3. Discrimination in the family

Gender equality in the private and family spheres is paramount to achieving women’s empowerment in other key areas. However, the household and the family are often places where discriminatory social norms and traditional roles are the strongest and the most difficult to challenge (OECD, 2019[1]). These discriminatory social norms have far-reaching consequences as traditional gender stereotypes do not remain confined to the home and are frequently replicated at the community level, as well as across other institutions such as schools, the workplace and other governance systems. For example, traditional gender stereotypes that restrict women and girls to household tasks and caring responsibilities may prevent them from pursuing professional careers, entering the job market, or assuming leadership roles in the community or society in general. Rigid gender norms within the family may also result in girls marrying before the age of 18, having an inferior status and less decision-making power in the home, being economically dependent on men, and enduring violence and harassment in the private sphere.

The present chapter is divided into two main sections. Each section focuses on one of the two main forms of discrimination that women and girls face in the family sphere – girl child marriage and unequal distribution of unpaid care and domestic work and decision-making power. These two challenges are critical because they generate many negative spillovers in other aspects of women’s and girls’ lives. For each section, the analysis presents the current situation of women in Tanzania and across the country’s 31 regions. The analysis then uncovers the role played by discriminatory social institutions and specifically discriminatory social norms, attitudes and stereotypes in explaining Tanzania’s unequal outcomes – namely, the large share of women who are married before the age of 18, their disproportionate share of unpaid care and domestic work, and their limited decision-making power in the household. Finally, each section uncovers some of the main determinants of these discriminatory social norms and attitudes that constrain the outcomes of Tanzanian women.

Recognised as a human rights violation, child marriage3 not only threatens the well-being, health and future of girls, but also restricts a country’s development and imposes significant economic costs (UNICEF, n.d.[2]). Girl child marriage is often accompanied by adolescent pregnancies, which entail high risks of maternal mortality, morbidity and infant mortality, and may lead to greater social isolation and increased risks of intimate partner violence (Izugbara, 2018[3]; Lee-Rife et al., 2012[4]). Child marriage is also associated with lower educational attainment – especially at the secondary school level. This, in turn, curbs the empowerment, decision-making power and financial independence of women and girls, as well as their ability to contribute to the socioeconomic development of their households, communities and society as a whole. These negative effects of child marriage may have long-term consequences, including for future generations. For instance, child marriage induces higher risks of adolescent pregnancies, which may lead to risks of stunting, wasting or underweight among children – with negative implications for the long-term development of human capital and welfare outcomes for all citizens (UNICEF, n.d.[2]; Harper et al., 2014[5]; Lee-Rife et al., 2012[4]).

Against this backdrop, Tanzania has enacted a National Plan of Action to End Violence against Women and Children in Tanzania 2017/18-2021/22, which explicitly targets child marriage. The plan acknowledges the negative consequences of the harmful practice and aims to reduce the prevalence of child marriage to 10% by 2022 (Ministry of Health, Community Development, Gender, Elderly and Children, 2016[6]). The action plan is accompanied by the National Integrated Communications and Outreach Strategy to End Violence against Women and Children and an Outreach Toolkit 2017/2018–2021/2022. Zanzibar’s National Plan of Action to End Violence against Women and Children 2017-2022 also acknowledges the importance of ending child marriage. The plan identifies child marriage as a harmful practice in and of itself as well as a risk factor in violence against women and children (Government of Zanzibar, 2017[7]).

In 2016, Tanzania amended the Education Act to prohibit child marriage for children attending primary or secondary school in Mainland Tanzania. It is hence unlawful for any person to marry a school boy or girl and vice-versa (Government of Tanzania, 2016[8]). Most recently, in November 2021, Tanzania announced that all students in Mainland Tanzania who dropped out of school would be given the opportunity to return. This includes girls who left primary or secondary school due to pregnancy and who were not allowed to return after childbirth (Dausen, 2021[9]). However, as of December 2021, the adoption of a formal policy or legal instrument to revise the Tanzania’s Education Act, which discriminates against pregnant girls, was yet to appear.4

In spite of a number of implemented programmes and associated progress (Box 3.1), concerted efforts are still needed to eliminate the practice completely, particularly in the context of the COVID-19 pandemic. The resultant external economic shocks, rising poverty levels and restricted access to services ranging from access to reproductive health services to child protection, have all increased the risk of girls being forced into child marriage (OECD, 2021[10]; UNICEF, 2021[11]; UNFPA, 2020[12]).

Child marriage is prevalent in Tanzania and affects primarily girls: 92.5% of all registered child marriages are girl child marriages.5 SIGI Tanzania finds that 19% of Tanzanian women aged 15 years and older have been or were married before the age of 18 years (Figure 3.1). When looking at child marriage among women aged 20 to 24 years – which corresponds to SDG indicator 5.3.1 (Box 3.2) – the share is slightly smaller at 16%6, showing that Tanzania fares better than the average rate of 31% for the East and Southern Africa region (UNICEF, 2021[14]).Gaps in the legal framework on marriage may contribute to the continued prevalence of girl child marriage. For example, Zanzibar does not have a minimum age for marriage, whereas in Mainland Tanzania, the law permits girls as young as 15 years to marry with a court order (see Annex 3.A).

Child marriage rates are significantly higher in rural areas than in urban settings. In rural areas, 23% of women aged 15 years and older report having been married, divorced or widowed before the age of 18 years, compared to 12% in urban areas (Figure 3.1). Likewise, the share of women aged between 20 and 24 years married before the age of 18 is significantly higher in rural settings (24%) than in urban areas (4%). Results at the regional level show that child marriage is unevenly distributed across Tanzania and is particularly acute in certain regions. Specifically, in six regions7 more than 25% of girls and women aged 15 years and older have been married before the age of 18. In Kaskazini Pemba, this share reaches 37% (Figure 3.1). According to the SDG indicator, four regions – Mara, Morogoro, Shinyanga and Simiyu – report girl child marriage rates above 30%.

Child marriage has decreased consistently over the last five decades, but progress appears to have stalled in recent years. A comparison of the prevalence rate of girl child marriage across different age brackets shows changes in rates of child marriage over time. As an example, if 20% of women aged between 40 and 44 years today were married before the age of 18 years, this means that 20 years ago, 20% of women then aged between 20 and 24 years married before the age of 18 years. Using this approach reveals how girl child marriage has evolved over several decades in Tanzania. In 2021, 33% of women aged between 65 and 69 years reported having been married for the first time before the age of 18 years, indicating that 45 years ago, the child marriage rate was roughly 33%. This rate decreased continuously for 20 years reaching 19% among women aged 50-54 years. The age bracket analysis, however, suggests that since the mid-1990s progress has stalled. In particular, the girl child marriage rate for the age brackets 20-24 to 50-54 does not vary significantly and ranges from 16% to 21% (Figure 3.2).8 These trends based on the SIGI Tanzania data are in line with other sources such as Demographic Health Survey results or census statistics, which show a similar long trend in the reduction of child marriage albeit with a plateau since the 1990s (United Nations, 2019[19]).

Moreover, progress in reducing girl child marriage has not been uniform and varies greatly between urban and rural areas. Immense progress has been achieved in urban areas: while 12% of women of all ages (15 years and above) have been married before the age of 18, the rate falls to 4% for women aged between 20-24 years, i.e. those women who have been married as a child over the past five years. The difference between the two rates is a rough estimation of the decrease in the practice. However, there has been no visible progress in rural areas in this regard. While the child marriage rate is 23% for rural women of all ages (15 years and above), for those aged between 20 and 24 years it is higher at 24%.

The trend analysis also reveals that large gains have been made in Zanzibar. While 21% of women of all ages have been married before the age of 18, the rate falls to 13% for women aged between 20 and 24 years – with variation in rates across regions. Some regions exhibit a downwards trend, while others show higher levels of child marriage among women aged 20-24 years compared to all women in Tanzania aged 15 years and older who have been married before the age of 18 years, suggesting a recent increase in child marriage. Given the limited size of the sample of women necessary to calculate the SDG indicator at the regional level, these trends should be considered with caution. Nevertheless, policies, interventions and programmes should concentrate on those regions where data suggest an important increase in child marriage – namely, Dodoma, Mara, Morogoro, Mtwara, Rukwa and Shinyanga.

Girl child marriage increases the likelihood of adolescent pregnancies (i.e. childbirth under the age of 20 years). Adolescent pregnancy remains an acute problem in Tanzania. At the national level, one in three women have had a child before the age of 20. In nine regions,9 more than one in four women currently aged 15 years and older were pregnant for the first time before the age of 20 (see Chapter 4). Data show that the age at which women are married and the age at which they have their first child are profoundly and significantly interrelated. In Tanzania, the older women are married, the older the age at which they have their first child (Figure 3.3, Panel A).10 Consequently, controlling for other socio-demographic factors, being a child bride significantly increases the likelihood of being pregnant as an adolescent.11 In this regard, regions with higher levels of child marriage in Tanzania report higher levels of adolescent pregnancies (Figure 3.3, Panel B). These high rates of adolescent pregnancies carry important health risks for both the young mother and her baby (see Chapter 4). In particular, maternal mortality and morbidity are higher among adolescent mothers compared to mothers who are older than 20 years, and babies of adolescent mothers face greater risks of lower birth weight, preterm delivery and severe neonatal conditions (WHO, 2020[20]). When the mother dies in childbirth, the surviving baby is vulnerable to several challenges ranging from nutritional problems caused by the lack of breastfeeding to increased risks of suffering from interrupted education or difficult living arrangements (Whetten et al., 2011[21]; Wang et al., 2013[22]).

At the same time, girl child marriage is perceived as a socially acceptable solution to out-of-wedlock pregnancies (Mbuyita, 2021[23]; Gemignani and Wodon, 2015[24]). In Tanzania, adolescent pregnancies that result from child marriage are often perceived as acceptable since they occur within the socially sanctioned institution of marriage. Early pregnancies are viewed as problematic when they occur out of wedlock. The mechanisms involved entail a combination of parents wishing to avoid the social shame associated with adolescent pregnancies occurring out of wedlock and the man responsible for the pregnancy wanting to avoid going to court where he risks being sentenced to 30 years in jail – the maximum penalty for impregnating a pupil/student or a girl under 18 years of age in Tanzania. Both sides often agree to an illegal marriage between the girl and the future father, who is then responsible for her care and welfare (Mbuyita, 2021[23]). For example, in Shinyanga, which has one of the highest girl child marriage rates, and where 50% of women report having their first child before the age of 18, the main causes of child marriage are reported to be adolescent pregnancies, peer pressure, girls’ desire to obtain money to buy small items and poverty in the household (Mbuyita, 2021[23]). In extreme cases, the same mechanism is used to cover up rape followed by a pregnancy. Child marriage between the girl who has been raped and her aggressor is often regarded as an acceptable “solution” that allows the girl’s family to protect their honour and the perpetrator to avoid legal punishment (Mbuyita, 2021[23]).

Girls who are married young are more likely to achieve lower educational attainment.12 In Tanzania, girl child marriage occurs on average at the age of 16 years – an age at which girls are expected to attend secondary school. Controlling for various socio-demographic factors, child marriage rates are significantly and positively associated with higher shares of women without formal education (Figure 3.4, Panel A). Conversely, they are negatively associated with higher shares of women with secondary education (Figure 3.4, Panel B). In many instances, child marriage leads to the interruption of girls’ schooling. In Mainland Tanzania, it is unlawful for any person to marry a girl (or a boy) who attends primary or secondary school (Government of Tanzania, 2016[8]). As a result, families that desire to marry their daughters might be tempted to remove them from school before organising the marriage. At the same time, a husband may oppose his young bride attending school and expect her to care for the household instead. In cases of adolescent pregnancies resulting from child marriage, constraints related to childrearing can impede girls from continuing their education. Until November 2021, a ban on school pregnancies that prohibited adolescent mothers from continuing or resuming their studies also played a role in limiting girls’ education (Dausen, 2021[9]).

Evidence from the SIGI Tanzania database suggests that the relationship between education and girl child marriage may run in both directions and that girls with low educational attainment or who have already dropped out of school are more at risk of marrying before the age of 18 (Lee-Rife et al., 2012[4]; UNICEF, n.d.[2]). Controlling for socio-demographic factors, higher shares of women with primary and secondary education are significantly associated with lower rates of girl child marriage.13 At the national level, 35% of women without formal education have been married before the age of 18 compared to 30% of women who did not complete primary education, 15% of those who completed primary education and 1% of those who completed secondary school.14 There are multiple underlying drivers. For instance, the perception of marriage as an opportunity for girls to increase their social status may encourage girls who are already out of school to marry as soon as possible. In contexts with structural barriers, such as poverty or limited access to education, marriage at a young age might also be perceived as a desirable option to pursue (Schaffnit et al., 2020[25]).

The consequences for human capital accumulation and women’s empowerment are far-reaching. Secondary schooling is essential for women’s empowerment and non-completion of secondary school often represents a barrier to further human capital accumulation, including in the professional sphere (Anyanwu, 2016[26]). Women with higher education, in comparison to those with lower education, also tend to invest more in the health and education of their children, and have fewer children, generating positive spillovers for the productivity and well-being of the next generation (Duflo, 2012[27]; Klasen and Lamanna, 2009[28]; Cabeza-García, Del Brio and Oscanoa-Victorio, 2018[29]). Keeping girls in school and ensuring that young mothers are reintegrated is therefore paramount. Tanzania has taken steps in this regard by establishing Vocational Training courses and Focal Development Colleges to create an enabling environment for adolescent girls. In addition, the Alternative Learning Skills Development project targets adolescents who have dropped out of school with a view to their re-integration (Government of Tanzania, 2019[30]). However, these measures have received criticism on the grounds that they may create alternative education pathways for pregnant girls and young mothers who are denied their right to compulsory education. Furthermore, some of these colleges are not cost-free and/or located far from the girls’ homes, which creates barriers to access and attendance (Human Rights Watch, 2021[31]). An announcement in November 2021 removing the ban preventing young mothers from returning to school seeks to address these shortcomings.

Bride price is a social institution in Tanzania that goes hand in hand with marriage. At the national level, 9 out of 10 women or girls have been married with a bride price, and in the large majority of cases (84%), the parents of the bride receive the payment, which has been made either by the groom’s parents or by the groom himself (Box 3.3). While the custom is equally common in both rural and urban areas, significantly more married women have been married with a bride price in Zanzibar than in Mainland Tanzania (98% compared to 90%, respectively). Moreover, social support of the practice is elevated in Tanzanian society, regardless of age, gender or residential status, with 90% of the population believing that marriage requires a bride price.

Although bride price is a custom rather than a social institution and is not necessarily harmful to women and girls, certain attitudes associated with the practice may severely diminish women’s empowerment and agency. Existing evidence and research on the effects associated with bride price are mixed (Voena and Corno, 2016[33]; Lowes and Nunn, 2018[32]). For instance, some empirical findings show that women married with a bride price experience lower levels of violence or are happier in their marriage than those who have not been married with a bride price. However, the practice often raises questions related to the husband’s ownership of the bride, incentives for child marriage, increased levels of adolescent pregnancies and the risk of being locked into an abusive relationship, especially where the tradition requires repayment of the bride price in case of divorce (Lowes and Nunn, 2018[32]). In Tanzania, for instance, 75% of the population believes that a husband owns his wife once he has paid a bride price. The consequences of this assumption can be far reaching, constraining women’s independence, well-being and decision-making power in the household.

Moreover, the transactional nature of bride price may encourage the practice of girl child marriage. In contexts where economic resources are limited, girl child marriage may be perceived as a means to alleviate a family’s economic burden. Firstly, the bride’s family will receive the bride price in the form of monetary or in-kind valuable assets; secondly, marrying a daughter may free up resources within the household that were formerly used to sustain the girl/bride (Voena and Corno, 2016[33]). Focus group discussions reveal that girls are often perceived as a source of income due to the existence of bride price, and that families sometimes seek marriage for their daughters for economic reasons. The COVID-19 pandemic has increased the risk of child marriage, particularly in the context of economic distress and limited access to health care and child protection services (OECD, 2021[10]; UNICEF, 2021[11]; UNFPA, 2020[12]). In this regard, there is evidence to suggest that alleviating families’ economic situations can help prevent child marriage.15 Pressure for early marriage can further stem from a potential decrease in the bride price payment for girls who have already given birth or who are no longer virgins. Consequently, in order to obtain a high payment, families may push for an early marriage that decreases the risk of an adolescent pregnancy out of wedlock (Mbuyita, 2021[23]). Moreover, larger bride price payments reported in Tanzania for younger brides may incentivise families to marry their daughters as early as possible (Schaffnit et al., 2019[34]).

Social acceptance of child marriage is associated with higher prevalence rates of this harmful practice. Controlling for various socio-demographic characteristics, the prevalence of girl child marriage increases significantly and strongly in tandem with the share of the population considering it appropriate for a girl to marry before the age of 18 years (Figure 3.5, Panel B).16 At the national level, 19% of the population holds such views, with 13% of the population believes that a girl should be married before her 18th birthday. In contrast, only 6% of the population considers it appropriate for a boy to marry before the age of 18 years.

Social acceptance of child marriage varies significantly between rural and urban areas as well as across regions. In rural areas, 24% of the population sets the appropriate age for a woman’s marriage below 18 years compared to 11% in urban areas (Figure 3.5, Panel A). Similarly, a greater share of the rural population (15%) compared to the urban population (9%) considers that a girl should be married before her 18th birthday. Social attitudes supporting and encouraging child marriage also vary greatly across regions. While in seven regions,17 the share of the population that sets the appropriate age for girls to be married at under 18 years is below 10%, in four regions18 the share accounts for more than one-third of the population.

Social norms, particularly those related to restrictive masculinities, may explain differences in perceptions regarding the appropriate age to marry for men and women. On average, respondents set the appropriate age to marry at 18.5 years for women compared to 22 years for men. At the same time, nearly all of Tanzania’s population (92%) agrees or strongly agrees that a “real” man should be the family’s breadwinner. Achieving sufficient economic stability to provide enough income to support a family, including finding a suitable home, requires time. As a result, many believe that men should marry older so they have enough time to create such conditions before starting a family. The fact that the appropriate marriage age for women is set lower than for men may reflect the way in which restrictive gender norms, including the importance of virginity at marriage and customs such as bride price, put pressure on women to marry earlier than men.

Restrictive gender norms place decision-making power in the hands of men and deny girls and women decision-making power over their lives and relationships, especially in regard to marriage (Jayachandran, 2015[35]). At the national level, only about one-third of the population considers that a daughter should have the agency to make decisions regarding her own marriage – including the choice of her husband. One-third of the population believe that both parents should make decisions collectively regarding their daughter’s marriage, while one-quarter hold the view that the father alone should be in charge. These norms regarding decision making are closely linked to restrictive masculinities that establish men as the head of the household and primary decision makers (OECD, 2021[36]). In this regard, 76% of Tanzanians agree or strongly agree that to be a “real” man, a man should have the last word in the home. In comparison, only 46% of the population believes that to be a “real” woman, a woman should have the last word in the home. Focus group discussions held with adolescent girls in Shinyanga revealed that their father would take the decisions regarding their marriage including the choice of husband and would arrange the bride price payment with the groom or his family. Mothers would only be informed once the decisions and arrangements had been made (Mbuyita, 2021[23]). Across Tanzania, the small share of people believing that a mother should take decisions regarding her child’s marriage and the high proportion of people who consider that a boy should be in charge of making decisions for his own marriage provide additional evidence of the gendered dimensions of decision making (Figure 3.6). These attitudes underscore the limited decision-making power that girls enjoy over their own marriage – particularly relative to boys – and that decision making often lies in the hands of men, especially fathers.

Attitudes denying girls control over their own marriage are more acute in Zanzibar and in certain regions of Mainland Tanzania. On average, 37% of the population in Mainland Tanzania considers that a girl should make decisions over her own marriage, compared to only 28% in Zanzibar. Conversely, 40% of Zanzibar’s population thinks that the father alone should be in charge of decisions for his daughter’s marriage (Figure 3.6). These attitudes vary significantly across different regions. In nine regions,19 more than half of the population considers that a girl should be responsible for decisions regarding her own marriage, whereas in four regions20 a similar proportion of the population considers that this power should rest with the father. By contrast, men and boys are generally granted greater power than women and girls regarding decision-making power over marriage. Both in Mainland Tanzania and Zanzibar, more than 45% of the population believes that a boy or young man should make decisions regarding his own marriage.

Discriminatory social norms and practices establishing male guardianship over women can promote child marriage. Indeed child marriage is significantly more likely to occur when women and girls are denied agency and autonomy in decisions over their own marriage.21 While half of the women who married after their 18th birthday took their own decisions regarding the marriage, the proportion was lower at less than 40% for those who married as a child. Conversely, the number of marriages where the father alone took decisions is greater for girls who were married before the age of 18 years (18%) than for those who were married later (14%). In these situations, gender norms that limit women’s and girl’s decision-making power and agency overlap with limitations on the basis of age and the fact that children often have limited decision-making power. This dynamic further marginalises the agency of girls, placing them at risk of child marriage.

Restrictive social norms guiding social understanding of what it means to be a “real” woman in Tanzanian society may be an even more important factor explaining the persistence of child marriage. The importance of remaining a virgin at marriage is a social norm that may induce child marriage, particularly in contexts where pregnancies out of wedlock are highly stigmatised and regarded as a threat to a family’s reputation (Gemignani and Wodon, 2015[24]). In addition, girls who remain unmarried for a long time may face mistrust and scrutiny regarding their virginity, which again can put the reputation of the girl and her family reputation at risk (Mbuyita, 2021[23]; Lee-Rife et al., 2012[4]). On average, 40% of Tanzania’s population agree that being a virgin at marriage is essential in order to be a “real” woman. Conversely, 30% of the population consider that virginity at marriage is essential for being a “real” man, reflecting somewhat asymmetric gender norms when it comes to sexuality. Prevalence of the belief that a “real” woman should still be a virgin at marriage varies significantly across regions, ranging from 15% in Geita to more than 65% of the population in Zanzibar.

Women are as likely as men to hold views that are supportive of girl child marriage. However, the likelihood that an individual will hold the belief that a girl should be married before the age of 18 or that this practice is appropriate decreases as age increases (Table 3.1). In addition, where a person lives matters. On average, the likelihood that child marriage will be upheld as an appropriate practice is higher in rural than in urban areas and higher in Zanzibar than in Mainland Tanzania.

Individuals with lower levels of education are also more prone to hold attitudes that encourage or justify the practice of girl child marriage. More precisely, individuals with at least some education are less likely than those without any formal education to declare that the minimum appropriate age for girls to marry is before 18 years or to believe that a girl should be married before the age of 18. Moreover, the likelihood of holding such discriminatory attitudes decreases as education increases, with the effect being strongest for individuals with secondary education or university-level education (Table 3.1). These results suggest that education is essential to empower both girls and boys and has the power to transform norms and aspirations for women that go beyond marriage and motherhood (Anyanwu, 2016[26]). For instance, as educational attainment increases, the likelihood of believing that a “real woman” should be married or should remain a virgin before her marriage decreases. Individuals with higher levels of education are also more likely to believe that a woman or a girl should be in charge of making decisions regarding her own marriage (Table 3.1).

Similar to education, individuals from wealthier households are less likely to hold discriminatory attitudes that condone girl child marriage. This is the case across all statements tested (Table 3.1). In all cases, the effect is strongest for individuals belonging to the highest wealth quintiles. The result are consistent with the strong interlinkages that exist between education and wealth.

Unequal distribution of unpaid care and domestic work as well as asymmetric power relationships within the household have serious implications for women and girls throughout their lifetimes. Gendered divisions of labour defining unpaid care and domestic work as the responsibility of women and girls constrain the time they are able to dedicate to education, training and paid employment. Furthermore, unequal decision-making power between men and women in the household may have consequences for women’s economic empowerment, curtailing their ability to seek a job or start a business and limiting their ownership and use of assets such as land and finances (see Chapter 2). In addition, men’s final say in household decisions, including those governing access to education and healthcare, can be detrimental to women’s own health and human capital development as well as that of their families. In this regard, the inequities that arise from intra-household dynamics have a bearing not only on women’s roles and responsibilities within the household, but also within society as a whole, impacting their empowerment and participation in public life.

Women undertake significantly more unpaid care and domestic work than men. On average, women spend 4.4 hours per day on unpaid care and domestic work, compared to 1.4 hours daily for men (Figure 3.7). Variations across rural and urban areas as well as between Mainland Tanzania and Zanzibar are very limited. Consequently, women spend on average three times more time on unpaid care and domestic work than men (Box 3.4).

In addition to their disproportionate share of unpaid care and domestic work, women face a double burden as they also undertake a large amount of paid work, resulting in a larger total workload. On average, at the national level, women in the labour force work 5 hours per day compared to 6 hours for men. As a result, women bear a significantly larger share of the total workload, spending, on average, a total of 9.4 hours a day on paid and unpaid work compared to 7.4 hours for men. The situation differs slightly between Mainland Tanzania and Zanzibar. In both locations, women undertake a significantly larger amount of unpaid care and domestic work than men. However, in Zanzibar, although women also spend 5 hours a day on paid work, men’s particularly large amount of time spent on paid work results in a similar total workload for both men and women (Figure 3.7).

Women’s unpaid work burden is particularly concentrated and disproportionate in basic and routine household tasks. When considering only certain basic unpaid care and domestic tasks such as cooking, cleaning the house and the bathroom or toilet, washing clothes, and taking care of the elderly and the sick, Tanzanian women spend four times more time than men on these basic but essential household tasks. For instance, women spend about 90 minutes cooking for the household, compared to only 12 minutes for men. Overall, women spend most time on cooking, washing clothes and cleaning the house, dedicating significantly more time to these tasks than men (Figure 3.8). Not surprisingly, men tend to undertake tasks that occur weekly or monthly, such as repairing the house, rather than routine and daily tasks. This household division of tasks constitutes a typical rule found across many countries and places (OECD, 2021[36]). Men also perform tasks that require leaving the household such as buying groceries or fetching firewood or water. This is particularly the case in Zanzibar where the female-to-male ratio for buying groceries is 0.7, indicating that men spend more time on this activity than women. Likewise, in Zanzibar, the ratios of time spent for fetching firewood and water stand at 1.3 and 1.7, respectively.

Traditionally, men assume household headship, which gives them the authority to make decisions on behalf of the rest of the household members on all matters, including expenditure or income, education, family planning and the marriage of children (Mbuyita, 2021[23]). Within households, fathers have a dominant role in taking important decisions related to children and, more precisely, education, health and marriage. Although decisions on children’s health and education are mostly shared equally between mothers and fathers, fathers still tend to have a greater say. The majority of Tanzanians (60%) live in households where decisions regarding children’s health or education are taken together by both parents. Yet, more than one-quarter and nearly one-third of them live in households where mothers are excluded from decision-making processes on children’s health and education, respectively, indicating that in these households the father alone is responsible for these decisions. Differences between rural and urban areas are significant with a higher proportion of households where only fathers take decisions regarding their children’s health and education in rural areas (Figure 3.9). At the same time, fathers also have a dominant role in making decisions regarding their children’s marriage. For instance, more than 50% of the population live in households where the father participates in the decision-making process for a son’s or daughter’s marriage. In contrast, only 28% and 38% of the population live in households where the mother is involved in decisions related to a son’s or daughter’s marriages, respectively.

Decision-making power over household consumption is largely in the hands of men, especially in rural areas. Men are often the exclusive decision makers within Tanzania’s households in regard to basic spending on food and clothing as well as large and/or productive purchases such as a house, land, equipment or farm inputs. Specifically, 37% of Tanzania’s population identify the male household head as the sole decision maker over basic consumption spending. This share is significantly higher in rural areas than in urban ones. While 27% of the urban population identify the male household head as the sole decision maker over basic consumption spending, this share reaches 43% in rural areas. Furthermore, for productive assets, more than 40% of the population identify the male household head as the individual who has the last word on important decisions such as choosing farm inputs, buying or renting a house, or buying transportation vehicles. Likewise, this proportion is about 10 percentage points higher in rural areas than in urban ones, highlighting women’s lower agency within rural households (Figure 3.10).

Men also exert primary control over household income, particularly in rural areas where such income is often earned through farming activities. Evidence from focus group discussions in the rural community of Shinyanga indicates that men are solely responsible for decisions concerning income generated from farming activities. Men have access to markets and also make decisions regarding the sale of crops. After the sale, men often choose how to spend the money without necessarily involving their wives/partners or other members of the household who participated in the farming and harvesting activities (Mbuyita, 2021[23]). When women control their own business or are engaged in small businesses such as selling vegetables from gardening, snacks or small fish, they tend to have more control over the income they generate. In many instances, they might be able to buy household items, food or small assets. However, even when acquiring a certain level of control over the income generated, women still face difficulties in buying large assets such as land or in accessing financial opportunities such as Village Savings and Loans Associations (Mbuyita, 2021[23]). As a result they may choose to do so secretly, without informing their husband (see Chapter 2 on land ownership).

Population attitudes consider women to be primarily responsible for most unpaid care and domestic work tasks. This is particularly true for tasks perceived as basic and routine unpaid care and domestic work. More than 60% of the population believes that tasks such as cooking for the household, cleaning the household, cleaning the bathroom/toilet and washing clothes are the exclusive responsibilities of women within the household. By contrast, less than 1% of the population considers these tasks to be the sole responsibility of men (Figure 3.11). Norms that ascribe gendered associations to these tasks also appear to be slightly more salient in rural settings than in urban areas, with the share of the population stating that these household tasks are the exclusive responsibility of women being higher in the former than in the latter.

From these attitudes emerges a gendered pattern according to which certain household tasks may be considered as feminine or masculine based on the traditional physical and emotional traits associated with men and women. Tasks that are more technical, such as repairing the house, are considered to be more suited to men, whereas those that are more socially oriented, such as caring for household members, are perceived to fall within the realm of women (OECD, 2021[36]). The association of certain tasks with women or men also depends on the physical or emotional skills required to accomplish them. For instance, focus group discussions revealed that tasks which require more physical force such as repairing the house and fetching water and firewood are considered to be masculine or shared, whereas tasks which require more emotional skills such as caring for the children are perceived to be more suitable for women (Figure 3.12).

These discriminatory attitudes are deeply entrenched and stem from customs and traditions transmitted from generation to generation. Controlling for various socio-demographic factors, women are more likely than men to believe that basic household tasks are the exclusive responsibility of women.23 This division of labour is normalised and considered acceptable by women. Focus group discussions clearly showed that the division of roles within the household is rigid with very little questioning around the allocation of responsibility, although women are conscious that the repartition of tasks is unequal. Failure to comply with the established norm is perceived as strange and abnormal. For instance, in qualitative interviews many women reported feeling embarrassed if they see a man performing what are perceived to be feminine household tasks (Mbuyita, 2021[23]). Men’s propensity to undertake unpaid care and domestic work is heavily related to whether they are married or not, highlighting the influence of traditional views on the role of married women. Married men are significantly less likely to perform basic unpaid care and domestic tasks than unmarried men, and the effect is stronger for single men than for men living with their partner. The results suggest a transfer of unpaid care and domestic duties to women’s shoulders upon marriage.

Imbalances between men and women start at a young age, strengthening the unequal gendered patterns associated with unpaid care and domestic work. Girls assist their mothers with domestic chores from a young age and to a greater extent than boys. For example, 91% of women aged over 15 years reported performing basic household tasks all the time or often when a teenager, compared to 63% of men. Likewise, 68% of surveyed individuals declared that girls aged between 7 and 15 years in their household perform basic household tasks all the time or often, compared to only 43% for boys (Figure 3.13, Panel A). Already tasked with a heavy unpaid domestic workload at a young age, girls internalise the intra-household division of responsibilities, and end up performing the majority of household chores as an adult. The consequences of this dynamic extend beyond an unbalanced distribution of unpaid care and domestic work, as highlighted during focus group discussions, with girls’ share of unpaid care and domestic work leaving them limited time to study and rest (Mbuyita, 2021[23]).

Evidence suggests that male role models who share basic household tasks on a more equitable basis may play a critical role in breaking the cycle of unequal distribution of unpaid care and domestic work. Men whose father undertook basic household tasks such as cooking, washing clothes, cleaning the house or the toilets/bathroom, and taking care of the elderly or the sick are more likely to take on household duties (Figure 3.13, Panel B).24 For instance, 65% of men whose father performed at least one basic household task all the time or often reproduced this pattern in their household. In contrast, only 50% of men whose father performed such tasks only sometimes, rarely or never, performed at least one basic household task all the time or often. Higher participation in unpaid care and domestic work by men’s fathers is also significantly associated with lower women-to-men ratios in unpaid care and domestic work.

Traditional views of women’s and men’s roles in the household also place men in charge of important decisions and expect women to obey their husbands. About three-quarters of Tanzania’s population hold the view that men should have the final word about important decisions in the home. Men are more prone to hold these discriminatory attitudes than women. Lack of education appears to be a strong and significant underlying factor in this regard.25 The effect of not holding discriminatory attitudes strengthens as education increases. Indeed, at the national level, 79% and 83% of the population with no formal education and an incomplete primary education, respectively, believe that a man should have the final word about important decisions in the home. Conversely, these shares drop to 65% for those that have completed secondary education and 56% for those that have obtained a tertiary-level diploma such as a university degree (Figure 3.14, Panel A).

Norms conferring decision-making power on men are strongly internalised by individuals. Focus group discussions revealed that women that do not comply with this norm are perceived as dangerous and may suffer from social stigma. Women who are perceived as vocal, educated and wealthy – attributes associated with being empowered enough to participate in decision-making processes – are likewise considered as a potential threat to marriage. Many men in the focus group discussions therefore admitted that they would be reluctant to get married to a woman presenting these characteristics (Mbuyita, 2021[23]). The few exceptional situations in which women may be allowed to exercise decision-making power within the household are when the husband is away on travel or when a woman is widowed and/or single and therefore lives alone.

Social norms that govern intra-household relationships between men and women task men with protecting and exercising guardianship over women in the household. In particular, married women are expected to seek their husband’s approval before leaving the private sphere of the household and going to public places. At the national level, 92% of the population believes that a woman should ask her husband or partner for permission if she wants to visit another city or travel abroad, and 86% thinks that she should ask permission before going to visit her own family. Overall, attitudes in favour of restrictions on women’s ability to go to public places without asking their husband’s permission are most acute when related to leisure activities, such as going to a sports field, cinema or a bar (Figure 3.14, Panel B). These deeply entrenched discriminatory attitudes are equally strong in rural and urban areas or in Mainland Tanzania and Zanzibar, and reveal a fundamental characteristic of restrictive masculinities in Tanzania – to be a “real” man, a man should protect and exercise guardianship over women in the household. In order to do so, social norms confer on men the ability to control women’s movements and their ability to take important decisions.

References

[26] Anyanwu, J. (2016), “Accounting for Gender Equality in Secondary School Enrollment in Africa”, African Development Review, Vol. 28/2, pp. 170-191, https://doi.org/10.1111/1467-8268.12188.

[29] Cabeza-García, L., E. Del Brio and M. Oscanoa-Victorio (2018), “Gender Factors and Inclusive Economic Growth: The Silent Revolution”, Sustainability, Vol. 10/2, p. 121, https://doi.org/10.3390/su10010121.

[9] Dausen, N. (2021), “Tanzania to allow students to restart education after giving birth”, Reuters, https://www.reuters.com/world/africa/tanzania-now-allow-students-resume-education-after-giving-birth-2021-11-24/ (accessed on 30 November 2021).

[27] Duflo, E. (2012), “Women Empowerment and Economic Development”, Journal of Economic Literature, Vol. 50/4, pp. 1051-1079, https://doi.org/10.1257/jel.50.4.1051.

[37] Erulkar, A. et al. (2020), “Designing and Evaluating Scalable Child Marriage Prevention Programs in Burkina Faso and Tanzania: A Quasi-Experiment and Costing Study”, Global Health: Science and Practice, Vol. 8/1, pp. 68-81, https://doi.org/10.9745/ghsp-d-19-00132.

[24] Gemignani, R. and Q. Wodon (2015), “Child Marriage and Faith Affiliation in Sub-Saharan Africa: Stylized Facts and Heterogeneity”, The Review of Faith & International Affairs, Vol. 13/3, pp. 41-47, https://doi.org/10.1080/15570274.2015.1075752.

[16] Girls not Brides (n.d.), About child marriage - Girls Not Brides, https://www.girlsnotbrides.org/about-child-marriage/ (accessed on 27 October 2021).

[13] Government of Tanzania (2019), Country Report on the Review and Progress made in Implementation of the Beijing Declaration and Platform for Action - Beijing +25, https://www.unwomen.org/sites/default/files/Headquarters/Attachments/Sections/CSW/64/National-reviews/United-Republic-of-Tanzania-en.pdf.

[30] Government of Tanzania (2019), Country Report on the Review and Progress made in Implementation of the Beijing Declaration and Platform for Action - Beijing +25, https://www.unwomen.org/sites/default/files/Headquarters/Attachments/Sections/CSW/64/National-reviews/United-Republic-of-Tanzania-en.pdf.

[8] Government of Tanzania (2016), The Written Laws (Miscellaneous Ammendments) Act (No.2) 2016.

[40] Government of Tanzania (2009), The Law of the Child Act, 2009.

[39] Government of Tanzania (1981), Penal Code, https://www.un.org/depts/los/LEGISLATIONANDTREATIES/PDFFILES/TZA_penal_code.pdf.

[43] Government of Tanzania (1978), The national education act, 1978.

[38] Government of Tanzania (1971), Law of Marriage Act, https://rita.go.tz/eng/laws/History%20Laws/Marriage%20Ordinance,%20(cap%2029).pdf.

[7] Government of Zanzibar (2017), National plan of action to end violence against women and children 2017-2022, https://www.unicef.org/tanzania/sites/unicef.org.tanzania/files/2018-10/Tanzania-2017-Zanzibar-National-Plan-of-Action-on-VAWC.pdf (accessed on 27 October 2021).

[5] Harper, C. et al. (2014), Unhappily ever after Slow and uneven progress in the fight, https://cdn.odi.org/media/documents/Unhappily_ever_after_the_fight_against_early_marriage.pdf.

[41] High Court of Tanzania (2017), Civil Appeal No. 204 of 2017, https://d3n8a8pro7vhmx.cloudfront.net/equalitynow/pages/1911/attachments/original/1571911712/20191023145222.pdf?1571911712.

[31] Human Rights Watch (2021), Tanzania: Pregnant Student Ban Harms Thousands, https://www.hrw.org/news/2021/10/06/tanzania-pregnant-student-ban-harms-thousands (accessed on 30 November 2021).

[3] Izugbara, C. (2018), “Spousal age differences and violence against women in Nigeria and Tanzania”, Health Care for Women International, Vol. 39/8, pp. 872-887, https://doi.org/10.1080/07399332.2018.1426001.

[35] Jayachandran, S. (2015), “The Roots of Gender Inequality in Developing Countries”, Annual Review of Economics, Vol. 7/1, pp. 63-88, https://doi.org/10.1146/annurev-economics-080614-115404.

[42] Jeppie, S., E. Moosa and R. Roberts (n.d.), Muslim Family Law in Sub-Saharan Africa. Colonial Legacies and Post-Colonial Challenges, Amsterdam University Press, Amsterdam, https://hdl.handle.net/1887/15230.

[28] Klasen, S. and F. Lamanna (2009), “The Impact of Gender Inequality in Education and Employment on Economic Growth: New Evidence for a Panel of Countries”, Feminist Economics, Vol. 15/3, pp. 91-132, https://doi.org/10.1080/13545700902893106.

[4] Lee-Rife, S. et al. (2012), “What Works to Prevent Child Marriage: A Review of the Evidence”, Studies in Family Planning, Vol. 43/4, pp. 287-303, https://doi.org/10.1111/J.1728-4465.2012.00327.X.

[32] Lowes, S. and N. Nunn (2018), “Bride Price and the Well-Being of Women”, in Anderson, S., L. Beaman and J. Platteau (eds.), Towards Gender Equity in Development, Oxford University Press, Oxford, http://www.wider.unu.edu (accessed on 4 November 2021).

[23] Mbuyita, S. (2021), SIGI Tanzania Qualitative Report, https://www.oecd.org/development/gender-development/.

[6] Ministry of Health, Community Development, Gender, Elderly and Children (2016), National Plan of Action to End Violence against Women and Children 2017/18-2021/22, https://www.unicef.org/tanzania/media/496/file/tanzania-2016-NPA-VAWC.pdf.

[36] OECD (2021), Man Enough? Measuring Masculine Norms to Promote Women’s Empowerment, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/6ffd1936-en.

[10] OECD (2021), SIGI 2021 Regional Report for Africa, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/a6d95d90-en.

[15] OECD (2021), SIGI Tanzania database, https://stats.oecd.org/.

[1] OECD (2019), SIGI 2019 Global Report: Transforming Challenges into Opportunities, Social Institutions and Gender Index, OECD Publishing, Paris, https://dx.doi.org/10.1787/bc56d212-en.

[44] OECD Development Centre/OECD (2019), “Gender, Institutions and Development (Edition 2019)”, OECD International Development Statistics (database), https://dx.doi.org/10.1787/ba5dbd30-en (accessed on 20 December 2021).

[17] OHCHR (n.d.), Child, early and forced marriage, including in humanitarian settings, https://www.ohchr.org/EN/Issues/Women/WRGS/Pages/ChildMarriage.aspx (accessed on 12 February 2020).

[34] Schaffnit, S. et al. (2019), “Parent–offspring conflict unlikely to explain ‘child marriage’ in northwestern Tanzania”, Nature Human Behaviour, Vol. 3/4, pp. 346-353, https://doi.org/10.1038/s41562-019-0535-4.

[25] Schaffnit, S. et al. (2020), “When marriage is the best available option: Perceptions of opportunity and risk in female adolescence in Tanzania”, Global Public Health, Vol. 16/12, pp. 1820-1833, https://doi.org/10.1080/17441692.2020.1837911.

[12] UNFPA (2020), Impact of the COVID-19 Pandemic on Family Planning and Ending Gender-based Violence, Female Genital Mutilation and Child Marriage, United Nations, https://www.unfpa.org/sites/default/files/resource-pdf/COVID-19_impact_brief_for_UNFPA_24_April_2020_1.pdf (accessed on 25 August 2020).

[14] UNICEF (2021), Child marriage data, https://data.unicef.org/topic/child-protection/child-marriage/# (accessed on 20 December 2021).

[11] UNICEF (2021), COVID-19: A threat to progress against child marriage, https://data.unicef.org/resources/covid-19-a-threat-to-progress-against-child-marriage/ (accessed on 7 September 2021).

[2] UNICEF (n.d.), Child marriage, https://www.unicef.org/protection/child-marriage.

[19] United Nations (2019), “World Marriage Data 2019 (POP/DB/Marr/Rev2019)”, Department of Economic and Social Affairs, Population Division, https://population.un.org/MarriageData/Index.html (accessed on 30 November 2021).

[18] United Nations Statistics Division (2021), SDG Indicators, https://unstats.un.org/sdgs/metadata/?Text=&Goal=5&Target=5.3 (accessed on 3 November 2021).

[33] Voena, A. and L. Corno (2016), “Selling daughters: age of marriage, income shocks and the bride price tradition”, Working Paper Series, IFS, https://doi.org/10.1920/wp.ifs.2016.1608.

[22] Wang, H. et al. (2013), “Economic Impact of Maternal Death on Households in Rural China: A Prospective Cohort Study”, PLoS ONE, Vol. 8/10, p. e76624, https://doi.org/10.1371/journal.pone.0076624.

[21] Whetten, R. et al. (2011), “Child work and labour among orphaned and abandoned children in five low and middle income countries”, BMC International Health and Human Rights, Vol. 11/1, https://doi.org/10.1186/1472-698x-11-1.

[20] WHO (2020), Adolescent pregnancy, https://www.who.int/news-room/fact-sheets/detail/adolescent-pregnancy (accessed on 27 October 2021).

In Mainland Tanzania, the Law of Marriage of 1971 (revised in 2002 as the Law of Marriage Act) defines marriage as the voluntary union of a man and a woman intended to last for their joint lives (Government of Tanzania, 1971[38]). While the Law of Marriage Act requires voluntary consent to enter into marriage, it permits a marriage contract to be concluded without the consent of the bride on the basis of an agreement between the guardian of the bride and the groom. The law recognises two types of marriages – monogamous and polygamous – depending on whether the same are contracted according to customary, religious or civil rite practices. Whereas a monogamous marriage is defined as a union between one man and one woman, a polygamous marriage is “a union in which the husband may, during the subsistence of the marriage, be married to or marry another woman or women”.

Under the Law of Marriage Act, the legal minimum age for marriage is 15 years for girls and 18 years for boys, although the law permits girls and boys as young as 15 years to marry on the basis of a court order. Furthermore, the Penal Code permits the marriage of girls aged less than 15 years, provided that the marriage is not consummated before the age of 15 (Government of Tanzania, 1981[39]). The Law of the Child Act (2009) does not stipulate a legal age of marriage or prohibit child marriages (Government of Tanzania, 2009[40]). In 2016, the High Court of Tanzania ruled against child marriage in a landmark case declaring child marriage provisions within the Law of Marriage Act unconstitutional. A government appeal against the decision of the High Court was overturned by the Court of Appeal, thus upholding the High Court’s ruling (High Court of Tanzania, 2017[41]). Efforts are underway to review the Law of Marriage Act (1971) to raise the age of marriage of girls from 14 to 18 (Ministry of Health, Community Development, Gender, Elderly and Children, 2016[6]). Meanwhile, in 2016 Tanzania amended the Education Act to prohibit child marriage involving children enrolled in primary or secondary school (Government of Tanzania, 2016[8]).

In Zanzibar, religious “Kadhi” courts essentially oversee all marriages in Zanzibar, although their intended purpose is to oversee legal matters among Zanzibar’s Muslim Population (more than 90% of the population in Zanzibar is Muslim) (Jeppie, Moosa and Roberts, n.d.[42]). Zanzibar does not have a minimum age for marriage.

Among Tanzania’s adult population, 63% are married, 10% live together and 27% are single. The majority of marriages in Tanzania are religious and/or customary. The average age of marriage among women (20.2 years) is more than five years younger than the average age for men (25.4 years), but varies according to the nature of the marriage. At the national level, the average age of first marriage is 19.5 years among women in customary marriages, 20.5 years for women in religious marriages and 22 years for women in civil marriages.

Marriage is perceived as a core social institution by the majority of men and women in Tanzania. More than 60% of Tanzania’s population agree that in order to be a “real” woman or a “real” man, one needs to be married. The data suggest that being married as a woman is more important in rural than urban areas, where 65% and 57% of the population, respectively, consider marriage essential for being a “real” woman. In a context where marriage is socially valued, individuals – and particularly women – may see marriage as a way to improve their social standing.

Notes

← 1. In 2019, the female-to-male ratio of time spent on unpaid, domestic, care and volunteer work in a 24-hour period stood, on average, at three. This means that in East African countries where data was available, women undertook spent three times more time on unpaid care and domestic work. Countries with available data are: Ethiopia, Kenya, Madagascar, Mauritius, Rwanda, Uganda and Tanzania (OECD Development Centre/OECD, 2019[44])

← 2. “Restrictive masculinities” refers to masculinities that confine men to their traditional role as the dominant gender group, undermining women’s empowerment and gender equality.

← 3. Child marriage refers to any formal marriage or informal union between a child under the age of 18 and an adult or another child (UNICEF, n.d.[2]). As child marriage disproportionality concerns girls – worldwide and in Tanzania – the term “girl child marriage” is used throughout the chapter to distinguish the practice from boy child marriage, when necessary. Otherwise, child marriage refers to girl child marriage.

← 4. Tanzania’s Education Act No. 25 of 1978 and its amendment CAP 353 R.E. of 2002 do not explicitly ban pregnant girls from attending school and returning to school after childbirth (Government of Tanzania, 1978[43]). Expulsion is permitted when a student has “committed an offence against morality” or has “entered into wedlock”. Ministry and school officials have interpreted pregnancy as an offense against morality leading to the school ban of pregnant girls (Human Rights Watch, 2021[31]).

← 5. At the national level, the average rate for boy child marriage is 1.7% (men aged 15+ years). This rate is higher in rural than in urban areas (2.4% and 0.3%, respectively).

← 6. In comparison, the 2015 DHS survey conducted in Tanzania estimated that 24.4% of women aged 15-19 years old had already been married, widowed, divorced or living together with their partner (United Nations, 2019[19]).This measure differs slightly from the one used by the SIGI Tanzania and SDG indicator 5.3.1.

← 7. Kigoma (25%), Kusini Pemba (26%), Mara (26%), Shinyanga (27%), Kagera (28%) and Simiyu (31%).

← 8. Girl child marriage for women aged between 15 and 19 years should not be compared to other age brackets and should be considered carefully. As not all women in this age bracket have reached 18 years old, the share of women married before the age of 18 years can still increase. This is not the case for other age brackets where women are older than 18 years.

← 9. Morogoro (40%), Arusha (41%), Tabora (41%), Mbeya (42%), Songwe (42%), Katavi (43%), Kaskazini Pemba (46%), Geita (49%) and Shinyanga (52%).

← 10. The results are based on an OLS regression. The dependent variable is age at first pregnancy and the main independent variable is age at marriage. Control variables include age, level of education, wealth quintiles (based on a set of measures including type of construction material, distance to water and distance to cooking energy) and regions. Coefficients and the marginal effects of age at marriage are significant at 1%.

← 11. The results are based on a probit model measuring the women’s likelihood of being pregnant as an adolescent (i.e. under the age of 20 years). The main independent variable is child marriage. Control variables include residential status (urban or rural), age, age squared, education level, a set of measures of household wealth and regions. The coefficients and the marginal effects for child marriage (i.e. married before the age of 18 years) are significant at 1%.

← 12. The results are based on an OLS regression. The dependent variable is education level, and the main independent variable is child marriage. Control variables include sex, residential status (urban or rural), residency (Mainland Tanzania or Zanzibar), employment status in agriculture, mother’s education level, father’s education level and a set of measures of household wealth. Coefficients and the marginal effects of child marriage are significant at 1%.

← 13. The results are based on an OLS regression. The dependent variable is girl child marriage, measured as the share of women aged 15 years or older who were married before the age of 18 years. The main independent variable is education level measured via four distinct variables: incomplete primary education level, complete primary education level, complete secondary education level and university level. Control variables include age, residential status (urban or rural), residency (Mainland Tanzania or Zanzibar) and a set of measures of household wealth. Coefficients and marginal effects for complete primary or secondary education level are significant at 1%. Coefficients and marginal effects for incomplete primary education level or university level are not significant.

← 14. In Tanzania, 14% of the population has a level of secondary education and only 2% have a level of tertiary education. Among Tanzanian women, 11% have a level of secondary education and 1% has a level of tertiary education, compared to 17% of men for secondary and 2% for tertiary education.

← 15. Evaluations of programmes targeting child marriage and education in Africa have found that providing parents with economic incentives – monetary or in-kind such as in the form of a goat – is an effective measure to end or, at least, delay child marriage (Lee-Rife et al., 2012[4])

← 16. The results are based on an OLS regression. The dependent variable is girl child marriage. The main independent variable is attitudes towards girl child marriage, measured as the share of the population which states that the appropriate minimum age for marriage for girls is below 18 years. Control variables include residential status (urban or rural), residency (Mainland Tanzania or Zanzibar), education level and a set of measures of household wealth. Coefficients and the marginal effects of attitudes towards girl child marriage are significant at 10%.

← 17. Morogoro (3%), Kusini Unguja (5%), Iringa (5%), Mjini Magharibi (5%), Dar es Salaam (6%), Lindi (7%) and Kaskazini Unguja (9%).

← 18. Kaskazini Pemba (35%), Kagera (37%), Kigoma (50%) and Simiyu (52%).

← 19. Geita, Lindi, Mbeya, Mtwara, Mwanza, Njombe, Ruvuma, Shinyanga and Songwe.

← 20. Kusini Pemba, Kaskazini Pemba, Manyara and Pwani.

← 21. The results are based on a probit model measuring women’s likelihood of being married as a child. Decision maker over marriage is the main independent variable. Control variables include residential status (urban or rural), age, age squared, education level, a set of measures of household wealth and regions. The girl herself as decision maker over her own marriage is the omitted variable. Coefficients and marginal effects for the father as decision-maker, and for both parents as decision makers are significant at 1%. Coefficients and marginal effects for the mother as decision-maker are not significant.

← 22. The 16 household tasks for which data were collected by the SIGI Tanzania survey were: cooking for the household, cleaning the house, washing clothes, cleaning the bathroom/toilet, taking care of the elderly, taking care of the sick or disabled, repairing the house, taking care of the yard, buying the groceries, fetching water for the household, fetching firewood for the household, leisure (culture, hobbies, sport), supervising the children’s homework, playing or doing sport with the children, bathing children and feeding children.

← 23. The results are based on a series of probit models measuring the population’s likelihood to consider unpaid care and domestic tasks as the exclusive responsibility of women. In total, attitudes are tested for eight activities – the six basic unpaid care and domestic work tasks listed in Box 3.4 as well as bathing children and feeding children. For each probit model, sex is the main independent variable. Control variables include residential status (urban or rural), age, age squared, size of the household, marital status, education level, a set of measures of household wealth and regions. Coefficients and marginal effects for sex are significant at 1% in all eight models.

← 24. The results are based on a probit model restricted to men only and measuring their likelihood of participating in at least one basic household task (cooking for the household, cleaning the house, washing clothes, cleaning the bathroom/toilet, taking care of the elderly, and taking care of the sick or disabled). The main independent variable are (i) whether the father or man who cared for the household during childhood participated all the time or often (or not) in at least one basic household tasks, and (ii) whether the mother or woman who cared for the household during childhood participated all the time or often (or not) in at least one basic household tasks. Control variables include residential status (urban or rural), age, age squared, size of the household, marital status, education level, a set of measures of household wealth and regions. Coefficients and marginal effects for the father’s participation in household tasks are significant at 1%. Coefficients and marginal effects for the mother’s participation in household tasks are not significant.

← 25. The results are based on a probit model measuring the population’s likelihood to consider that men should have the final word about important decisions in the home. Independent variables include sex, residential status (urban or rural), age, age squared, size of the household, marital status, education level, a set of measures of household wealth and regions. Coefficients and marginal effects for sex are significant at 1%. Coefficients and marginal effects for primary education complete are significant at 5% and coefficients and marginal effects for complete secondary education and university education are significant at 1%.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.