood collected on a HemoCue microcuvette and placed in a HemoCue photometer, which displays the result. For children under six months of age (or for children under one year of age who were particularly undernourished and bony) a heel puncture was made to draw a drop of blood. The results were recorded in the Household Questionnaire, as well as on a brochure, which was given to each woman, parent, or responsible adult, that explained what the results meant. For each person whose haemoglobin level was low (severe), and who agreed to have the condition reported, a referral was given to the respondent to be taken to a health facility. 1.7.2 HIV/AIDS Testing All eligible women and men who were interviewed were asked to voluntarily provide a few drops of blood for HIV testing. The protocol for the blood specimen collection and analysis was based on the anonymous linked protocol developed for DHS. The protocol allows for the merging of the HIV results to the socio-demographic data collected in the individual questionnaires, provided that information that could potentially identify an individual is destroyed before the linking takes place. This required that identification codes be deleted from the data file and that the back page of the Household Questionnaires that contain the bar code labels and names of respondents be destroyed prior to merging the HIV results with the individual data file. If, after explaining the procedure, the confidentiality of the data, and the fact that the test results would not be made available to the subject, a respondent consented to the HIV testing, a dried blood spot (DBS) specimen was obtained from a finger prick. Each respondent who consented to being tested for HIV was given an information brochure on AIDS, a list of fixed sites providing voluntary counseling and testing (VCT) services throughout the country, and a voucher to access free VCT services at any of these sites for the respondent and/or the partner. Each DBS sample was given a bar code label, with a duplicate label attached to the Household Questionnaire on the line showing consent for that respondent. A third copy of the same bar code label was affixed to a Blood Sample Transmittal Form in order to track the blood samples from the field to the laboratory. Filter papers were dried overnight in a plastic drying box, after which the nurse packed them in individual Ziploc bags for that particular sample point. Blood samples were periodically collected in the field along with the completed questionnaires and transported to the GSS headquarters in Accra for logging in, after which they were taken to the Noguchi Memorial Institute for Medical Research (NMIMR) at Legon, for HIV testing. In preparation for carrying out the HIV testing, a consultant from the Kenya Medical Research Institute was contracted by ORC Macro to spend a couple of weeks at NMIMR to assess their equipment and staff capacity. In addition, an ORC Macro official worked with laboratory scientists at NMIMR to conduct a validation study and set up the dried blood spot methodology to test for HIV using two Enzyme-Linked Immunosorbent Assay (ELISA) tests from different manufacturers that would also allow for sero-typing. Introduction | 7 Several meetings with ORC Macro staff, NMIMR staff, and staff of GSS, were then held to discuss the monitoring of sample collection in the field, the collection of samples from the field, and the delivery of the samples to the laboratory, with built-in checks to verify the samples collected and delivered. It was also emphasized at the meeting that the period between the collecting of blood samples in the field and the time of refrigeration should not exceed 14 days. The DBS filter paper samples with bar codes were received by NMIMR. Upon receipt, the samples were counted and checked against the transmittal sheet to verify the bar code identifications and kept in a cold room at 4 degrees centigrade until testing was started in September. Samples were taken out of the cold room and kept for at least 30 minutes at room temperature before testing. One-quarter-inch disks were punched from the dried blot spots and were submerged in phosphate buffered saline and Tween 20 for overnight elution at 4 degrees centigrade. The following day, serum was eluted and appropriate dilutions were made according to the testing protocol for the test kits used in the GDHS. These dilutions were determined following the validation study on the same test kits for both the DBS and venous blood samples. Eluted serum was tested following the manufacturer’s recommendations for each of the test kits used in the GDHS. All specimens were tested with a screening test, Vironostika HIV Uni-Form Plus O (ELISA I). All samples positive on the first screening test as well as 10 percent of the negatives were further tested in parallel with Wellcozyme HIV-1 Recombinant and Murex HIV-2 (ELISA II) for serotyping. Results for all the ELISAs were obtained by relating the absorbance value or optical density (OD) of a specimen to the OD of the serum controls. According to the testing algorithm, samples positive on the first ELISA and positive on both the second ELISAs were regarded as postivie for HIV-1 and HIV-2; samples positive on the first ELISA and positive on Wellcozyme HIV-1 Recombinant and negative on Murex HIV-2 were categorized as positive for HIV-1; similarly, samples that were positive on the first ELISA and negative on Wellcozyme HIV-1 Recombinant and positive on Murex HIV-2 were categorized as positive for HIV-2. Samples negative on the first ELISA and negative on ELISAs for serotyping were regarded as negative. Samples that had discordant results on ELISA I and ELISA II were tested again with ELISA I and ELISA II. The results were obtained and interpreted in the same manner as indicated above for the repeat ELISA testing. Discordant samples from the repeat ELISAs, were tested with a confirmatory test, PEPTI-LAV 1-2. In addition, all samples that tested positive on ELISAs and samples whose repeat ELISA results were discordant were also tested with PEPTI-LAV 1-2. Samples with “grey zone” or discordant results on the two assays (i.e., repeat ELISA’s and PEPTI-LAV 1-2) were tested by immunoblotting (Western Blot) with NEW LAV-BLOT I and NEW LAV-BLOT II using appropriate interpretative criteria based on the test kit. There were some indeterminate samples from the first round of testing that were also included for immunoblotting. The result on immunoblotting (Western Blot) was regarded as the final result. 1.8 PRETEST, TRAINING, AND FIELDWORK 1.8.1 Pretest A pretest of the Household, Women’s, and Men’s questionnaires used in the GDHS was conducted in May 2003 in English and five major local languages. The pretest training was conducted by GSS staff for two weeks from 5-17 May 2003. In addition, nurses recruited from the Ghana Health Service were trained in testing for haemoglobin and collecting blood samples for HIV/AIDS. Five teams were formed to conduct the pretest. Each team consisted of a supervisor, four interviewers, and a nurse. Urban and rural areas were chosen for the pretest to get a better overall sense of the response level and acceptance of HIV/AIDS testing. The lessons learned from the pretest were used to finalize the survey instruments and logistical arrangements. 8 | Introduction 1.8.2 Training and Fieldwork A total of 102 interviewers, 23 nurses, and 12 data entry operators participated in the main survey training that took place from 6-27 July 2003. All participants were trained in interviewing techniques and the contents of the GDHS questionnaires. The training was conducted following the standard DHS training procedures, including class presentations, mock interviews, and tests using the Household, Women’s, and Men’s Questionnaires. All interviewers were trained in taking height and weight measurements. In addition to interviewer training, 23 persons (most of whom were nurses from the Ghana Health Service) were trained for a period of 10 days in anaemia testing, collection of blood samples for the HIV testing, and in informed consent procedures. An additional 20 interviewers were also trained in blood collecting techniques. In addition to in-class practice, the nurses were taken to the local health clinic to practice blood-collecting techniques on women, men, and children. Interviewers and nurses were selected based on their in-class participation, performance in the field practices, fluency in the Ghanaian languages, and assessment tests. The most experienced trainees, those who took part in the pretest, and those who did extremely well, were selected to be supervisors and editors. Trainees selected as supervisors and field editors were given an additional two-days training on how to supervise fieldwork and edit questionnaires. In addition, there was one standby supervisor and nine interviewers ready for relief assignment whenever necessary. Ten regional statisticians acted as regional coordinators, and GSS staff coordinated and supervised fieldwork activities. Fifteen teams were constituted for data collection. Each team was made up of a supervisor, an editor, a nurse, four interviewers, and a driver. Fieldwork lasted for three months from late July to late October. ORC Macro provided technical assistance on all aspects of the survey; staff from ORC Macro participated in field supervision of interviews, height and weight measurements, and blood sample collection. 1.9 DATA PROCESSING The processing of the GDHS results began shortly after the fieldwork commenced. Completed questionnaires were returned periodically from the field to the GSS headquarters in Accra, where they were entered and edited by data processing personnel who were specially trained for this task. Twelve data entry operators from GSS were trained for one week on data entry procedures using CSPro. All data were entered twice (100 percent verification). In addition, tables were run periodically to monitor the quality of the data collected. The concurrent processing of the data was an advantage for data quality because field coordinators were able to advise teams of problems detected during the data entry. The data entry and editing phase of the survey was completed in mid-December 2003. 1.10 RESPONSE RATES Table 1.2 shows response rates for the 2003 GDHS. Response rates are important because high non- response may affect the reliability of the results. A total of 6,628 households were selected in the sample, of which 6,333 were occupied at the time of fieldwork. The difference between selected and occupied households is largely due to structures being vacant or destroyed. Successful interviews were conducted in 6,251 households, yielding a response rate of 99 percent. Introduction | 9 In the households interviewed in the survey, a total of 5,949 eligible women age 15-49 were identified; interviews were completed with 5,691 of these women, yielding a response rate of 96 percent. In the same households, a total of 5,345 eligible men age 15-59 were identified and interviews were completed with 5,015 of these men, yielding a male response rate of 94 percent. The response rates are slightly lower for the urban than rural sample, and among men than women. The principal reason for non-response among both eligible women and men was the failure to find individuals at home despite repeated visits to the household. The lower response rate for men reflects the more frequent and longer absences of men from the household, principally related to their employment and life style. Response rates for the HIV testing component were lower than those for the interviews. Details of the HIV testing response rates are discussed in Chapter 13. Table 1.2 Results of the household and individual interviews Number of households, number of interviews, and response rates, according to residence, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence ––––––––––––––––– Result Urban Rural Total ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Household interviews Households selected 2,720 3,908 6,628 Households occupied 2,571 3,762 6,333 Households interviewed 2,517 3,734 6,251 Household response rate 97.9 99.3 98.7 Interviews with women Number of eligible women 2,500 3,449 5,949 Number of eligible women interviewed 2,374 3,317 5,691 Eligible woman response rate 95.0 96.2 95.7 Interviews with men Number of eligible men 2,063 3,282 5,345 Number of eligible men interviewed 1,903 3,112 5,015 Eligible man response rate 92.2 94.8 93.8 Household Population and Housing Characteristics | 11 HOUSEHOLD POPULATION AND HOUSING CHARACTERISTICS 2 This chapter provides a descriptive summary of the social, economic, and demographic characteristics of households sampled in the survey with a focus on some basic background characteristics such as age, sex, education, place of residence, and socio-economic condition of households. This information is crucial for the interpretation of key demographic and health indicators from which to draw meaningful policies and programmes for intervention. This information is also a basis for gauging the representativeness of the survey. The basic characteristics of the sampled population, that is, age, sex, education, and place of residence, form the basis of the background information by which most key demographic and health indices are analysed throughout this report. New to the DHS in general, and the 2003 GDHS in particular, is the wealth quintile, which is an indicator of the level of wealth that is consistent with expenditure and income measures. The wealth quintile was constructed using information on household ownership of a number of consumer items, ranging from a television to a bicycle or car, as well as dwelling characteristics, such as source of drinking water, sanitation facilities, and type of material used for flooring. Each asset was assigned a weight (factor score) generated through principal components analysis, and the resulting asset scores were standardised in relation to a normal distribution with a mean of zero and standard deviation of one. Each household was then assigned a score for each asset, and the scores were summed for each household; individuals were ranked according to the total score of the household in which they resided. The sample was then divided into quintiles from one (lowest) to five (highest). A single asset index was developed for the whole sample; separate indices were not prepared for the urban and rural population. In the 2003 GDHS, a household is defined as a person or a group of persons, related or unrelated, who live together in the same house or compound, share the same housekeeping arrangements, and are catered for as one unit. The Household Questionnaire was used to collect information on all usual residents and visitors who spent the night preceding the survey in the household. This mode of data collection allows the analysis of either the de jure (usual residents) or de facto (those who are present at the time of the interview) populations. 2.1 HOUSEHOLD POPULATION BY AGE AND SEX Age and sex are important variables in analysing demographic trends. Table 2.1 presents the distribution of the de facto household population in the 2003 GDHS survey by five-year age groups, according to sex and urban-rural residence. Figure 2.1 and Table 2.1 show the population by sex for Ghana. The data show that there are slightly more women (53 percent) than men (47 percent) in the overall population. There is a slightly higher concentration of women in the urban than rural areas (55 and 51 percent). The age structure is typical of a young population characterised by high fertility. This type of population structure imposes a heavy burden on the social and economic assets of a country. Ghana’s population is still young, with 44 percent of the population under 15 years, with the percentage in the older age groups (65 years and above) constituting just 5 percent of the population. 12 | Household Population and Housing Characteristics Figure 2.1 Population Pyramid 80+ 75-79 70-74 65-69 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5-9 0-4 0246810 0 2 4 6 8 10 Age Male Percent Female GDHS 2003 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, Ghana 2003 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Urban Rural Total ––––––––––––––––––––– –––––––––––––––––––– ––––––––––––––––––––– Age Male Female Total Male Female Total Male Female Total –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– <5 13.3 10.7 11.9 17.5 16.2 16.8 15.8 13.9 14.8 5-9 14.0 11.8 12.8 17.5 15.0 16.2 16.1 13.6 14.8 10-14 16.1 12.8 14.3 16.9 13.3 15.0 16.6 13.1 14.7 15-19 11.5 11.5 11.5 9.0 7.5 8.2 10.0 9.3 9.6 20-24 7.8 10.1 9.1 5.0 6.8 5.9 6.2 8.2 7.2 25-29 8.0 8.3 8.2 5.6 7.1 6.4 6.6 7.6 7.1 30-34 6.4 6.7 6.5 5.0 6.1 5.6 5.5 6.4 6.0 35-39 4.9 5.9 5.4 4.3 5.8 5.1 4.5 5.9 5.2 40-44 3.6 4.6 4.2 3.7 4.4 4.0 3.7 4.5 4.1 45-49 3.7 3.8 3.7 4.0 3.5 3.7 3.9 3.6 3.7 50-54 2.7 3.7 3.3 2.5 3.9 3.2 2.6 3.8 3.3 55-59 1.7 2.4 2.1 1.7 2.6 2.1 1.7 2.5 2.1 60-64 2.3 2.3 2.3 2.1 2.3 2.2 2.2 2.3 2.3 65-69 1.5 1.5 1.5 1.9 1.6 1.7 1.7 1.6 1.6 70-74 1.1 1.7 1.4 1.3 1.5 1.4 1.2 1.6 1.4 75-79 0.7 0.7 0.7 0.9 1.0 0.9 0.8 0.8 0.8 80 + 0.6 1.3 1.0 0.9 1.3 1.1 0.8 1.3 1.0 Don't know/missing 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Number 4,575 5,539 10,115 6,925 7,326 14,250 11,500 12,865 24,365 Household Population and Housing Characteristics | 13 Data from the GDHS show an excess of males in the 10-14 age group, and a deficit in the 20-24 and 40-44 age groups. The excess of males in the 10-14 age group could be the effect of age shifting by interviewers in the DHS, out of the eligible age range (15-59) for the individual interviews, to reduce their workload. This effect is also obvious for females in the 50-54 age group, who are not eligible for the individual interview. The deficit of males in the 20-24 and 40-44 age groups is also reflected in the 2000 Census (GSS, 2002), and may be attributable to differential outmigration of males in search of jobs. 2.2 HOUSEHOLD COMPOSITION The size and composition of households and the sex of the head of household are important aspects that impact on household welfare. Table 2.2 shows information collected in the 2003 GDHS on sex composition and household size. The mean household size is 4.0, with household size in rural areas (4.3) larger than in urban areas (3.6). Two-thirds (66 percent) of households are headed by males, while a third (34 percent) are headed by females. The percentage of female-headed households is higher in urban (40 percent) than in rural areas (29 percent). Single-person households are more common in urban (25 percent) than rural areas (18 percent). This may be due to an influx of unmarried young persons migrating to urban areas in search of employment or to further their education. Table 2.2 Household composition Percent distribution of households by sex of head of house- hold and by household size, according to residence, Ghana 2003 –––––––––––––––––––––––––––––––––––––––––––––––––––– Residence ––––––––––––––––– Characteristic Urban Rural Total –––––––––––––––––––––––––––––––––––––––––––––––––––– Sex of head of household Male 60.3 71.1 66.2 Female 39.7 28.9 33.8 Total 100.0 100.0 100.0 Number of usual members 1 24.5 17.5 20.7 2 14.2 11.3 12.6 3 14.2 13.1 13.6 4 15.4 14.5 14.9 5 11.4 13.6 12.6 6 8.3 10.6 9.5 7 5.3 7.9 6.7 8 2.7 4.3 3.6 9+ 4.0 6.9 5.6 Total 100.0 100.0 100.0 Number of households 2,870 3,381 6,251 Mean size 3.6 4.3 4.0 –––––––––––––––––––––––––––––––––––––––––––––––––––– Note: Table is based on de jure members, i.e., usual resi- dents. 14 | Household Population and Housing Characteristics 2.3 EDUCATIONAL ATTAINMENT OF HOUSEHOLD MEMBERS Education is important in that it helps individuals to make informed decisions that impact their health and well-being. Ghana’s system of education has undergone several stages of restructuring over the past 25 years (Sedgwick, 2000). The current system of formal education was introduced in 1989. It is based on a three-tier system: six years of primary education, followed by three years of junior secondary school (JSS), and a further three years at the senior secondary school (SSS) level. From the mid-1970s till the introduction of the current system of education, the six years of primary education was followed by five years of secondary education―three years of JSS and two years of SSS. Prior to the mid-1970s, students who completed six years of primary education had a choice. They could attend four years of middle school or attend five years of secondary school with a small group having the further option to pursue an additional two years of pre-university education. Upon completion of formal schooling, a student could choose to further his or her education at the tertiary level. In addition to university education, there are a host of institutions offering vocational, technical, and professional training that may be tertiary or non-tertiary. The different systems of formal education have been taken into account in tabulating the educational attainment of women and men interviewed in the 2003 GDHS. Table 2.3.1 shows the percent distribution of the de facto female household population age six years and over by highest level of education attended or completed, according to background characteristics. Thirty-seven percent of women have never been to school, about 30 percent have some primary or have completed primary school, 31 percent have some secondary or have completed secondary school, and about 2 percent have more than secondary school education. The data reveal that the proportion of women with no education is higher among older women, suggesting some improvement in education over the years. This may be due to the impact of the Free Compulsory Universal Basic Education (fCUBE) programme introduced in 1996. Education varies by place of residence. Urban women are more likely to be educated than rural women. For example, 26 percent of urban females have no education, compared with 47 percent of rural females. The proportion of urban females with some secondary education or higher (47 percent) is more than twice as high as that of rural females (21 percent). It is notable that females in the northern half of the country (Northern, Upper East, and Upper West regions) are seriously disadvantaged educationally. More than two-thirds of women in these regions have never been to school, compared with one-fifth in the Greater Accra region. In addition, 13 percent of females in Greater Accra have completed secondary education or higher, compared with less than 2 percent in the Northern and Upper West regions. With the exception of the three northern regions, the majority of females in all regions have been to school. It is worth noting that the proportion of female household members who have never attended school decreases with higher wealth status. Sixty-five percent of women in the lowest wealth quintile have no education compared with only 15 percent in the highest quintile. Table 2.3.2 shows that 26 percent of males have never been to school, 33 percent have had some primary or have completed primary education, 37 percent have had some secondary or completed secondary education, and about 4 percent have more than secondary education. One-third of males in rural areas have no education compared with only 15 percent in urban areas. There is a marked urban- rural differential in secondary and higher education: 16 percent of males in urban areas have completed secondary or higher education compared with only 4 percent in rural areas. Household Population and Housing Characteristics | 15 Table 2.3.1 Educational attainment of household population: women 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, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– More Median Some than Don’t years Background No Some Completed secon- Completed secon- know/ of characteristic education primary primary1 dary secondary2 dary missing Total Number schooling ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Age 6-9 47.5 51.6 0.0 0.0 0.0 0.0 0.9 100.0 1,457 0.0 10-14 15.2 65.0 6.3 13.3 0.0 0.0 0.3 100.0 1,685 3.0 15-19 14.7 15.3 7.5 58.3 3.9 0.1 0.1 100.0 1,191 6.8 20-24 21.6 11.8 6.3 44.8 11.5 3.9 0.0 100.0 1,053 8.0 25-29 33.9 11.3 6.1 36.5 9.8 2.3 0.1 100.0 981 5.7 30-34 35.5 13.1 4.5 41.0 2.6 3.3 0.0 100.0 821 5.3 35-39 41.6 14.2 5.7 33.6 2.7 2.2 0.1 100.0 756 3.0 40-44 35.3 12.9 3.9 41.6 3.0 2.6 0.7 100.0 577 5.4 45-49 40.2 15.4 2.0 37.5 0.7 4.0 0.1 100.0 465 3.3 50-54 61.3 8.5 2.6 23.5 0.4 2.6 1.1 100.0 492 0.0 55-59 68.0 9.3 1.4 16.4 1.1 2.5 1.4 100.0 320 0.0 60-64 81.1 3.7 2.6 11.2 0.0 1.0 0.4 100.0 297 0.0 65+ 86.6 5.2 0.2 6.6 0.3 0.9 0.4 100.0 674 0.0 Residence Urban 25.9 22.5 4.6 38.1 5.8 2.8 0.4 100.0 4,841 5.3 Rural 46.8 27.8 4.0 19.5 0.9 0.6 0.4 100.0 5,944 0.0 Region Western 29.5 30.8 5.7 29.0 2.8 2.0 0.1 100.0 960 3.3 Central 39.1 29.1 4.6 23.7 2.2 1.2 0.2 100.0 904 1.6 Greater Accra 20.3 21.9 4.6 40.1 9.2 3.6 0.2 100.0 1,547 6.4 Volta 31.2 33.0 4.8 26.9 2.3 1.6 0.3 100.0 1,023 2.5 Eastern 29.6 27.0 6.2 32.4 1.9 2.2 0.6 100.0 1,166 3.7 Ashanti 28.3 25.7 4.1 37.7 2.7 1.1 0.4 100.0 2,154 4.1 Brong Ahafo 37.0 27.5 5.2 27.4 1.8 0.6 0.4 100.0 1,061 2.4 Northern 74.4 16.4 1.3 6.2 0.9 0.7 0.1 100.0 989 0.0 Upper East 71.1 18.3 1.1 6.7 1.3 1.0 0.6 100.0 661 0.0 Upper West 66.1 19.2 2.8 8.7 1.5 0.3 1.3 100.0 321 0.0 Wealth quintile Lowest 65.1 22.7 2.5 8.8 0.2 0.1 0.6 100.0 1,992 0.0 Second 47.2 29.6 3.8 18.4 0.4 0.2 0.4 100.0 2,046 0.0 Middle 36.5 28.9 5.5 27.4 1.2 0.2 0.3 100.0 2,172 2.1 Fourth 28.0 26.3 5.4 35.9 3.2 0.9 0.3 100.0 2,204 4.1 Highest 15.4 20.1 3.9 44.8 9.5 6.0 0.3 100.0 2,372 7.9 Total 37.4 25.4 4.3 27.8 3.1 1.6 0.4 100.0 10,785 2.1 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Note: Total includes 13 women with missing information on age who are not shown separately. 1 Completed grade 6 at the primary level 2 Completed grade 12 at the secondary level Across the regions the pattern among the male population is similar to that exhibited by the females. Males in the three northern regions are disadvantaged, with 54-59 percent never having been to school compared with less than 20 percent in the other regions, except Brong Ahafo (24 percent) and the Central (21 percent) regions. The variation in education among the male population according to wealth quintile is similar to that among the female population. Wealthy males are less likely to have no education. For example, 7 percent of males in the highest wealth quintile have no education compared with 53 percent in the lowest. 16 | Household Population and Housing Characteristics Table 2.3.2 Educational attainment of household population: men Percent distribution of the de facto male household population age six and over by highest level of education attended or com- pleted, according to background characteristics, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– More Some than Don’t Median Background No Some Completed secon- Completed secon- know/ years of characteristic education primary primary1 dary secondary2 dary missing Total Number schooling ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Age 6-9 45.2 53.6 0.1 0.1 0.0 0.0 1.0 100.0 1,544 0.0 10-14 14.8 67.7 6.1 11.1 0.0 0.0 0.3 100.0 1,907 2.7 15-19 9.3 18.9 7.9 60.8 3.7 0.0 0.1 100.0 1,148 6.9 20-24 12.9 8.7 5.6 50.5 17.6 4.7 0.0 100.0 707 8.4 25-29 15.6 9.1 4.7 43.6 16.3 10.3 0.3 100.0 758 8.5 30-34 21.4 7.5 4.0 48.5 13.1 5.2 0.5 100.0 634 9.0 35-39 25.4 8.4 4.1 46.7 7.7 6.6 1.1 100.0 522 9.1 40-44 24.4 6.3 2.1 50.9 6.8 9.3 0.3 100.0 420 9.3 45-49 26.3 9.0 4.2 45.9 6.0 8.5 0.2 100.0 445 9.2 50-54 25.7 7.6 3.5 48.2 4.8 9.3 0.9 100.0 301 9.1 55-59 34.7 5.6 3.2 35.9 4.7 15.4 0.6 100.0 197 9.0 60-64 52.7 8.2 3.3 27.6 3.5 4.1 0.5 100.0 253 0.0 65+ 66.3 6.8 1.7 20.6 2.1 2.3 0.2 100.0 523 0.0 Residence Urban 15.2 25.5 4.4 38.1 9.7 6.6 0.5 100.0 3,865 6.9 Rural 33.3 31.4 4.0 26.9 2.5 1.5 0.4 100.0 5,511 2.2 Region Western 14.5 32.5 6.8 37.3 5.9 2.9 0.0 100.0 817 5.4 Central 21.0 34.4 3.7 35.1 2.6 3.2 0.0 100.0 719 4.1 Greater Accra 12.8 21.9 3.7 37.0 15.4 8.4 0.8 100.0 1,194 8.5 Volta 18.0 30.7 5.2 37.5 3.6 4.7 0.3 100.0 825 5.2 Eastern 18.9 28.0 6.7 37.9 3.2 4.6 0.8 100.0 1,031 5.4 Ashanti 16.4 29.1 3.0 43.0 5.6 2.4 0.4 100.0 1,773 5.8 Brong Ahafo 23.5 34.6 4.3 29.4 4.6 3.4 0.2 100.0 1,020 3.6 Northern 58.6 25.3 2.2 9.0 3.2 1.4 0.2 100.0 1,060 0.0 Upper East 54.1 28.3 3.8 10.2 1.6 0.6 1.3 100.0 651 0.0 Upper West 54.2 26.3 2.6 11.3 2.5 2.8 0.4 100.0 286 0.0 Wealth quintile Lowest 52.6 28.8 3.4 12.8 1.1 0.6 0.7 100.0 1,865 0.0 Second 30.4 35.3 4.8 26.6 1.7 0.7 0.5 100.0 1,899 2.3 Middle 22.5 33.4 5.0 35.3 2.3 1.1 0.3 100.0 1,911 3.8 Fourth 16.5 26.1 4.6 40.7 7.4 4.4 0.3 100.0 1,855 6.3 Highest 7.0 20.8 3.0 42.4 15.1 11.3 0.4 100.0 1,846 8.9 Total 25.9 29.0 4.2 31.5 5.5 3.6 0.4 100.0 9,376 3.9 –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Note: Total includes 18 men with missing information on age who are not shown separately. 1 Completed grade 6 at the primary level 2 Completed grade 12 at the secondary level Men are more educated than women at all levels of education, implying that females continue to lag behind males in education. The median number of years of schooling completed is twice as high among men (3.9 years) as among women (2.1 years). It is disappointing to note that the level of education has deteriorated over the last five years for both women and men. The proportion of women with no education rose from 34 percent in 1998 (GSS and MI, 1999) to 37 percent in 2003, with the median number of years of schooling falling slightly from 2.3 to 2.1 over the five years. Similarly, the proportion Household Population and Housing Characteristics | 17 of men with no education rose from 21 percent (GSS and MI, 1999) to 26 percent with the median number of years of schooling falling from 4.9 to 3.9 over the last five years. Nevertheless, the male- female gap in educational attainment has narrowed over the same period. The 2003 GDHS collected information on school attendance among the population 6-24 years that allows the calculation of net attendance ratios (NARs) and gross attendance ratios (GARs). 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 measure of the secondary-school-age (12-18 years) population that is attending secondary school. By definition, the NAR cannot exceed 100 percent. The GAR however, measures participation at each level of schooling among persons age 6-24. The GAR is almost always higher than the NAR for the same level because the GAR includes participation by those who may be older, because they may have started school late, may have repeated one or more grades in school, or may have dropped out of school and later returned, or may be younger than the official age range for that level. Table 2.4 presents data on NAR and GAR for the de jure household population by level of schooling and sex, according to place of residence and wealth quintile. Sixty percent of children age 6-11, who should be attending primary school, are currently doing so. At the same time, the GAR at the primary school level is 95 percent, indicating that more than a third (35 percent) of young Ghanaians attending primary school are above or below primary school age. Not surprisingly, both the NAR and GAR are much lower at the secondary than at the primary school level. Slightly more than one-third of students age 12-18 who should be attending secondary school are in school at that level. The GAR for secondary school is 41 percent, indicating that the proportion of underage or overage youths in secondary school is not so large. The results show similar proportions of NAR for females and males at primary and secondary school level, indicating that there is no gender gap in school attendance among the Ghanaian school age population who should be attending school at a given level. However, the GARs at primary and secondary school levels are slightly higher for males than females, indicating a relatively higher overage or underage attendance among males than females. As expected, school attendance ratios at both the primary and secondary levels are lower in rural than in urban areas. For instance, the NAR at the primary school level in rural areas is 56 percent compared with 68 percent in urban areas. Similarly, the GAR at secondary school is 31 percent in rural areas compared with 53 percent in urban areas. Regional differences are obvious for the NAR and GAR with attendance ratios notably lower among the three northern regions (Northern, Upper East, and Upper West) compared with all other regions, and especially in the case of the GAR at the primary school level. There is a strong relationship between household economic status and school attendance that can be seen at both the primary and secondary levels and among males and females. For example, the NAR increases from 43 percent among students from poorer households (lowest wealth quintile) in primary school to 78 percent among students from richer households (highest wealth quintile). Similarly, the GAR rises three-fold from 20 percent among secondary school attendees in the lowest wealth quintile to 63 percent among those in the highest wealth quintile. The Gender Parity Index (GPI) represents the ratio of the GAR for females to the GAR for males. It is presented at both the primary and secondary levels and offers a summary measure of gender differences in school attendance rates. A GPI less than 1 indicates that a smaller proportion of females than males attend school. In Ghana, the GPI is slightly less than 1 (0.9) for both primary and secondary school attendance, indicating that the gender gap is relatively small. There are no marked differences in the GPI by place of residence. The Northern Region has the widest gap (0.8) for primary school attendance and the Brong Ahafo Region has the widest gap (0.7) for secondary school attendance. 18 | Household Population and Housing Characteristics Table 2.4 School attendance ratios Net attendance ratios (NAR), gross attendance ratios (GAR), and gender parity index for the de jure household population by level of schooling and sex, according to background characteristics, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Net attendance ratio1 Gross attendance ratio2 Gender Background –––––––––––––––––––––––––– ––––––––––––––––––––––––– parity characteristic Male Female Total Male Female Total index3 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– PRIMARY SCHOOL ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence Urban 69.9 66.7 68.3 110.0 100.4 105.2 0.91 Rural 56.2 55.2 55.8 92.0 86.8 89.6 0.94 Region Western 70.6 68.2 69.4 109.3 102.8 106.1 0.94 Central 61.3 60.6 61.0 98.5 99.5 99.0 1.01 Greater Accra 72.3 71.1 71.7 108.9 102.9 105.9 0.95 Volta 60.7 66.1 63.5 106.9 107.0 106.9 1.00 Eastern 58.9 60.9 59.8 99.6 92.0 96.1 0.92 Ashanti 69.0 65.9 67.5 104.8 98.5 101.7 0.94 Brong Ahafo 66.3 56.5 62.0 119.8 109.0 115.1 0.91 Northern 47.4 39.5 43.8 70.8 55.1 63.7 0.78 Upper East 42.4 46.4 44.2 70.3 61.0 66.0 0.87 Upper West 41.9 41.2 41.5 74.9 68.2 71.5 0.91 Wealth quintile Lowest 43.8 41.8 42.9 72.6 66.7 69.9 0.92 Second 57.9 53.6 55.9 100.9 89.4 95.5 0.89 Middle 66.0 62.4 64.4 107.9 101.3 104.8 0.94 Fourth 67.6 68.3 67.9 109.2 101.5 105.4 0.93 Highest 78.9 77.0 77.9 109.2 105.7 107.4 0.97 Total 61.0 59.6 60.4 98.4 92.0 95.4 0.94 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– SECONDARY SCHOOL ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence Urban 45.4 44.8 45.1 55.3 50.1 52.6 0.90 Rural 26.8 25.9 26.4 33.0 28.3 30.8 0.86 Region Western 36.8 41.9 39.3 41.9 44.3 43.1 1.06 Central 34.3 30.5 32.4 39.6 34.9 37.2 0.88 Greater Accra 47.7 48.6 48.2 59.2 56.6 57.7 0.96 Volta 36.2 35.4 35.8 50.5 38.7 44.6 0.77 Eastern 37.8 35.6 36.8 45.8 37.2 41.6 0.81 Ashanti 42.8 39.2 40.9 48.1 42.2 45.1 0.88 Brong Ahafo 32.8 26.6 30.0 40.5 29.9 35.7 0.74 Northern 17.4 15.8 16.7 24.6 19.2 22.4 0.78 Upper East 16.5 23.2 19.4 24.1 26.6 25.2 1.11 Upper West 20.2 22.5 21.2 27.4 26.4 27.0 0.96 Wealth quintile Lowest 15.4 15.2 15.3 22.1 17.0 19.8 0.77 Second 27.3 19.7 23.9 33.6 21.5 28.2 0.64 Middle 34.5 34.7 34.6 40.8 36.7 38.8 0.90 Fourth 40.4 42.3 41.4 51.0 47.1 49.0 0.92 Highest 57.5 53.6 55.3 66.6 60.8 63.3 0.91 Total 34.6 35.0 34.8 42.4 38.8 40.6 0.91 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 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 at- tending 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 pri- mary-school-age population. The GAR for secondary school is the total number of secondary school students, expressed as a percentage of the official secondary-school-age population. If there are significant numbers of overage and underage stu- dents 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. Household Population and Housing Characteristics | 19 Figure 2.2 shows the age-specific attendance rates (ASAR) for the de jure household population age 6-24 by sex. The ASAR shows participation in schooling at any level, from primary through higher education. The closer the ASAR is to 100, the higher the participation of a given age population at that level. Less than 50 percent of children age seven and below are attending school. School attendance rises markedly up to age 11, remains high up to age 14, and then gradually declines. There are no marked differences in the proportion of males and females attending school up to age 15, after which there are significantly higher proportions of males than females attending school. Figure 2.2 Age-Specific Attendance Rates 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Age 0 20 40 60 80 100 Percent Male Female GDHS 2003 2.4 HOUSING CHARACTERISTICS There is a strong correlation between the socio-economic condition of households and the vulnerability of its members, and especially children, to common diseases. The amenities and assets available to households are important in determining the general socio-economic status of the population. The GDHS included questions on a household’s access to electricity, source of drinking water, type of sanitation facilities, flooring materials, and ownership of durable goods. Table 2.5 presents the distribution of households by household characteristics, according to residence. One in two households in Ghana has electricity. Three-fourths of households in urban areas (77 percent) have electricity compared with one-fourth (24 percent) of rural households. The 2003 data show an increase in the use of electricity among rural households over the last five years (GSS and MI, 1999), while access to electricity in urban households has declined over the same time period. The decline in the use of electricity by urban households may be attributed to the rapid development in housing projects, some of which are not yet connected to the national power grid. The availability of and accessibility to potable water may, to a large extent, minimise the prevalence of water-borne diseases among household members, especially young children. The source of drinking water is important because potentially fatal diseases, such as diarrhoeal diseases, guinea worm, bilharzia, typhoid, cholera, and dysentery, are common in the country. The main sources of drinking water are piped, protected well or borehole, and rivers or streams. Sixteen percent of households have 20 | Household Population and Housing Characteristics access to piped water in their dwelling, yard or plot, while 23 percent access drinking water from a public tap. Twelve percent of households get their drinking water from open wells, and 28 percent obtain drinking water from protected wells or boreholes. Fourteen percent of households obtain drinking water from rivers and streams. Not surprisingly, rural households have less access to clean drinking water than urban households. For example, one in three urban households have piped water in their dwelling, yard or plot, compared with 2 percent of rural households. The major source of drinking water for rural households is a well or borehole. For example, two-fifths of rural households get their drinking water from a protected public well or borehole, compared with one-tenth of urban households. The overall access to piped drinking water has remained the same over the last five years (GSS and MI, 1999). It takes eight in ten urban households and about half of rural households less than 15 minutes to reach their nearest source of drinking water. The median time to source in urban households is 4 minutes compared with 14 minutes among rural households. The vast majority of households have access to water all the time. Rural households are more likely than urban households to have access to water all the time. The availability of toilet facilities in households ensures a more efficient and hygienic method of human waste disposal. Most households in Ghana (42 percent) have traditional pit latrines, 26 percent have improved ventilated pit latrines (KVIP), and 11 percent have flush toilets. One-fifth of households have no toilet facility. Lack of a toilet facility is more common in rural areas (31 percent) than in urban areas (7 percent). Two-fifths of urban households have KVIP toilets, compared with 14 percent of households in rural areas. Traditional pit toilets are twice as common in rural areas (54 percent) as in urban areas (27 percent). Flush toilets are more common in urban households Table 2.5 Household characteristics Percent distribution of households by household characteristics, according to residence, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence Household ––––––––––––––––– characteristic Urban Rural Total ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Electricity Yes 76.9 24.1 48.3 No 23.1 75.8 51.6 Missing 0.0 0.1 0.1 Total 100.0 100.0 100.0 Source of drinking water Piped into dwelling 11.5 0.7 5.7 Piped into compound/plot 21.9 1.2 10.7 Public tap 39.1 8.9 22.8 Open well in dwelling/yard 2.5 1.3 1.9 Open public well 5.9 12.8 9.6 Protected well in dwelling/yard 2.4 1.7 2.0 Protected public well 7.8 41.1 25.8 Spring 0.3 0.9 0.6 River, stream 2.0 24.3 14.1 Pond, lake 0.5 2.4 1.5 Dam 0.1 3.7 2.0 Rainwater 0.5 0.2 0.3 Tanker truck 2.3 0.4 1.3 Satchel water 3.0 0.3 1.5 Other 0.4 0.0 0.2 Missing 0.0 0.1 0.1 Total 100.0 100.0 100.0 Time to water source Percentage <15 minutes 79.8 46.5 61.8 Median time to source 4.3 14.3 9.4 Water availability All the time 75.9 91.9 84.5 Several hours per day 11.4 4.8 7.8 A few times a week 9.1 1.7 5.1 Less frequently 2.7 1.0 1.8 Not at all 0.7 0.4 0.5 Don’t know 0.1 0.0 0.1 Missing 0.1 0.2 0.2 Total 100.0 100.0 100.0 Sanitation facility Flush toilet 21.2 1.7 10.7 Traditional pit toilet 26.7 54.0 41.5 Ventilated improved pit latrine 40.8 13.5 26.0 No facility, bush, field 6.7 30.6 19.6 Bucket, pan 4.5 0.2 2.2 Missing 0.1 0.1 0.1 Total 100.0 100.0 100.0 Sharing toilet facilities No 14.6 8.6 11.4 Yes 78.5 60.8 68.9 No facility 6.7 30.6 19.6 Missing 0.1 0.1 0.1 Total 100.0 100.0 100.0 Flooring material Earth/sand/mud 3.2 17.8 11.1 Mud mixed with dung 0.1 2.4 1.3 Wood/palm/bamboo/parquet 0.4 0.1 0.2 Linoleum 19.5 5.7 12.0 Ceramic tiles/terrazo 3.9 0.3 2.0 Cement 54.5 71.5 63.7 Carpet 18.4 2.1 9.6 Missing 0.1 0.1 0.1 Total 100.0 100.0 100.0 Continued… Household Population and Housing Characteristics | 21 (21 percent) than in rural households (2 percent). Access to flush toilets has risen over the last five years, from 8 percent in 1998 (GSS and MI, 1999) to 11 percent in 2003. The majority of households (69 percent) share toilet facilities with one or more households. The type of flooring material used in dwellings is a proxy indicator of the socio- economic status of the household as well as its likely exposure to disease-causing agents. Most households in Ghana (87 percent) have finished floors (terrazzo, tiles, cement, carpet, and linoleum), with only 12 percent of households having rudimentary or natural flooring material (earth, sand, mud or mud mixed with dung). There has been little change over the last five years in the percentage of households with finished flooring. Rural households are much more likely to have cement floors (72 percent) than urban households (55 percent). The second most common flooring material in rural areas is earth, sand, or mud (18 percent). About one-fifth of urban households have linoleum floors and almost the same proportion have carpeted floors. Two common sources of cooking fuel in the country are firewood (59 percent) and charcoal (30 percent). One in four urban households uses firewood, while 87 percent of rural households depend on firewood as their main source of cooking fuel. On the other hand, more than 50 percent of urban households use charcoal compared with 10 percent of rural households. Liquified petroleum gas (LPG) or natural gas is used more commonly by urban households (15 percent) than households in rural areas (1 percent). However, even in urban areas, few households use electricity for cooking (1 percent), presumably because of the higher cost. The GDHS also included questions pertaining to disposal of household waste, possibility of eviction, and membership in mutual health organisations (MHO) or health insurance schemes (HIS). Data on these are also shown in Table 2.5. The majority of households (62 percent) dispose of their household waste in the street or an empty plot, with a much higher proportion of rural households than urban households disposing of household waste in this manner. Fourteen percent of households have their waste collected by the government and this is predominantly done in urban areas. Seven percent of households burn their waste, 4 percent have their waste collected by a community association, 4 percent dump it in their compound, while 3 percent each bury or compost their household waste. When asked about the possibility of eviction, most household respondents said that this was not likely at all (71 percent). Twelve percent of households mentioned that it was very likely that they could Table 2.5—Continued ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence Household ––––––––––––––––– characteristic Urban Rural Total ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Cooking fuel Electricity 0.6 0.1 0.3 LPG, natural gas 14.6 1.4 7.4 Biogas 0.7 0.0 0.3 Kerosene 1.3 0.4 0.8 Coal, lignite 0.7 0.0 0.3 Charcoal 54.1 10.1 30.3 Firewood, straw 25.6 87.4 59.0 Dung 0.0 0.1 0.1 Other 2.3 0.4 1.3 Total 100.0 100.0 100.0 Disposal of household waste Collected by government 30.6 0.4 14.2 Collected by community assoc. 4.2 3.9 4.0 Collected by private company 4.8 0.3 2.3 Dumped in compound 2.5 5.9 4.3 Dumped in street/empty plot 47.3 74.0 61.7 Burned 7.7 6.4 7.0 Buried 2.5 3.5 3.0 Composted 0.2 5.3 3.0 Other 0.2 0.1 0.1 Missing 0.1 0.2 0.3 Total 100.0 100.0 100.0 Possibility of eviction Very likely 21.0 4.9 12.3 Somewhat likely 20.7 10.9 15.4 Not at all likely 56.7 83.6 71.3 Don’t know 1.5 0.6 1.0 Total 100.0 100.0 100.0 Member of HIS Yes 3.5 2.8 3.1 No 95.9 96.3 96.2 Don’t know 0.6 0.8 0.7 Missing 0.0 0.2 0.1 Total 100.0 100.0 100.0 Number of households 2,870 3,381 6,251 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– HIS = Health Insurance Scheme 22 | Household Population and Housing Characteristics be evicted, while 15 percent of households mentioned that it was somewhat likely. Urban households are more likely to report that they face possible eviction than rural households. A very small percentage of households in Ghana (3 percent) belong to an HIS. Among those who belong to an HIS, the majority belong to an MHO (43 percent) or have government health coverage (20 percent), with 29 percent belonging to a private health insurance scheme (data not shown). About half of those who belong to any kind of insurance scheme mentioned that they have benefited from it in the past, and a large majority (91 percent) of those who are not members of an insurance scheme indicate that they would consider joining one in the future (data not shown). 2.6 HOUSEHOLD DURABLE GOODS Respondents were asked about ownership of particular household goods such as radios and television sets (access to media), refrigerators (access to food storage), telephones (access to other means of communication), and modes of transport (bicycle, motorcycle, car, or truck). Ownership of these items is also indicative of the household’s social and economic well-being. Table 2.6 presents data on the percentage of households possessing various durable consumer goods, by residence. The results show that 71 percent of households own a radio, 26 percent have a television, 23 percent have bicycles, and 19 percent own refrigerators. It is striking to note that nearly one in four households possess none of the durable items identified. There has been a noticeable rise in ownership of consumer durable goods over the last five years, with the most marked increase in the ownership of refrigerators, which increased from 2 percent in 1998 (GSS and MI, 1999) to 19 percent in 2003. Table 2.6 Household durable goods Percentage of households possessing various durable consumer goods, by residence, Ghana 2003 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Residence ––––––––––––––––– Durable consumer goods Urban Rural Total ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Radio 76.1 66.6 71.0 Television 44.9 9.9 26.0 Telephone 14.3 0.7 6.9 Refrigerator 34.6 5.5 18.8 Video deck 20.6 2.2 10.6 Bicycle 16.0 29.1 23.1 Motorcycle 2.6 1.7 2.1 Car/truck 9.0 2.4 5.4 Tractor 0.4 0.2 0.3 Horse/cart 0.3 1.4 0.9 None of the above 18.3 27.1 23.1 Number of households 2,870 3,381 6,251 Sixty-seven percent of households in rural areas have a radio. Smaller proportions, however, own consumer items such as televisions, telephones, refrigerators, and cars. Twenty-nine percent of rural households own bicycles, compared with 16 percent of urban households. Televisions, refrigerators, telephones, and cars or trucks are mostly restricted to urban areas, presumably due to the lack of electricity or affordability in rural areas. Characteristics of Survey Respondents | 23 CHARACTERISTICS OF SURVEY RESPONDENTS 3 The purpose of this chapter is to provide a descriptive summary of the demographic and socio- economic profile of respondents in the 2003 GDHS. This basic information on women and men in the reproductive age group is crucial for the interpretation of the 2003 GDHS findings within the context of reproduction, health, and women’s status and empowerment. The percent distribution of respondents by the various demographic and socio-economic characteristics can also be used as an approximate indicator of the representativeness of the survey sample to the general population. The main background characteristics described in detail that will be used in subsequent chapters on reproduction and health are: age at the time of the survey, marital status, residence, education, and wealth quintiles. This chapter also includes information on literacy, exposure to mass media, employment and earnings, and women’s position and decisionmaking power in relation to others in the household. 3.1 BACKGROUND CHARACTERISTICS OF RESPONDENTS Table 3.1 shows data on the background characteristics of the 5,691 female respondents age 15- 49 and the 5,015 male respondents age 15-59, interviewed in the 2003 GDHS, by background characteristics, including age, marital status, urban-rural residence, region, education, religion, and ethnicity. The age distribution shows that more than one in two females (55 percent) and males (51 percent) are under age 30. The proportion in each age group tends to decrease with increasing age for both sexes. The data show that most of the respondents are currently married or living together, although the proportion of women (62 percent) who are married or living together is higher than men (53 percent). Four in ten men (41 percent) have never married, compared with only about three in ten women (28 percent). Nine percent of women and 6 percent of men are divorced, separated, or widowed. The distribution of respondents by urban-rural residence shows that men are slightly more likely to live in rural areas (55 percent) than women (52 percent). By region, the distribution of respondents varies markedly. For example, about one-fifth of respondents are from the Ashanti Region, about one- sixth are from Greater Accra, and about one-tenth each are from the Western, Eastern, Brong Ahafo, and Northern regions. Less than 3 percent of women and men are from the Upper West Region. Twenty-eight percent of women and 18 percent of men have no education. About one-fifth of women and one-sixth of men have only primary education and two-fifths have only middle/JSS level of education. Men are twice as likely to have attained the secondary level of education as women (23 and 12 percent, respectively). The majority of respondents are Christians―77 percent of women and 70 percent of men. Sixteen percent of women and 19 percent of men are Moslems. The ethnic composition shows that Akans are the predominant group, with 51 percent of women and 47 percent of men, followed by the Mole Dagbon (13 percent of women and 18 percent of men). 24 | Characteristics of Survey Respondents Table 3.1 Background characteristics of respondents Percentage distribution of women and men by background characteristics, Ghana 2003 Number of women Number of men Background characteristic Weighted percent Weighted Unweighted Weighted percent Weighted Unweighted Age 15-19 20.2 1,148 1,113 22.1 1,107 1,095 20-24 17.8 1,012 997 13.6 684 692 25-29 16.7 951 966 15.0 754 727 30-34 14.1 802 818 12.6 633 633 35-39 12.7 722 724 9.9 498 518 40-44 10.2 579 572 8.2 412 411 45-49 8.4 477 501 8.8 441 441 50-54 na na na 5.9 294 300 55-59 na na na 3.8 192 198 Marital status Never married 28.4 1,616 1,509 40.7 2,042 2,002 Married 54.2 3,087 3,273 48.6 2,439 2,514 Living together 8.1 462 421 4.6 233 212 Divorced/separated 7.3 416 368 5.4 272 257 Widowed 1.9 110 120 0.6 29 30 Residence Urban 48.4 2,755 2,374 44.9 2,250 1,903 Rural 51.6 2,936 3,317 55.1 2,765 3,112 Region Western 9.7 553 524 9.5 476 457 Central 7.6 431 352 7.4 370 300 Greater Accra 16.6 942 835 14.6 733 621 Volta 8.6 492 442 8.8 440 386 Eastern 10.6 601 506 10.7 539 453 Ashanti 20.1 1,142 927 19.1 956 785 Brong Ahafo 10.0 569 638 10.5 528 593 Northern 8.8 499 610 10.5 527 638 Upper East 5.4 310 395 6.3 317 395 Upper West 2.7 153 462 2.6 130 387 Education No education 28.2 1,608 1,917 17.6 881 1,118 Primary 20.0 1,135 1,112 16.0 803 857 Middle/JSS 40.0 2,279 2,044 43.2 2,165 1,967 Secondary+ 11.8 669 618 23.2 1,165 1,073 Religion Roman Catholic 13.9 788 905 14.6 731 794 Anglican/Methodist/Presbyt. 17.8 1,016 907 15.9 799 709 Other Christian 45.6 2,597 2,352 39.4 1,978 1,785 Moslem 15.6 887 1,013 18.7 939 1,050 Traditional/spiritualist 2.7 152 210 4.7 238 317 No religion 4.4 250 302 6.5 327 355 Other/Missing 0.0 1 2 0.0 4 5 Ethnicity Akan 50.7 2,885 2,481 47.3 2,370 2,025 Ga/Dangme 8.2 465 437 7.5 374 338 Ewe 13.1 745 698 13.0 654 614 Guan 2.6 145 159 3.7 186 191 Mole-Dagbani 12.8 730 1,119 17.5 878 1,235 Grussi 2.4 134 171 2.4 121 157 Gruma 2.5 142 178 3.0 151 188 Hausa 1.3 74 62 1.1 56 50 Other 6.4 362 380 4.4 223 214 Missing 0.2 9 6 0.1 3 3 Total 100.0 5,691 5,691 100.0 5,015 5,015 Note: Education categories refer to the highest level of education attended, whether or not that level was completed. na = Not applicable Characteristics of Survey Respondents | 25 3.2 EDUCATIONAL ATTAINMENT AND LITERACY Education provides people with the knowledge and skills that can lead to a better quality of life. The level of education has been found to be highly associated with the health of mothers and their children and on their reproductive health behaviours. Tables 3.2.1 and 3.2.2 present the distribution of women and men by highest level of schooling attended or completed, and the median number of years of schooling, according to background characteristics. The data show that 28 percent of women have never been to school, 14 percent have some primary education only, while 6 percent have completed primary education. In addition, 44 percent have some secondary education, 8 percent have completed secondary school, with 3 percent having attained higher than secondary education. The data also show that men are more educated than women at all levels of education. For example, about twice as many men as women have completed secondary education or gone on to a higher level. On average, men have two more years of schooling than women—the median years of schooling among women and men is 6.2 and 8.3 years, Table 3.2.1 Educational attainment by background characteristics: women Percent distribution of women by highest level of schooling attended or completed, and median number of years of schooling, ac- cording to background characteristics, Ghana 2003 Highest level of schooling attended or completed Background characteristic No education Some primary Completed primary1 Some sec- ondary Completed secondary2 More than secondary Total Number of women Median years of schooling Age 15-19 12.3 15.7 7.7 59.4 4.8 0.1 100.0 1,148 6.9 20-24 19.6 13.8 6.4 45.5 9.6 5.1 100.0 1,012 8.0 25-29 32.2 11.9 6.9 38.0 8.9 2.1 100.0 951 5.8 30-34 33.9 15.3 5.4 39.6 2.7 3.2 100.0 802 5.1 35-39 40.4 12.7 5.5 36.4 2.4 2.6 100.0 722 4.2 40-44 35.9 12.3 4.6 42.8 2.2 2.2 100.0 579 5.4 45-49 39.9 15.7 3.0 36.0 1.3 4.1 100.0 477 3.4 Residence Urban 16.3 10.6 5.4 54.5 8.9 4.3 100.0 2,755 8.3 Rural 39.5 17.1 6.6 34.1 1.7 1.1 100.0 2,936 3.5 Region Western 22.3 17.3 7.2 45.8 4.5 2.9 100.0 553 6.5 Central 25.2 20.7 7.3 41.0 3.5 2.2 100.0 431 5.5 Greater Accra 12.4 11.2 5.2 54.0 12.0 5.2 100.0 942 8.5 Volta 20.7 21.9 5.4 44.2 5.1 2.8 100.0 492 6.3 Eastern 15.9 14.4 9.1 53.4 3.9 3.3 100.0 601 7.3 Ashanti 16.8 12.4 6.6 57.5 5.1 1.5 100.0 1,142 7.9 Brong Ahafo 27.4 15.5 7.4 45.3 2.7 1.7 100.0 569 6.0 Northern 78.8 6.2 2.3 9.2 2.0 1.5 100.0 499 0.0 Upper East 72.4 9.8 2.0 12.3 1.9 1.6 100.0 310 0.0 Upper West 63.3 11.5 3.7 18.3 2.3 0.8 100.0 153 0.0 Wealth quintile Lowest 63.6 14.1 4.8 16.5 0.6 0.3 100.0 970 0.0 Second 37.8 21.6 6.3 32.9 1.0 0.4 100.0 949 3.2 Middle 26.0 16.0 8.0 47.5 2.0 0.5 100.0 1,071 5.9 Fourth 18.6 14.0 6.9 53.9 5.3 1.3 100.0 1,245 7.6 Highest 8.4 7.2 4.4 58.5 13.2 8.3 100.0 1,457 9.0 Total 28.2 13.9 6.0 44.0 5.2 2.6 100.0 5,691 6.2 1 Completed grade 6 at the primary level 2 Completed grade 12 at the secondary level 26 | Characteristics of Survey Respondents respectively. High dropout of girls at primary and secondary levels may explain some of the differences in educational attainment between women and men. The Ghanaian government has been considering measures to enhance girls’ retention rates in schools. One such measure allows girls who drop out of school due to pregnancy to continue with their education after delivery. The data show that the educational attainment among both women and men has improved over time, as seen by the changes between age cohorts. For example, 40 percent of women in the oldest age cohort (45-49) have no education compared with 12 percent among those age 15-19, while the corresponding percentages for men are 35 and 8 percent, respectively. Table 3.2.2 Educational attainment by background characteristics: men Percent distribution of men by highest level of schooling attended or completed, and median number of years of schooling, according to background characteristics, Ghana 2003 Highest level of schooling attended or completed Background characteristic No educa- tion Some primary Completed primary1 Some sec- ondary Completed secondary2 More than secondary Total Number of men Median years of schooling Age 15-19 8.1 17.9 8.8 60.3 4.8 0.1 100.0 1,107 7.1 20-24 10.8 9.6 5.8 51.2 17.6 5.0 100.0 684 8.4 25-29 14.5 8.3 5.6 44.4 16.6 10.6 100.0 754 8.5 30-34 19.5 8.1 4.5 49.8 11.9 6.1 100.0 633 9.0 35-39 26.6 8.7 4.5 46.6 7.4 6.2 100.0 498 9.0 40-44 23.4 6.3 2.6 53.6 6.0 8.1 100.0 412 9.3 45-49 26.3 10.2 4.0 43.8 6.1 9.6 100.0 441 9.1 50-54 25.0 8.4 3.4 49.9 3.2 10.1 100.0 294 9.2 55-59 34.5 5.3 4.3 35.4 6.8 13.8 100.0 192 9.0 Residence Urban 8.0 6.3 4.9 54.6 15.7 10.4 100.0 2,250 9.2 Rural 25.4 13.9 6.0 47.0 4.8 2.9 100.0 2,765 6.7 Region Western 7.3 11.0 9.2 58.3 9.4 4.8 100.0 476 8.5 Central 9.3 14.8 5.8 60.3 4.6 5.2 100.0 370 8.3 Greater Accra 5.8 5.1 2.8 50.4 22.6 13.3 100.0 733 9.9 Volta 7.9 13.7 5.9 58.4 6.1 8.0 100.0 440 8.5 Eastern 7.9 8.9 8.5 62.3 5.0 7.4 100.0 539 8.8 Ashanti 9.4 7.0 5.0 63.8 10.6 4.2 100.0 956 8.7 Brong Ahafo 14.2 12.2 5.2 53.2 8.9 6.3 100.0 528 8.3 Northern 59.5 11.6 4.0 14.6 7.3 3.0 100.0 527 0.0 Upper East 48.9 18.8 5.6 21.7 3.6 1.4 100.0 317 0.0 Upper West 44.9 16.4 4.2 23.4 5.5 5.5 100.0 130 2.2 Wealth quintile Lowest 47.9 16.5 5.7 26.1 2.6 1.3 100.0 872 0.8 Second 22.2 18.4 6.9 47.6 3.6 1.2 100.0 903 6.2 Middle 14.7 10.6 7.7 60.5 4.4 2.0 100.0 975 8.0 Fourth 8.7 6.8 5.3 59.2 12.7 7.3 100.0 1,060 8.9 Highest 2.2 3.5 2.8 54.2 21.0 16.3 100.0 1,204 9.9 Total 17.6 10.5 5.5 50.4 9.7 6.3 100.0 5,015 8.3 1 Completed 6 grade at the primary level 2 Completed 12 grade at the secondary level Characteristics of Survey Respondents | 27 Tables 3.2.1 and 3.2.2 also show that educational attainment varies greatly by urban-rural residence. Respondents in rural areas have substantially lower levels of educational attainment than their urban counterparts. Forty percent of women and 25 percent of men in the rural areas have never been to school in contrast to 16 percent of women and 8 percent of men in urban areas. Educational attainment is highest in Greater Accra and lowest in the Northern region. This is not surprising because Greater Accra is the most urbanized region in the country and has better educational opportunities. Nevertheless, even in this region, twice as many women as men have no education. Not surprisingly, there is a direct relationship between educational attainment and wealth. Women and men in the highest wealth quintile are most educated, in contrast to respondents with little or no education who are concentrated in the lowest wealth quintile. For example, 8 percent of women and 2 percent of men from the highest wealth quintile have no education, in contrast to 64 and 48 percent of women and men, respectively, in the lowest wealth quintile. Literacy is widely acknowledged as benefiting both the individual and society and, in particular among women, is associated with a number of positive outcomes, including intergenerational health and nutrition benefits. In the 2003 GDHS, literacy was ascertained by a respondent’s ability to read none, part, or all of a simple statement in any language that the respondent is likely to be able to read. The questions on literacy were asked only of respondents who had not attended school or attended primary or middle/JSS only. Respondents for whom no card with the required language was available, and those who were blind or visually impaired, are excluded from the estimation of percent literate, because their literacy cannot be gauged. Tables 3.3.1 and 3.3.2 show the percent distribution of women and men by level of schooling attended and by level of literacy. More than half (55 percent) of women and nearly three-quarters (73 percent) of men are literate, while 45 percent of women and 27 percent of men cannot read at all. As in the case of educational attainment, men are more likely to be literate than women. Forty percent of rural women compared with 71 percent of urban women are literate. Similarly, 62 percent of rural men compared with 87 percent of urban men are literate. Regional variations in the level of literacy are marked, ranging from a high of 75 percent among women in Greater Accra to a low of 14 percent among women in the Northern region. Almost nine in ten men in Greater Accra (89 percent) are literate, compared with one in three in the Northern and Upper East regions. There is a strong relationship between wealth and literacy levels. Women (83 percent) and men (95 percent) categorised in the highest wealth quintile are literate compared with only 20 percent of women and 37 percent of men in the lowest wealth quintile. Four in five women and three in five men in the lowest quintile cannot read at all. 28 | Characteristics of Survey Respondents Table 3.3.1 Literacy: women Percent distribution of women by level of schooling attended and by level of literacy, and percent literate, according to back- ground characteristics, Ghana 2003 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all No card/ visually impaired Missing Total Number of women Percent literate1 Age 15-19 64.3 4.7 3.4 27.6 0.0 0.0 100.0 1,148 72.4 20-24 60.2 1.2 1.0 37.4 0.0 0.1 100.0 1,012 62.5 25-29 49.0 0.6 0.3 49.9 0.2 0.0 100.0 951 50.0 30-34 45.4 1.1 1.1 52.1 0.0 0.3 100.0 802 47.7 35-39 41.4 0.4 0.9 57.2 0.1 0.0 100.0 722 42.7 40-44 47.1 1.0 0.7 51.0 0.2 0.0 100.0 579 48.9 45-49 41.4 1.4 1.8 55.4 0.0 0.0 100.0 477 44.6 Residence Urban 67.7 2.0 1.2 28.9 0.1 0.0 100.0 2,755 71.0 Rural 36.9 1.4 1.6 60.1 0.0 0.1 100.0 2,936 39.9 Region Western 53.2 1.1 2.3 43.4 0.0 0.0 100.0 553 56.6 Central 46.7 2.0 1.4 49.9 0.0 0.0 100.0 431 50.1 Greater Accra 71.2 2.2 1.0 25.5 0.1 0.0 100.0 942 74.5 Volta 52.0 2.6 1.5 43.9 0.0 0.0 100.0 492 56.1 Eastern 60.6 2.0 1.5 35.9 0.0 0.0 100.0 601 64.1 Ashanti 64.2 1.5 1.5 32.6 0.0 0.1 100.0 1,142 67.4 Brong Ahafo 49.7 2.0 2.4 45.6 0.3 0.0 100.0 569 54.3 Northern 12.7 0.6 0.3 86.0 0.1 0.4 100.0 499 13.6 Upper East 15.8 0.7 0.1 83.4 0.0 0.0 100.0 310 16.6 Upper West 21.5 1.7 1.1 75.7 0.0 0.0 100.0 153 24.3 Wealth quintile Lowest 17.4 1.0 1.1 80.2 0.1 0.2 100.0 970 19.6 Second 34.3 1.3 2.0 62.3 0.2 0.0 100.0 949 37.6 Middle 50.0 1.3 1.6 47.1 0.0 0.0 100.0 1,071 52.9 Fourth 60.5 2.4 1.7 35.1 0.1 0.1 100.0 1,245 64.8 Highest 80.0 2.1 0.7 17.2 0.0 0.0 100.0 1,457 82.8 Total 51.8 1.7 1.4 45.0 0.1 0.1 100.0 5,691 55.0 1 Refers to women who attended secondary school or higher and women who can read a whole sentence or part of a sen- tence Excludes from the denominator those women for whom no card with required language was available and those who are blind or visually impaired. Characteristics of Survey Respondents | 29 Table 3.3.2 Literacy: men Percent distribution of men by level of schooling attended and by level of literacy, and percent literate, according to back- ground characteristics, Ghana 2003 No schooling or primary school Background characteristic Secondary school or higher Can read a whole sentence Can read part of a sentence Cannot read at all No card/ visually impaired Missing Total Number of men Percent literate1 Age 15-19 65.2 8.5 5.6 20.6 0.0 0.1 100.0 1,107 79.4 20-24 73.8 1.6 3.6 20.6 0.3 0.0 100.0 684 79.3 25-29 71.7 0.9 1.6 25.7 0.0 0.1 100.0 754 74.2 30-34 67.8 1.1 2.0 27.9 1.0 0.1 100.0 633 71.8 35-39 60.2 1.8 2.1 34.8 1.1 0.0 100.0 498 64.8 40-44 67.7 0.8 1.7 28.8 0.8 0.2 100.0 412 70.9 45-49 59.5 2.2 2.0 35.3 1.2 0.0 100.0 441 64.3 50-54 63.2 3.0 1.6 31.2 1.0 0.0 100.0 294 68.5 55-59 56.0 2.9 2.2 38.1 0.8 0.0 100.0 192 61.6 Residence Urban 80.7 2.4 3.1 13.0 0.8 0.0 100.0 2,250 86.9 Rural 54.8 3.7 2.8 38.3 0.3 0.1 100.0 2,765 61.5 Region Western 72.4 4.1 2.7 20.4 0.3 0.0 100.0 476 79.5 Central 70.1 2.6 2.5 24.9 0.0 0.0 100.0 370 75.1 Greater Accra 86.3 1.2 1.0 10.9 0.6 0.0 100.0 733 89.1 Volta 72.5 5.0 2.5 19.9 0.2 0.0 100.0 440 80.1 Eastern 74.8 3.4 3.2 18.7 0.0 0.0 100.0 539 81.3 Ashanti 78.6 2.4 3.0 15.9 0.1 0.0 100.0 956 84.1 Brong Ahafo 68.4 3.8 4.6 23.2 0.0 0.0 100.0 528 76.8 Northern 24.9 4.5 4.1 65.8 0.2 0.6 100.0 527 33.7 Upper East 26.7 2.2 2.9 62.5 5.8 0.0 100.0 317 33.7 Upper West 34.5 2.6 4.4 58.2 0.2 0.0 100.0 130 41.6 Wealth quintile Lowest 29.9 3.7 3.1 62.3 0.6 0.3 100.0 872 37.1 Second 52.5 5.4 3.5 38.3 0.3 0.0 100.0 903 61.6 Middle 67.0 3.1 3.5 25.9 0.6 0.0 100.0 975 74.0 Fourth 79.2 2.8 3.2 14.3 0.5 0.0 100.0 1,060 85.6 Highest 91.5 1.3 1.8 4.9 0.6 0.0 100.0 1,204 95.1 Total 66.4 3.1 2.9 27.0 0.5 0.1 100.0 5,015 72.9 1 Refers to men who attended secondary school or higher and men who can read a whole sentence or part of a sentence. Excludes from the denominator those men for whom no card with required language was available and those who are blind or visually impaired. 3.3 ACCESS TO MASS MEDIA The 2003 GDHS collected information on respondents’ exposure to both broadcast and print media. This information is a means of assessing the potential effectiveness of using these media to broadcast messages on such important topics as reproductive health and HIV/AIDS. Tables 3.4.1 and 3.4.2 show that access to mass media, especially the broadcast media, is generally high in Ghana. Seventy-four percent of women and 89 percent of men listen to the radio at least once a week, and 44 percent of women and 51 percent of men watch television at least once a week. Exposure to the print 30 | Characteristics of Survey Respondents media is relatively low. Twelve percent of women and 28 percent of men read a newspaper at least once a week. Men are twice as likely as women to be exposed to all three media sources (23 and 10 percent, respectively). Media exposure is higher among younger women (age 15-24) than older women (25 years and above). However, among men, exposure is lowest among those age 15-19 and highest among those age 20-29. Urban women and men tend to have greater exposure to all three media sources than their rural counterparts. The high level of illiteracy among rural women is reflected in the lower proportion of these women (4 percent) exposed to the print media than urban women (21 percent). Table 3.4.1 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, Ghana 2003 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 19.6 54.5 73.0 14.7 18.5 1,148 20-24 14.3 51.9 78.0 11.4 16.5 1,012 25-29 9.6 43.7 75.7 8.3 20.5 951 30-34 7.7 36.8 72.0 6.7 25.9 802 35-39 9.3 35.9 71.0 7.8 25.2 722 40-44 9.9 37.3 74.2 7.6 22.4 579 45-49 11.7 33.6 73.3 9.3 23.7 477 Residence Urban 21.3 66.0 80.9 17.6 12.2 2,755 Rural 4.0 23.1 67.7 2.6 29.6 2,936 Region Western 16.7 45.5 76.0 12.8 19.4 553 Central 7.5 35.2 61.2 4.7 31.5 431 Greater Accra 26.7 74.4 85.4 22.6 8.3 942 Volta 10.9 26.5 73.9 6.4 21.2 492 Eastern 10.3 44.5 80.4 8.8 15.9 601 Ashanti 12.9 54.0 81.6 10.6 13.2 1,142 Brong Ahafo 7.0 40.9 80.0 5.4 16.1 569 Northern 3.0 15.6 45.9 2.5 51.4 499 Upper East 2.3 16.8 62.6 2.1 34.7 310 Upper West 2.4 10.7 46.7 1.1 51.7 153 Education No education 0.0 18.4 56.7 0.0 40.1 1,608 Primary 1.4 34.4 70.6 0.6 24.8 1,135 Middle/JSS 12.6 56.1 82.8 9.4 11.8 2,279 Secondary+ 60.1 79.7 92.2 51.0 1.7 669 Wealth quintile Lowest 1.4 7.2 51.2 0.1 47.1 970 Second 1.9 16.5 69.0 1.2 29.0 949 Middle 4.3 33.4 74.2 2.5 22.0 1,071 Fourth 10.0 53.6 79.4 6.5 14.2 1,245 Highest 34.5 85.5 88.1 30.3 4.2 1,457 Total 12.4 43.9 74.1 9.9 21.2 5,691 Characteristics of Survey Respondents | 31 Table 3.4.2 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, Ghana 2003 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 21.8 55.2 84.0 18.5 12.2 1,107 20-24 30.4 60.3 91.8 24.5 5.0 684 25-29 33.4 57.2 93.2 28.1 4.9 754 30-34 30.0 48.4 90.4 24.0 7.1 633 35-39 28.7 44.1 89.8 22.7 8.6 498 40-44 31.0 41.4 89.7 23.0 8.9 412 45-49 30.0 43.9 87.5 22.4 9.0 441 50-54 25.3 38.8 90.7 19.6 7.9 294 55-59 30.0 41.7 92.8 24.6 7.2 192 Residence Urban 46.4 76.3 92.4 40.7 3.7 2,250 Rural 13.8 29.7 86.9 8.4 11.7 2,765 Region Western 31.4 52.6 90.5 21.7 4.7 476 Central 21.4 37.3 82.3 12.2 13.2 370 Greater Accra 55.9 78.8 95.2 49.6 2.1 733 Volta 20.7 33.5 89.0 13.6 8.5 440 Eastern 27.2 53.6 95.9 23.3 2.8 539 Ashanti 35.8 66.8 97.5 30.3 1.6 956 Brong Ahafo 24.4 51.5 94.0 20.4 4.6 528 Northern 8.0 22.6 67.7 5.6 28.7 527 Upper East 8.1 28.9 80.9 6.1 14.5 317 Upper West 9.0 12.8 74.9 3.5 24.3 130 Education No education 0.4 17.8 75.4 0.4 22.9 881 Primary 3.6 34.0 82.0 2.2 14.5 803 Middle/JSS 25.5 56.2 94.0 19.7 3.5 2,165 Secondary+ 72.2 76.7 96.4 60.0 1.1 1,165 Wealth quintile Lowest 4.4 10.4 73.8 1.3 24.9 872 Second 8.9 20.4 89.2 3.7 9.7 903 Middle 16.0 44.5 91.6 9.5 5.0 975 Fourth 37.1 67.9 92.2 28.3 4.2 1,060 Highest 63.0 92.3 96.3 58.9 0.7 1,204 Total 28.4 50.6 89.3 22.9 8.1 5,015 More than one in five women and one in two men in Greater Accra are exposed to all three media sources. Media exposure is markedly lower in the three northern regions and is especially low in the Upper West, where only 1 percent of women and 4 percent of men are exposed to all three media sources. Exposure to mass media is closely related to the level of education of respondents. Half of all women and three-fifths of all men with at least secondary education are exposed to all three media sources. Three- fifths and nearly three-quarters of highly educated women and men read a newspaper at least once a week. 32 | Characteristics of Survey Respondents There is a high correlation between wealth and media exposure, with the gap between those in the highest quintile and all other quintiles being especially wide. For example, 30 percent of women in the highest wealth quintile are exposed to all three media compared with 7 percent or less of women in the other four quintiles. Since 1998, the proportions of both women and men who have no media exposure have declined markedly, from 30 percent to 21 percent among women and 15 percent to 8 percent among men (GSS and MI, 1999). However, the proportion of women and men who report reading the newspaper and watching television at least once a week has declined, while the proportion who reported listening to the radio has increased. Some of these reported differences by type of media source may be due to a change in the way the questions were worded between the two surveys. 3.4 EMPLOYMENT 3.4.1 Employment Status Tables 3.5.1 and 3.5.2 present the percent distribution of women and men by employment status, according to background characteristics. There is little difference in the overall employment status of women and men. Three-fourths of women and men reported being currently employed, while 3 percent reported being employed in the 12 months preceding the survey, but not employed at the time the survey was fielded. About one-fifth of women and men were not employed in the 12 months prior to the survey. Current employment increases with age from 33 and 26 percent among the youngest cohort of women and men, respectively, to about 95 percent among the oldest cohort of respondents. Low current employment among young women and men may be due to the fact that a proportionately larger number of young people are still in school. Currently and formerly married women are more likely to be currently employed than never married women. However, among men, married men are more likely to be employed than formerly married men and those never married. Current employment rises with the number of living children. For example, about one in two respondents with no children are currently employed compared with 94 percent of women and 98 percent of men with five or more children. Current employment is also higher among rural respondents than urban respondents. This could be due to the fact that it is easier to find employment in the largely informal sector in the rural areas than in urban areas. Unemployment could also be higher in the urban areas because there is greater demand for skilled labour, which is harder to acquire. There is little variation in employment status of respondents by region. More than 80 percent of women in the Upper West and Volta regions and men in the Upper East and Northern regions are currently employed compared with about 70-80 percent of respondents in all other regions. Current employment is inversely related to education, falling from 86 percent among women with no education to 60 percent among women with at least secondary education. The corresponding data for men is 97 and 67 percent, respectively. A similar pattern is seen by wealth quintile for both women and men. Characteristics of Survey Respondents | 33 Table 3.5.1 Employment status: women Percent distribution of women by employment status, according to background characteristics, Ghana 2003 Employed in the 12 months preceding the survey Background Characteristic Currently employed Not currently employed Not employed in the 12 months preced- ing the survey Total Number of women Age 15-19 33.0 3.3 63.6 100.0 1,148 20-24 66.5 4.7 28.8 100.0 1,012 25-29 86.8 2.6 10.5 100.0 951 30-34 90.4 2.1 7.4 100.0 802 35-39 94.7 1.5 3.6 100.0 722 40-44 92.3 1.9 5.8 100.0 579 45-49 94.8 0.5 4.7 100.0 477 Marital status Never married 42.0 3.8 54.2 100.0 1,616 Married or living together 88.3 2.2 9.4 100.0 3,549 Divorced/separated/widowed 87.7 2.4 10.0 100.0 526 Number of living children 0 48.2 3.7 48.1 100.0 1,872 1-2 82.2 3.3 14.6 100.0 1,602 3-4 91.3 1.5 7.0 100.0 1,227 5+ 94.3 1.3 4.4 100.0 990 Residence Urban 69.2 3.4 27.4 100.0 2,755 Rural 80.6 2.0 17.3 100.0 2,936 Region Western 72.3 1.1 26.6 100.0 553 Central 79.1 5.9 15.0 100.0 431 Greater Accra 71.1 4.2 24.7 100.0 942 Volta 82.2 1.2 16.6 100.0 492 Eastern 73.0 1.7 25.3 100.0 601 Ashanti 71.3 4.0 24.6 100.0 1,142 Brong Ahafo 79.2 1.1 19.7 100.0 569 Northern 79.5 0.8 19.7 100.0 499 Upper East 73.7 2.5 23.7 100.0 310 Upper West 85.1 1.0 13.9 100.0 153 Education No education 86.2 1.6 12.1 100.0 1,608 Primary 76.5 3.3 20.1 100.0 1,135 Middle/JSS 70.9 2.9 26.2 100.0 2,279 Secondary+ 60.3 3.3 36.4 100.0 669 Wealth quintile Lowest 83.5 1.3 15.0 100.0 970 Second 82.4 2.1 15.5 100.0 949 Middle 76.5 1.8 21.8 100.0 1,071 Fourth 71.9 3.8 24.3 100.0 1,245 Highest 66.4 3.7 29.9 100.0 1,457 Total 75.1 2.7 22.2 100.0 5,691 34 | Characteristics of Survey Respondents Table 3.5.2 Employment status: men Percent distribution of men by employment status, according to background characteristics, Ghana 2003 Employed in the 12 months preceding the survey Background characteristic Currently employed Not currently employed Not em- ployed in the 12 months pre- ceding the survey Missing/ don't know Total Number of men Age 15-19 26.0 4.3 68.3 1.5 100.0 1,107 20-24 66.8 5.5 27.4 0.3 100.0 684 25-29 88.8 4.6 6.6 0.0 100.0 754 30-34 94.8 2.6 2.4 0.2 100.0 633 35-39 97.0 1.6 1.4 0.0 100.0 498 40-44 97.7 0.9 1.4 0.0 100.0 412 45-49 96.7 1.4 1.9 0.0 100.0 441 50-54 96.7 2.1 1.2 0.0 100.0 294 55-59 94.4 1.0 4.6 0.0 100.0 192 Marital status Never married 45.9 5.3 47.9 0.9 100.0 2,042 Married or living together 96.9 1.6 1.5 0.0 100.0 2,671 Divorced/separated/widowed 88.1 4.5 7.4 0.0 100.0 302 Number of living children 0 50.8 5.3 43.1 0.8 100.0 2,300 1-2 95.1 2.2 2.7 0.0 100.0 981 3-4 96.4 1.6 1.8 0.1 100.0 816 5+ 98.3 0.8 0.9 0.0 100.0 917 Residence Urban 69.4 4.1 25.9 0.6 100.0 2,250 Rural 80.6 2.6 16.6 0.3 100.0 2,765 Region Western 73.3 3.8 22.7 0.2 100.0 476 Central 74.4 2.9 21.4 1.3 100.0 370 Greater Accra 72.3 5.2 21.9 0.6 100.0 733 Volta 73.3 2.7 23.6 0.4 100.0 440 Eastern 76.9 2.0 20.8 0.2 100.0 539 Ashanti 74.6 2.7 22.2 0.5 100.0 956 Brong Ahafo 70.6 3.3 25.8 0.3 100.0 528 Northern 85.6 2.7 11.6 0.0 100.0 527 Upper East 82.5 2.5 15.0 0.0 100.0 317 Upper West 78.3 6.2 15.4 0.0 100.0 130 Education No education 97.4 1.2 1.4 0.0 100.0 881 Primary 71.6 2.6 25.7 0.2 100.0 803 Middle/JSS 73.0 3.1 23.3 0.5 100.0 2,165 Secondary + 66.6 5.5 27.3 0.6 100.0 1,165 Wealth quintile Lowest 85.6 3.0 11.4 0.0 100.0 872 Second 80.4 2.1 17.4 0.1 100.0 903 Middle 74.6 2.4 22.3 0.7 100.0 975 Fourth 69.8 4.1 25.9 0.2 100.0 1,060 Highest 70.7 4.2 24.3 0.8 100.0 1,204 Total 75.6 3.3 20.8 0.4 100.0 5,015 Characteristics of Survey Respondents | 35 3.4.2 Occupation Tables 3.6.1 and 3.6.2 show data on employed women and men by their occupation, according to background characteristics. More than one-third of working women (36 percent) and half of men (50 percent) are employed in the agricultural (Figure 3.1). Four times as many women (42 percent) as men (11 percent) work in sales and services. Sixteen percent of employed women and 23 percent of employed men are skilled manual workers. Three times as many men (10 percent) as women (3 percent) are engaged in professional, technical, and managerial positions. Occupation varies by age groups. Among women, the proportion engaged in agriculture or in professional, technical, or managerial occupations, increases with age. For example, one in four working women age 15-19 are in agricultural occupations compared with nearly one in two women age 45-49. On the other hand, the proportion engaged in sales and services and in skilled manual labour decreases with age. A similar pattern is observed for men for skilled manual labour only. A higher proportion of never-married women than ever-married women are engaged in most occupations with the exception of agriculture, which accounts for two in five currently married women who are working compared with more than one in four formerly married women and 12 percent of never- married women. There is no clear pattern in occupation by marital status among men. Among working men, about half of those who are currently or formerly married are engaged in the agricultural sector compared to two in five never-married men. Three-tenths of never-married men are engaged as skilled manual labour, compared with one-fifth of ever married men. There is a direct relationship between the number of living children and agricultural occupation among both women and men. The proportion engaged in agriculture increases with the number of living children. Among women, the proportion engaged in all other occupations is higher among those with no children than among those with one or more children. Not surprisingly, most working women and men in rural areas are engaged in the agricultural sector, in contrast to women and men in urban areas, who are mostly engaged in sales and service and skilled manual work. Agriculture is the predominant occupation among women in the Northern, Upper East, Brong Ahafo, and Upper West regions. In addition to these regions, the majority of men in the Eastern, Volta, and Central regions are also employed in the agricultural sector. In contrast, the majority of working women living in Greater Accra, Ashanti, and Eastern regions are in sales and service jobs. Among working men, the highest proportions engaged in the professional, technical and managerial, clerical, sales and service, and skilled manual work are from Greater Accra. Education is related to the type of occupation of respondents. Fifty-nine percent of working women and 83 percent of working men who have never been to school are engaged in agriculture. On the other hand, the majority of women and men with secondary or higher education are employed in non- agricultural occupations. Women and men in the lowest quintile are predominantly engaged in agriculture, while those in the highest wealth quintile are mostly engaged in sales and service, skilled manual, or professional, technical, and managerial work. 36 | Characteristics of Survey Respondents Table 3.6.1 Occupation: women Percent distribution of women employed in the 12 months preceding the survey by occupation, according to background characteristics, Ghana 2003 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agriculture Missing Total Number of women Age 15-19 0.6 1.5 50.2 21.2 0.0 25.9 0.5 100.0 417 20-24 1.9 3.8 46.2 20.8 0.0 26.6 0.8 100.0 721 25-29 2.4 1.2 43.6 18.3 0.8 33.2 0.6 100.0 851 30-34 3.3 0.9 43.3 14.2 0.7 37.1 0.5 100.0 743 35-39 4.1 1.2 40.5 14.0 0.1 39.1 0.9 100.0 694 40-44 4.1 1.0 36.5 13.9 0.4 43.2 0.9 100.0 545 45-49 5.6 1.0 31.8 12.0 0.1 48.0 1.6 100.0 455 Marital status Never married 4.0 4.7 55.2 22.3 0.6 12.2 0.9 100.0 740 Married or living together 2.9 1.0 38.0 14.6 0.3 42.4 0.8 100.0 3,213 Dvorced/separated/widowed 2.9 0.6 48.4 19.4 0.3 27.5 0.8 100.0 473 Number of living children 0 4.8 4.5 52.5 21.9 0.4 15.5 0.5 100.0 971 1-2 2.3 0.9 48.6 17.7 0.2 29.6 0.7 100.0 1,369 3-4 3.7 0.9 38.8 14.8 0.5 40.8 0.6 100.0 1,140 5+ 1.9 0.2 25.8 10.9 0.3 59.5 1.4 100.0 947 Residence Urban 4.5 3.1 61.3 21.0 0.3 9.4 0.4 100.0 2,001 Rural 1.9 0.2 26.1 12.7 0.4 57.5 1.1 100.0 2,425 Region Western 4.4 1.4 31.6 21.6 0.0 39.8 1.2 100.0 406 Central 1.9 0.3 40.7 19.6 0.0 37.5 0.0 100.0 366 Greater Accra 5.0 4.9 60.0 24.3 0.0 5.3 0.4 100.0 709 Volta 2.3 0.7 36.8 17.8 0.3 41.5 0.7 100.0 410 Eastern 4.7 1.5 50.8 9.4 0.3 33.3 0.0 100.0 449 Ashanti 2.3 1.5 52.3 13.1 0.5 28.2 2.1 100.0 860 Brong Ahafo 2.3 0.8 34.2 8.6 0.0 53.3 0.8 100.0 457 Northern 2.5 0.0 18.1 17.1 1.2 60.5 0.5 100.0 401 Upper East 2.0 0.0 30.6 12.3 0.0 55.1 0.0 100.0 236 Upper West 0.6 0.5 21.1 23.6 3.1 51.1 0.0 100.0 131 Education No education 0.1 0.0 24.4 14.8 0.6 59.1 0.9 100.0 1,412 Primary 0.3 0.1 41.3 15.7 0.1 41.6 0.8 100.0 906 Middle/JSS 1.7 0.6 56.8 18.8 0.3 21.2 0.8 100.0 1,683 Secondary+ 24.7 13.6 43.7 14.1 0.0 3.3 0.6 100.0 425 Wealth quintile Lowest 0.3 0.0 14.9 11.9 1.0 71.5 0.5 100.0 823 Second 1.0 0.0 24.3 11.0 0.2 63.1 0.5 100.0 802 Middle 1.6 0.4 36.8 16.0 0.0 43.7 1.4 100.0 838 Fourth 3.4 1.4 61.0 21.8 0.6 10.9 1.0 100.0 943 Highest 8.0 5.1 64.7 19.7 0.0 1.8 0.7 100.0 1,020 Total 3.1 1.5 42.0 16.4 0.3 35.8 0.8 100.0 4,426 Characteristics of Survey Respondents | 37 Table 3.6.2 Occupation: men Percent distribution of men employed in the 12 months preceding the survey by occupation, according to background characteristics, Ghana 2003 Background characteristic Professional/ technical/ managerial Clerical Sales and services Skilled manual Unskilled manual Agriculture Missing Total Number of men Age 15-19 2.6 0.3 8.1 25.2 2.0 57.6 4.1 100.0 335 20-24 7.2 2.3 11.1 33.0 3.0 40.1 3.2 100.0 494 25-29 10.5 2.2 12.7 27.8 1.7 41.9 3.2 100.0 705 30-34 8.7 2.6 13.5 27.3 1.0 46.2 0.6 100.0 617 35-39 10.5 1.4 10.9 20.9 1.9 51.6 2.8 100.0 491 40-44 13.7 2.8 7.3 16.1 0.4 59.6 0.3 100.0 406 45-49 14.1 2.1 8.3 18.8 0.5 55.0 1.3 100.0 433 50-54 13.1 2.3 12.5 13.7 1.4 55.4 1.6 100.0 290 55-59 17.7 1.0 13.9 7.9 1.2 57.2 1.1 100.0 183 Marital status Never married 9.6 2.6 11.6 29.6 2.2 40.1 4.3 100.0 1,045 Married or living together 11.1 1.8 10.4 20.9 0.9 53.7 1.2 100.0 2,630 Divorced/separated/widowed 6.7 2.3 14.6 20.7 4.1 49.5 2.1 100.0 279 Number of living children 0 9.9 2.5 12.2 28.6 2.5 40.5 3.8 100.0 1,291 1-2 10.1 1.6 12.2 27.2 1.5 45.2 2.1 100.0 955 3-4 10.7 2.0 11.9 21.0 0.7 53.0 0.7 100.0 800 5+ 11.1 1.8 7.3 13.1 0.6 65.2 1.0 100.0 909 Residence Urban 18.0 3.8 21.4 36.4 2.1 14.9 3.5 100.0 1,654 Rural 4.9 0.7 3.5 13.7 1.1 74.9 1.1 100.0 2,300 Region Western 7.2 2.4 11.4 26.2 1.8 45.8 5.0 100.0 367 Central 9.5 1.4 4.5 29.5 1.0 52.0 2.0 100.0 286 Greater Accra 15.3 6.1 25.2 36.6 2.6 10.6 3.7 100.0 568 Volta 11.8 1.1 4.9 20.7 1.2 58.2 2.1 100.0 334 Eastern 14.0 1.3 9.0 21.7 0.5 52.5 0.9 100.0 425 Ashanti 12.6 1.8 11.5 31.7 3.0 37.6 1.9 100.0 739 Brong Ahafo 9.8 1.4 8.0 12.5 0.9 65.6 1.7 100.0 390 Northern 4.5 0.4 5.6 8.4 0.4 80.2 0.4 100.0 465 Upper East 4.0 0.7 12.2 13.0 0.0 69.0 1.1 100.0 270 Upper West 8.1 0.0 6.8 7.5 1.0 75.3 1.3 100.0 110 Education No education 0.5 0.3 7.4 6.5 0.8 83.1 1.3 100.0 869 Primary 0.8 0.7 4.8 22.1 0.7 67.1 3.8 100.0 595 Middle/JSS 4.8 1.7 11.3 33.7 2.5 43.7 2.3 100.0 1,649 Secondary+ 38.4 5.3 18.6 20.4 0.8 15.1 1.3 100.0 840 Wealth quintile Lowest 2.5 0.1 2.0 5.4 0.3 88.8 0.9 100.0 772 Second 2.8 0.3 2.6 9.3 0.7 82.8 1.5 100.0 745 Middle 5.6 0.7 6.0 20.2 1.7 63.3 2.6 100.0 751 Fourth 14.9 3.2 15.5 40.4 2.5 20.2 3.2 100.0 784 Highest 23.5 5.2 26.0 37.3 2.0 3.8 2.2 100.0 902 Total 10.4 2.0 11.0 23.2 1.5 49.8 2.1 100.0 3,954 38 | Characteristics of Survey Respondents Figure 3.1 Occupation of Women Age 15-49 and Men Age 15-59 GDHS 2003 Agriculture 36% Unskilled manual 2% Unskilled manual <1% Skilled manual 16% Agriculture 50% Professional/ technical/ managerial 3% Missing 2% Missing <1% Professional/ technical/ managerial 10% Sales and services 42% Clerical 2% Clerical 2% Sales and Skilled manual 23% services 11% 3.4.3 Type of Employer, Form of Earnings, and Continuity of Employment Tables 3.7.1, 3.7.2, and Figure 3.2, present data on women and men by type of employment. The majority of women and men (about 60 percent) who work receive cash earnings. A significant proportion of women (25 percent) and men (30 percent) receive earnings in cash and in-kind. Eight percent of women and 5 percent of men are not paid at all. Three-quarters of women in both agricultural and non-agricultural occupations are self-employed, 14 percent are employed by a non-family member, and 10 percent are employed by a family member. Three-quarters of employed women work all year, while one in five works seasonally. Most women in both agricultural and non-agricultural occupations work all year; however, one in three women who work in agriculture and 13 percent of women engaged in non-agricultural occupations work seasonally. Similar information was not collected for men. Characteristics of Survey Respondents | 39 Table 3.7.1 Type of employment: women 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 non-agricultural), Ghana 2003 Employment characteristic Agricultural work Non- agricultural work Total Type of earnings Cash only 30.9 78.2 61.0 Cash and in-kind 48.4 12.2 25.3 In-kind only 12.2 2.4 6.0 Not paid 8.5 7.3 7.7 Total 100.0 100.0 100.0 Type of employer Employed by family member 17.3 5.5 9.8 Employed by non-family member 5.5 18.1 13.9 Self-employed 77.0 75.8 75.9 Missing 0.2 0.6 0.5 Total 100.0 100.0 100.0 Continuity of employment All year 65.3 83.1 76.6 Seasonal 34.0 12.9 20.5 Occasional 0.6 3.9 2.8 Missing 0.1 0.0 0.1 Total 100.0 100.0 100.0 Number of women 1,583 2,809 4,426 Note: Total includes 30 women with missing information on type of employment who are not shown separately. Table 3.7.2 Type of employment: men Percent distribution of men employed in the 12 months preceding the survey by type of earnings, according to type of employment (agricul- tural or non-agricultural), Ghana 2003 Type of Earnings Agricultural work Non- agricultural work Total Cash only 36.0 84.3 60.2 Cash and in-kind 51.6 8.6 29.9 In-kind only 7.5 1.5 4.4 Not paid 5.0 5.7 5.2 Missing 0.0 0.0 0.2 Total 100.0 100.0 100.0 Number of men 1,970 1,901 3,954 Note: Total includes 78 men with missing information on type of employment who are not shown separately. 40 | Characteristics of Survey Respondents Figure 3.2 Type of Earnings of Employed Women Age 15-49 and Men Age 15-59 Women Men Cash only 61% Not paid 8% Cash and in-kind 25% Not paid 5% Cash only 60% In-kind only 6% Cash and in-kind 30% In-kind only 4% Note: Total may not add to 100 due to rounding. GDHS 2003 3.4.4 Control Over Earnings and Women’s Contribution to Household Expenditures Women’s autonomy is dependent not only on their access to income but also on the amount of control they have over their earnings. Employed women who earn cash were asked about who mainly decides how their income is used. Table 3.8 shows that women in Ghana have considerable autonomy over the use of their earnings. Nearly three-fourths of women who earn cash report that they are solely responsible for decisions on the use of their earnings, while 18 percent report that they jointly decide how the money should be spent either with their husband or someone else. Almost one in ten women stated that they have no say in how their earnings are spent. Sole decisionmaking rises with age. One in five never-married women have no say in how their earnings are used, while a similar proportion of currently married women report that they make joint decision. There is little difference in sole decisionmaking by the number of children, but joint decisionmaking rises with the number of children women have. On the other hand, there is an inverse relationship between the number of children women have and the percentage who have no say in how their earnings are spent. For example, twice as many women who have no children have no say in how their earnings are spent compared with women with one or more children. Autonomy over cash earnings is higher among urban than rural women, and relatively high among women residing in Greater Accra, Western, and Upper East regions, and lowest among women in the Central region. Education and wealth exert only a small influence on control over earnings. For example, 78 percent of women with at least secondary education or in the highest wealth quintile compared with 70 percent of women with no education or in the lowest wealth quintile have the sole say in how their earnings are spent. Characteristics of Survey Respondents | 41 Table 3.8 Decision on use of earnings and contribution of earnings to household expenditures Percent distribution of women employed in the 12 months preceding the survey receiving cash earnings by person who decides how earn- ings are to be used and by proportion of household expenditures met by earnings, according to background characteristics, Ghana 2003 Person who decides how earnings are used Proportion of household expenditures met by earnings Background characteristic Woman only Jointly1 Someone else only2 Total Almost none/none Less than half About half or more than half All Total Number of women Age 15-19 62.6 9.3 28.1 100.0 47.6 27.2 21.6 3.5 100.0 228 20-24 69.9 16.7 13.3 100.0 19.9 34.0 39.3 6.9 100.0 584 25-29 72.1 18.3 9.5 100.0 11.9 37.0 44.2 6.8 100.0 776 30-34 75.4 17.3 7.3 100.0 9.1 26.3 52.1 12.5 100.0 682 35-39 71.2 20.9 7.7 100.0 4.4 26.0 52.6 16.8 100.0 638 40-44 75.2 19.1 5.7 100.0 4.6 22.1 56.1 17.1 100.0 497 45-49 81.0 15.5 3.2 100.0 3.4 20.9 60.8 14.8 100.0 417 Marital status Never married 75.5 5.4 19.1 100.0 37.3 26.8 27.9 8.1 100.0 498 Married or living together 69.1 22.0 8.8 100.0 7.8 31.6 51.4 9.2 100.0 2,887 Divorced/separated/widowed 96.2 2.1 1.7 100.0 8.1 10.2 50.7 31.1 100.0 436 Number of living children 0 75.2 8.4 16.4 100.0 30.5 28.3 32.4 8.7 100.0 695 1-2 74.9 16.3 8.8 100.0 10.5 32.0 48.3 9.2 100.0 1,246 3-4 71.2 21.4 7.4 100.0 6.4 26.5 54.0 13.1 100.0 1,027 5+ 70.8 22.4 6.7 100.0 4.3 26.0 54.3 15.3 100.0 853 Residence Urban 78.0 12.9 9.0 100.0 14.5 27.9 46.1 11.5 100.0 1,782 Rural 68.7 21.7 9.6 100.0 9.1 29.1 50.2 11.5 100.0 2,039 Region Western 82.7 9.6 7.7 100.0 12.6 25.1 55.4 6.9 100.0 371 Central 54.0 34.9 11.1 100.0 11.7 21.8 45.4 21.0 100.0 308 Greater Accra 84.8 4.6 10.6 100.0 12.9 24.0 49.7 13.4 100.0 616 Volta 76.1 10.9 13.0 100.0 6.4 22.6 56.8 14.3 100.0 378 Eastern 77.3 19.0 3.7 100.0 7.5 14.9 60.0 17.7 100.0 418 Ashanti 66.1 25.3 8.4 100.0 16.4 34.1 42.3 7.2 100.0 791 Brong Ahafo 59.4 33.8 6.5 100.0 12.2 28.5 45.7 13.3 100.0 390 Northern 79.0 9.8 11.2 100.0 4.6 47.6 41.9 5.9 100.0 310 Upper East 82.1 3.7 14.2 100.0 19.2 45.5 30.6 4.7 100.0 165 Upper West 66.3 16.0 17.7 100.0 4.6 42.7 46.6 6.2 100.0 74 Education No education 70.3 18.9 10.8 100.0 8.3 32.9 49.1 9.7 100.0 1,144 Primary 72.7 18.1 9.1 100.0 12.0 25.0 50.5 12.4 100.0 785 Middle/JSS 73.9 17.5 8.3 100.0 13.5 26.4 46.9 13.1 100.0 1,499 Secondary+ 78.0 13.0 9.1 100.0 13.4 30.9 46.8 8.9 100.0 393 Wealth quintile Lowest 70.2 17.7 12.1 100.0 8.1 33.5 50.4 8.0 100.0 610 Second 63.8 24.6 11.4 100.0 10.3 29.0 49.5 11.0 100.0 707 Middle 74.6 17.9 7.5 100.0 10.7 27.1 48.0 14.3 100.0 746 Fourth 75.6 15.2 9.0 100.0 12.9 25.8 49.4 11.9 100.0 837 Highest 78.3 14.0 7.7 100.0 14.7 28.5 45.2 11.6 100.0 921 Total 73.0 17.6 9.3 100.0 11.6 28.5 48.3 11.5 100.0 3,821 Note: Percentages may not add to 100 due to the exclusion of women with missing information. 1 With husband or someone else 2 Includes husband 42 | Characteristics of Survey Respondents Information on the contribution of respondent’s income to household expenditures was also gathered in the 2003 GDHS. It is expected that employment and earnings are more likely to empower women if their earnings are important for meeting the needs of their households. However, often women’s income is so small that it can barely meet household needs. Table 3.8 shows that the earnings of very young women (age 15-19) are less likely to contribute a major share of household expenditures than those of older women (20-49). Not surprisingly, working women who are divorced, separated, or widowed tend to contribute to a major portion of household expenditure. Women’s contribution to household expenditure increases with the number of children they have. There is little difference between urban and rural women’s contribution to household expenditure. The majority of working women in all regions except Upper East, Northern, and Ashanti regions meet half or more or all of household expenditure with their earnings. There is no clear relationship between women’s contribution to household expenditure and education or wealth. Table 3.9 shows the relationship between women’s control over their earnings and their contribution to household expenditure by marital status. The table shows that women who are not currently married (that is, those who have never married, or are divorced, separated, or widowed) are somewhat more likely (85 percent) than currently married women (69 percent) to make sole decisions on how their earnings are spent. Nine percent of married women do not have a say in how their earnings are spent, with 8 percent reporting that their husbands alone decide on how their earnings are spent. On the other hand, 11 percent of unmarried women have no say in how their earnings are used. The data also show that among unmarried women earning cash, the greater a woman’s contribution to household expenditure, the more likely is she to make sole decisions on how her earnings are spent. Nevertheless, 8 percent of married women who contribute all of their earnings to household expenditure report that their husband or someone else makes sole decisions on how their earnings are spent, whereas only 3 percent of unmarried women who contribute all their earnings to household expenditure report that they have no say in how it is spent. Table 3.9 Women's control over earnings Percent distribution of women who received cash earnings for work in the past 12 months by person who decides how earnings are used, and the proportion of household expenditures met by earnings, according to current marital status, Ghana 2003 Currently married or living together Not married1 Contribution to household expenditures Woman only Jointly with husband Jointly with someone else Husband only Some- one else only Total Number of women Woman only Jointly with someone else Someone else only Total Number of women Almost none/ none 74.5 15.8 0.5 6.8 2.4 100.0 224 71.1 7.1 21.9 100.0 221 Less than half 66.9 22.8 0.3 9.3 0.6 100.0 912 84.0 3.6 12.4 100.0 178 Half or more 68.2 22.6 0.9 7.6 0.7 100.0 1,485 89.6 2.9 7.5 100.0 360 All 77.2 15.3 0.0 7.1 0.4 100.0 265 95.1 2.2 2.6 100.0 176 Total 69.1 21.4 0.6 8.0 0.8 100.0 2,887 85.2 3.9 10.9 100.0 934 Note: Total includes 1 woman (currently married or living together) with missing information on contribution to household expendi- tures, who is not shown separately. Percentages for currently married women may not add to 100 due to exclusion of women with missing information. 1 Never-married, divorced, separated, or widowed women Characteristics of Survey Respondents | 43 3.5 WOMEN’S EMPOWERMENT In addition to information on women’s education, employment status, and control over earnings, the 2003 GDHS collected information from both women and men on other measures of women’s empowerment. Respondents were asked about women’s role in household decisionmaking, their acceptance of wife-beating, and their opinions about whether a wife can deny sex to her husband for certain specified reasons. Such information provides insight into women’s control over their environment and their attitudes towards gender roles, both of which are relevant to understanding women’s ability to make independent decisions about their own health care and that of their children’s. 3.5.1 Women’s Participation in Decisionmaking In order to assess women’s weight in household decisionmaking, women were asked who in their family usually has the final say on five different types of decisions, namely: their own health care, large household purchases, daily household purchases, visits to family or relatives, and what food to cook each day. The percent distribution of women according to the person who usually has the final say in different decisions is shown in Table 3.10. The data are presented separately for women who are currently married and women who have never married or who are divorced, separated, or widowed. Table 3.10 Women's participation in decisionmaking Percent distribution of women by person who has the final say in making specific decisions, according to current marital status and type of decision, Ghana 2003 Currently married or living together Not married1 Decision Woman only Jointly with hus- band Jointly with some- one else Hus- band only Some- one else only Decision not made/not applica- ble Total Num- ber of women Woman only Jointly with some- one else Some- one else only Decision not made/not applicable Total Num- ber of women Own health care 37.0 20.6 0.9 34.9 6.6 0.0 100.0 3,549 33.1 7.3 58.6 1.0 100.0 2,142 Large household purchases 20.9 30.2 1.5 40.9 6.3 0.2 100.0 3,549 25.5 5.9 66.4 2.2 100.0 2,142 Daily household purchases 28.8 32.3 1.4 31.8 5.5 0.2 100.0 3,549 26.5 5.8 65.4 2.3 100.0 2,142 Visits to family or relatives 20.9 37.9 1.5 33.7 5.4 0.5 100.0 3,549 29.6 6.1 62.0 2.3 100.0 2,142 What food to cook each day 39.9 26.5 1.7 26.1 5.6 0.2 100.0 3,549 27.0 7.8 63.1 2.1 100.0 2,142 Note: Percentages may not add to 100 due to the exclusion of women with missing information. 1 Never-married, divorced, separated or widowed women The data show that the majority of Ghanaian women, irrespective of their marital status, do not have sole authority over any of the five main household decisions. Thirty-one percent of all women, irrespective of their marital status, have no final say in any of the decisions, while 35 percent of women have a final say in all the five decisions (Figure 3.3). Among unmarried women, decisionmaking is highly dominated by someone else (59-66 percent), while among married women, decisionmaking is somewhat dominated by husbands (26-41 percent). Married women also report that decisionmaking is made jointly with their husband (21-38 percent). Especially disconcerting is the fact that most women do not have sole authority over their own health care. Only about one-third of married and unmarried women make sole decisions about their own health care. A third of married women report that their husbands make sole decisions about their health care and three-fifths of unmarried women report that someone else makes 44 | Characteristics of Survey Respondents sole decisions about their health care. One-fifth of married women report that decisions on their own health care are made jointly with their husbands. Forty-one percent of married women report that their husbands alone decide on large household purchases compared with 21 percent of women who report that they alone are the sole decisionmakers on large household purchases, and 30 percent of women who report it to be a joint decision. Among the five decisions, married women seem to have the greatest say over what to cook each day (40 percent), while unmarried women have the greatest say over their own health care (33 percent). Joint decisions with husbands are especially important when it comes to visiting family or relatives (38 percent). Table 3.11 shows the percentage of women who say that they alone or jointly with someone else have the final say in the five specific household decisions, by background characteristics. More than one- third of women say that they alone or jointly with someone else have the final say in all five decisions. The data show that women’s participation in decisionmaking rises with age. Women who are divorced, separated, or widowed are more likely to participate in decisionmaking than currently married women. Decisionmaking among women also rises with the number of children she has, indicating her greater involvement in decisions that may have an impact on the welfare of her children. There are no significant differences in decisionmaking by urban-rural place of residence and women’s education. Women’s involvement in decisionmaking is highest in the Central region and lowest in the Upper East. Employed women who receive cash earnings are more likely to have a greater say in all five decisions than unemployed women and those who are employed but not for cash. Women in the lowest wealth quintile are least likely to report involvement in all five decisions, but beyond that there is little difference among higher wealth quintiles. 31 12 7 8 7 35 None 1 2 3 4 5 Number of decisions in which woman participates in final say 0 10 20 30 40 Pe rc en ta ge o f w om en Figure 3.3 Women’s Participation in Decisionmaking: Number of Decisions in Which Women Participate in the Final Say, Based on Five Household Decisions GDHS 2003 Characteristics of Survey Respondents | 45 Table 3.11 Women's participation in decisionmaking by background characteristics Percentage of women who say that they alone or jointly have the final say in specific decisions, by background characteristics, Ghana 2003 Alone or jointly has final say in: Background characteristic Own health care Making large pur- chases Making daily pur- chases Visits to family or relatives What food to cook each day All specified decisions None of the specified decisions Number of women Age 15-19 18.1 8.8 9.4 11.5 13.1 6.8 75.9 1,148 20-24 44.7 32.8 38.4 41.4 42.3 25.3 39.5 1,012 25-29 58.6 49.3 58.3 59.1 64.9 37.6 18.8 951 30-34 65.0 59.3 67.0 65.3 71.8 46.4 14.8 802 35-39 65.0 62.8 73.6 69.8 76.6 51.3 12.5 722 40-44 67.6 65.9 72.5 70.9 76.3 53.1 12.3 579 45-49 71.3 67.7 78.1 74.3 82.6 57.7 9.8 477 Marital status Never married 26.6 16.0 16.5 21.0 19.2 13.5 67.8 1,616 Married or living together 58.5 52.6 62.5 60.4 68.1 39.9 17.7 3,549 Divorced/separated/ widowed 82.8 78.4 80.9 80.8 82.4 72.6 9.8 526 Number of living children 0 30.9 20.1 22.5 26.5 25.2 16.7 60.5 1,872 1-2 58.1 52.0 59.4 58.3 64.9 40.5 21.6 1,602 3-4 63.5 58.5 67.5 63.9 72.0 44.9 14.5 1,227 5+ 65.8 61.6 71.7 70.0 77.2 50.9 12.2 990 Residence Urban 51.6 45.9 49.9 50.0 52.5 37.3 34.9 2,755 Rural 51.7 43.3 52.3 52.0 58.4 33.7 27.7 2,936 Region Western 51.4 43.3 50.3 52.4 63.2 35.4 27.6 553 Central 69.7 63.5 66.1 64.4 68.1 57.3 23.2 431 Greater Accra 49.1 45.5 48.8 51.0 50.4 38.9 39.8 942 Volta 27.2 32.9 41.3 43.2 44.0 17.7 40.6 492 Eastern 69.7 60.6 62.8 64.4 66.5 55.3 23.2 601 Ashanti 56.8 51.0 56.0 56.2 60.8 37.8 26.8 1,142 Brong Ahafo 53.1 51.9 54.9 54.8 55.8 43.9 36.5 569 Northern 34.9 26.9 49.8 43.9 53.4 13.6 28.6 499 Upper East 53.2 9.3 14.0 10.3 24.4 6.1 29.8 310 Upper West 32.7 18.5 42.8 35.9 47.0 13.7 40.7 153 Education No education 53.1 41.1 53.6 51.0 60.3 31.5 22.5 1,608 Primary 52.2 47.4 52.7 53.0 57.1 38.6 32.7 1,135 Middle/JSS 50.9 47.2 50.9 50.5 53.6 37.2 34.8 2,279 Secondary+ 49.8 39.2 43.4 49.7 47.7 33.4 37.4 669 Employment Not employed 27.0 17.1 19.5 23.9 25.8 13.5 62.4 1,412 Employed for cash 63.8 58.7 67.1 65.4 70.2 46.9 16.3 3,708 Employed not for cash 33.2 20.3 25.5 24.8 33.3 14.5 51.0 565 Wealth quintile Lowest 46.4 33.0 45.2 41.4 50.3 25.0 30.5 970 Second 53.5 46.5 54.6 55.2 61.2 36.8 27.6 949 Middle 55.4 47.7 52.9 53.7 58.0 37.5 28.9 1,071 Fourth 51.3 48.9 54.7 54.0 57.7 39.7 31.3 1,245 Highest 51.5 45.0 48.5 50.3 51.6 36.3 35.7 1,457 Total 51.7 44.6 51.2 51.1 55.5 35.4 31.2 5,691 Note: Total includes 6 women with missing information on employment who are not shown separately. 46 | Characteristics of Survey Respondents 3.5.2 Attitudes toward Wife-beating Violence against women is receiving considerable attention because it has serious consequences for mental and physical well-being, including their reproductive and sexual health (WHO, 1999). To assess the acceptability of domestic violence, respondents interviewed in the 2003 GDHS were asked whether they thought a husband is justified in hitting or beating his wife for any of the following reasons: if she burns the food, if she argues with him, if she goes out without telling him, if she neglects the children, and if she refuses to have sexual relations with him. The data on attitude toward wife-beating are shown in Tables 3.12.1 and 3.12.2 for women and men, respectively. Nearly one in two women believe that a husband is justified in beating his wife for at least one of the specified reasons (Table 3.12.1). Thirty-seven percent of women believe that a husband is justified in beating his wife if she neglects the children, while one in three women think wife-beating is justified if a wife goes out without telling her husband, and 30 percent believe that wife-beating is justified if she argues with her husband. Twenty percent and 14 percent of women, respectively, believe that a man is justified in beating his wife if she burns the food or refuses to have sex with him. Surprisingly, younger women, who are presumably more educated, are more likely than older women to agree with at least one reason for wife-beating. Currently married women are slightly more likely than never-married or formerly married women to accept wife-beating. Women who reside in rural areas and in the more conservative northern regions are much more likely than urban women and women in the other regions to accept wife-beating for at least one reason. As a woman’s level of education or wealth rises, she is less likely to agree that wife-beating is justified for any reason. There is no clear relationship between attitudes towards wife-beating and women’s employment status or decisionmaking power. Men were also asked their opinions about wife-beating to understand attitudes that may prompt male violence against women (Table 3.12.2). Interestingly, men (32 percent) are less likely than women (49 percent) to feel that wife-beating is justified for at least one reason, and are much less likely to agree that wife-beating is justified for each of the five specific reasons. Characteristics of Survey Respondents | 47 Table 3.12.1 Women's attitude toward wife-beating Percentage of women who agree that a husband is justified in hitting or beating his wife for specific reasons, by background characteris- tics, Ghana 2003 Husband is justified in hitting or beating his wife if she: Background characteristic Burns the food Argues with him Goes out without tell- ing him Neglects the children Refuses to have sex with him Percentage who agree with at least one of the specified reasons Number of women Age 15-19 13.2 27.0 34.5 36.7 16.4 50.6 1,148 20-24 14.2 29.7 34.5 38.1 19.1 48.4 1,012 25-29 13.9 30.9 33.9 39.3 20.8 51.0 951 30-34 15.7 32.3 37.3 39.4 24.5 50.4 802 35-39 13.9 30.0 31.6 36.7 20.8 47.9 722 40-44 11.2 26.8 32.8 32.9 19.5 42.9 579 45-49 14.8 29.5 31.5 33.3 20.1 43.6 477 Marital status Never married 10.5 22.0 29.2 32.0 13.8 44.1 1,616 Married or living together 15.8 33.1 36.6 39.9 23.0 51.0 3,549 Divorced/separated/widowed 11.6 28.0 31.2 33.6 18.2 45.5 526 Number of living children 0 11.0 24.1 30.2 32.3 15.0 45.0 1,872 1-2 15.6 32.3 36.1 39.5 20.7 50.8 1,602 3-4 14.8 30.8 36.5 39.9 22.8 49.6 1,227 5+ 15.4 33.2 34.7 38.6 24.5 50.5 990 Residence Urban 8.9 24.3 28.4 30.8 15.5 41.6 2,755 Rural 18.5 34.3 39.2 42.9 24.1 55.1 2,936 Region Western 16.5 29.8 37.0 38.7 20.6 52.9 553 Central 6.1 39.9 38.7 48.0 19.8 56.2 431 Greater Accra 5.7 13.5 20.1 23.3 8.1 30.3 942 Volta 9.4 16.4 20.0 26.2 9.7 37.6 492 Eastern 12.7 28.3 30.6 31.3 14.6 46.4 601 Ashanti 7.5 27.9 30.5 33.5 17.8 44.2 1,142 Brong Ahafo 8.9 19.6 27.8 19.7 13.3 38.5 569 Northern 37.4 53.1 62.0 68.4 48.5 76.3 499 Upper East 41.3 56.7 61.5 70.0 50.7 81.2 310 Upper West 29.2 59.1 56.0 65.6 30.0 80.6 153 Education No education 23.9 43.4 46.7 51.4 33.0 62.3 1,608 Primary 15.4 30.8 37.8 40.6 20.3 54.4 1,135 Middle/JSS 8.9 23.3 29.0 31.0 13.9 42.8 2,279 Secondary+ 4.0 14.6 13.7 17.5 8.5 25.0 669 Employment Not employed 12.6 26.7 31.2 34.3 16.5 47.2 1,412 Employed for cash 13.2 29.4 33.2 36.0 19.9 46.9 3,708 Employed not for cash 21.6 36.9 46.2 51.3 28.2 62.9 565 Number of decisions in which woman has final say1 0 14.0 26.8 32.5 36.4 16.9 48.1 1,777 1-2 21.6 36.9 41.5 46.5 26.0 58.2 1,055 3-4 14.1 30.8 36.4 39.1 22.1 50.0 842 5 9.7 27.3 30.3 32.0 18.5 43.3 2,017 Wealth quintile Lowest 28.5 45.0 49.2 54.0 35.5 67.2 970 Second 17.3 32.3 40.4 41.7 22.4 55.2 949 Middle 12.9 29.8 33.8 37.7 18.8 49.6 1,071 Fourth 10.3 27.6 30.4 34.0 16.8 46.6 1,245 Highest 5.7 18.6 22.9 25.0 11.5 32.7 1,457 Total 13.9 29.5 34.0 37.1 19.9 48.5 5,691 Note: Total includes 6 women with missing information on employment who are not shown separately. 1 Either by herself or jointly with others 48 | Characteristics of Survey Respondents Table 3.12.2 Men's attitude toward wife-beating Percentage of men who agree that a husband is justified in hitting or beating his wife for specific reasons, by background characteris- tics, Ghana 2003 Husband is justified in hitting or beating his wife if she: Background characteristic Burns the food Argues with him Goes out without tell- ing him Neglects the children Refuses to have sex with him Percentage who agree with at least one of the specified reasons Number of men Age 15-19 12.3 21.4 26.3 32.6 14.3 43.8 1,107 20-24 9.8 17.0 21.0 23.9 12.0 34.7 684 25-29 6.4 16.2 21.1 26.2 8.7 34.1 754 30-34 8.3 14.2 18.9 21.4 9.3 28.7 633 35-39 5.7 12.6 16.6 21.1 7.8 27.5 498 40-44 4.0 11.7 12.7 14.9 6.5 21.8 412 45-49 6.2 12.7 15.8 17.4 8.0 23.3 441 50-54 4.9 11.7 15.3 17.7 8.3 23.6 294 55-59 5.6 14.0 15.6 15.0 9.4 24.3 192 Marital status Never married 9.9 18.2 22.2 26.8 12.1 36.7 2,042 Married or living together 6.9 14.4 18.3 21.6 9.1 28.9 2,671 Divorced/separated/widowed 4.8 12.2 16.8 19.0 6.2 28.3 302 Number of living children 0 9.8 17.9 22.3 27.2 12.3 36.8 2,300 1-2 6.9 14.8 19.6 21.8 7.9 30.5 981 3-4 5.0 11.8 15.5 19.5 7.4 24.9 816 5+ 7.4 15.2 17.7 19.8 9.4 27.9 917 Residence Urban 5.2 12.0 15.3 17.5 7.1 25.0 2,250 Rural 10.3 18.9 23.5 28.5 12.6 37.8 2,765 Region Western 12.7 25.3 25.1 30.7 14.2 41.7 476 Central 6.0 12.5 20.3 20.5 7.8 29.8 370 Greater Accra 3.0 7.9 9.1 12.0 4.7 16.5 733 Volta 4.8 6.6 10.3 12.7 3.5 18.5 440 Eastern 5.1 12.3 14.9 17.9 5.8 23.1 539 Ashanti 3.3 9.0 12.1 13.4 5.9 21.6 956 Brong Ahafo 5.0 13.0 19.5 18.0 6.2 29.8 528 Northern 22.9 34.4 45.7 54.8 32.0 65.4 527 Upper East 14.1 27.7 28.1 41.2 14.3 54.3 317 Upper West 19.0 37.8 44.5 59.0 20.8 69.2 130 Education No education 15.1 27.4 33.4 41.2 22.0 50.9 881 Primary 13.1 23.8 29.3 32.6 15.4 46.4 803 Middle/JSS 5.7 13.3 16.7 19.6 6.6 27.8 2,165 Secondary+ 3.4 6.2 8.6 11.3 4.0 15.7 1,165 Employment Not employed 8.9 15.3 20.0 25.7 11.0 35.4 1,224 Employed for cash 7.1 14.9 18.7 21.5 9.0 29.0 3,448 Employed not for cash 13.9 27.3 29.7 36.7 18.2 50.2 341 Number of decisions in which wife has say1 0 9.4 17.6 21.8 25.8 11.7 35.3 3,386 1-2 7.4 15.5 18.6 24.5 10.2 32.0 713 3-4 3.3 9.2 13.2 14.1 4.9 19.0 674 5 3.2 9.6 13.7 16.5 2.3 22.2 242 Wealth index Lowest 16.4 27.0 32.1 41.2 19.1 52.8 872 Second 8.5 17.3 22.7 25.4 12.3 35.0 903 Middle 8.9 16.8 23.3 25.0 9.9 33.8 975 Fourth 5.6 12.8 14.6 18.9 7.4 27.3 1,060 Highest 2.9 8.3 10.4 12.4 4.5 17.4 1,204 Total 8.0 15.8 19.8 23.6 10.1 32.0 5,015 Note: Total includes 2 men with missing information on employment who are not shown separately. 1 Either by herself or jointly with others Characteristics of Survey Respondents | 49 3.5.3 Attitudes toward Refusing Sex Women’s rights and control over their own sexuality are important aspects of their empowerment. In addition, their control over when and with whom they have sex has an impact on their health, especially with respect to the transmission of STIs such as HIV/AIDS. Respondents in the GDHS were asked whether a wife is justified in refusing sex when: she knows that her husband has a sexually transmitted disease; she knows that her husband has sex with other women; she has recently given birth; and when she is tired or not in the mood for sex. Tables 3.13.1 and 3.13.2 show the percentage of women and men who believe that a wife is justified in refusing to have sex with her husband for these specified reasons, by background characteristics. Sexual autonomy is relatively high among Ghanaian women. Nearly two-thirds of women and men agree that women are justified in denying sex to their husbands for all four reasons, with little variation between women and men by background characteristics. Women and men who reside in rural areas, those who live in the Upper West region, poorly educated respondents, and respondents from the lowest wealth quintile are somewhat less likely than their counterparts to agree that a woman is justified in refusing sex with her husband for all four reasons. There is no clear relationship between women’s sexual autonomy and her decisionmaking power or beliefs about wife-beating. Male respondents in the 2003 GDHS were also asked whether they thought that a husband had the right to take specific actions—get angry and reprimand her, to refuse to give her money or other means of financial support, to use force and have sex with her even if she does not want to, and to have sex with another woman—if his wife refused to have sex with him. Table 3.14 shows that one in four men agree with at least one of the four specified actions. Fifteen percent of men say it is justifiable for men to get angry and reprimand the wife, 12 percent say it is okay to have sex with another woman, 10 percent say it is justifiable to refuse financial support, and 5 percent say it is acceptable to use force if she refuses to have sex with him. Differences by background characteristics are not large. However, men in the youngest cohort (15-19), rural men, men in the Upper West and Northern regions, men with no education, men who are employed but not for cash, and men in the lowest wealth quintile are less tolerant of women’s sexual autonomy than other men. 50 | Characteristics of Survey Respondents Table 3.13.1 Women's attitude toward wives refusing sex with husbands Percentage of women who believe that a wife is justified in refusing to have sex with her husband for specific reasons, by background characteris- tics, Ghana 2003 Wife is justified in refusing sex with husband if she: Background characteristic Knows husband has a sexually transmitted disease Knows husband has sex with other women Has recently given birth Is tired or not in the mood Percentage who agree with all of the speci- fied reasons Percentage who agree with none of the specified reasons Number of women Age 15-19 84.7 79.1 79.1 73.1 63.0 9.8 1,148 20-24 86.5 81.8 85.0 75.2 62.5 6.2 1,012 25-29 86.4 78.8 84.4 74.4 62.4 8.0 951 30-34 84.9 76.8 84.5 74.9 61.1 6.0 802 35-39 86.5 76.7 82.8 72.6 62.1 8.7 722 40-44 88.6 79.7 86.4 76.8 65.9 6.2 579 45-49 87.4 79.4 86.1 74.7 63.4 7.6 477 Marital status Never married 86.0 80.2 80.3 72.7 62.7 8.8 1,616 Married or living together 85.8 77.8 84.1 74.4 61.8 7.5 3,549 Divorced/separated/widowed 89.6 83.5 90.3 79.6 69.6 4.8 526 Number of living children 0 86.4 80.8 82.3 74.0 63.3 7.8 1,872 1-2 85.8 78.3 84.3 75.5 62.1 7.2 1,602 3-4 86.1 77.9 84.1 73.3 62.2 7.9 1,227 5+ 86.6 78.0 84.1 74.7 63.5 7.8 990 Residence Urban 88.1 81.2 84.4 76.3 63.9 6.6 2,755 Rural 84.4 76.9 82.8 72.6 61.7 8.6 2,936 Region Western 92.3 85.3 90.4 79.4 68.9 2.8 553 Central 87.4 88.5 87.2 60.9 51.9 4.6 431 Greater Accra 87.7 81.5 86.0 77.9 67.5 7.2 942 Volta 82.9 76.3 76.9 74.5 61.3 12.8 492 Eastern 80.3 81.2 83.7 75.8 66.3 10.0 601 Ashanti 85.9 74.7 81.2 74.6 62.1 9.4 1,142 Brong Ahafo 84.4 80.4 83.8 78.6 67.6 8.2 569 Northern 83.3 72.9 79.7 68.8 54.5 7.7 499 Upper East 95.5 72.9 88.0 79.2 63.1 0.9 310 Upper West 83.7 73.5 74.9 57.8 45.5 8.3 153 Education No education 84.5 73.2 81.7 69.4 56.0 8.2 1,608 Primary 83.3 78.9 82.1 74.2 63.0 9.8 1,135 Middle/JSS 88.0 82.0 84.6 77.2 66.5 6.9 2,279 Secondary+ 88.8 82.6 87.0 77.0 65.9 4.9 669 Employment Not employed 86.7 80.8 82.9 75.1 63.6 7.3 1,412 Employed for cash 86.3 79.5 84.6 75.1 63.6 7.5 3,708 Employed not for cash 84.2 71.3 78.6 67.8 55.2 8.9 565 Number of decisions in which woman has final say1 0 84.9 78.2 79.8 73.6 62.8 9.5 1,777 1-2 84.4 73.0 83.2 72.0 57.8 7.9 1,055 3-4 87.0 78.6 81.8 70.3 59.2 7.4 842 5 87.9 83.0 87.8 78.0 66.9 5.9 2,017 Number of reasons wife-beating is justified 0 85.3 79.5 83.4 74.5 64.5 8.9 2,928 1-2 85.8 77.0 81.3 73.5 59.7 7.7 1,279 3-4 88.2 77.1 85.4 72.2 58.0 4.7 1,047 5 88.4 85.7 87.3 81.1 71.5 5.9 437 Wealth quintile Lowest 86.2 75.3 83.6 73.0 60.7 7.1 970 Second 83.7 76.0 80.7 72.4 62.0 9.7 949 Middle 83.5 80.0 81.8 72.2 62.1 9.4 1,071 Fourth 86.2 79.5 82.8 74.7 62.0 8.0 1,245 Highest 89.7 82.2 87.4 78.0 65.8 5.0 1,457 Total 86.2 79.0 83.6 74.4 62.8 7.6 5,691 Note: Total includes 6 women with missing information on employment who are not shown separately. 1 Either by herself or jointly with o

View the publication