Child development accounts (CDAs): An asset-building strategy to empower girls in Uganda

Abstract

This study explores an innovative intervention for orphaned children in Uganda. It combines standard health care with an economic empowerment component. We refer to this combination as a family asset-based intervention, which provides each child with a child development account (CDA), a matched savings account for secondary schooling; financial education; and a mentor. This article examines the educational outcomes of the girls in this study. The results from the first two waves of the study indicate that CDAs have the potential to begin to help negate the effects of past gender inequalities and to help provide a path for young girls to move forward.

Introduction

Although great strides have been made in reducing gender inequality in many African countries, women still face higher rates of poverty compared to their male counterparts. Without access to opportunities to further their education and accumulate productive assets in order to become self-reliant, women will continue to disproportionately experience poverty in Africa (African Union, 2004; Ssewamala, 2004). This article describes a family asset-based intervention program, Suubi-Uganda, which, among several other intervention components, included child development accounts (CDAs) that provided matched savings accounts to HIV/AIDS-orphaned children. A child development account is a matched savings product, held in the child’s name, housed at a formal financial institution that facilitates monetary savings and the accumulation of assets. In the Ugandan case, the funds accumulated in CDAs constitute a tangible asset that can provide poor orphaned children with a means to expand their life options if they are provided with the training and skills to use them effectively on their own behalf (Ssewamala, 2005a; Ssewamala et al., 2008). Specifically, the article reports on educational outcomes for girls who participated in the study, and demonstrates how CDAs, over and above the current educational policies in Uganda, may be an effective strategy in addressing the intersection of gender inequality and poverty in a low-resource setting.

The study, Suubi, took place in Uganda, a sub-Saharan African country where gender inequality is ubiquitous (Ministry of Finance, Ugandan Government, 2005). In a national commitment to address gender inequality brought about by history, cultural tradition, and other factors, the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW) was passed in 1995. The National Gender Policy established in 1997 and revised in 2005, also attempts to reconcile issues that exacerbate the effects of gender inequalities (The Republic of Uganda, 2007).

Yet, even with these current human rights instruments and measures in place, gender inequality in Uganda still leaves girls and women at a disadvantage in terms of available opportunities and choices in numerous economic and social situations, including the areas of HIV prevalence, literacy levels, access to justice, financial services, employment and overall access to productive assets, such as land (The Republic of Uganda, 2007). To compound this problem, Uganda has a high percentage of orphans due to a combination of the HIV/AIDS epidemic and the recent civil conflicts in the northern part of the country which has left many vulnerable children, particularly orphan girls, at an even greater risk of contracting HIV/AIDS and living in poverty (Ssewamala and Ismayilova, 2008). Due to the increasing number of orphaned children and poverty, traditional care by the extended family system is overwhelmed. Families are unwilling to take in more children with the fear of additional costs associated with their health and educational needs. This leaves many orphaned children, including girls, to care for themselves and their younger siblings.

Two of the most important factors associated with breaking the cycle of poverty and leading to equality and empowerment for girls are education (Kasente, 2003) and access to productive assets. Providing these opportunities when girls are still young increases their life chances in the future. In line with CEDAW, providing access to education through a public sponsored program, particularly beyond primary school – when free and universal compulsory education ends in Uganda – will help ensure increased opportunities for education and access to asset building opportunities in the future for girls, especially as they move through adolescence into adulthood.

Education and gender in Uganda

In 1997, the Government of Uganda established a policy of Universal Primary Education (UPE), which, at the time, entitled up to four children per family to receive free primary education (grades 1 through 7). The program was later extended to enable all children of school-going age in each family to access free primary education. As a direct result of the UPE initiation, primary school enrollment significantly increased from approximately 2.5 million (the number reported in 1996) to 6.8 million in 2000 (Kisubi, 2008). The first year (1997) saw a 123 percent increase in primary school enrollment. The total primary school enrollment has been growing since then and is currently estimated at 8.7 million children (The Republic of Uganda, 2010). As these numbers changed, the gender gap in school enrollment started narrowing as well. To illustrate, because of the UPE initiative, girls’ enrollment increased from 46 percent to 50 percent between 1996 and 2003 (MoES, 2004). Nationwide, the gross enrollment ratio gap was reduced to just 1 percent (121% boys versus 120% girls) between eligible boys and girls who were attending primary school in 2009 (The Republic of Uganda, 2010). This marked positive progress at the primary school level.

However, when students reach secondary school and fees are instituted, attendance decreases and dropout rates increase significantly, particularly for girls coming from resource-constrained families. When families have limited resources, girls are more often withheld from secondary school compared to boys (UNAIDS, 2006). For example, the ratio of female to male secondary school enrollment in Uganda is 83 percent compared to the primary ratio of 100 percent (UNICEF, 2010) – a 17 percentage point difference. Furthermore, secondary school dropout rates are higher for girls (15.3%) than boys (11.7%). The cost of secondary schooling is a major factor in dropouts. Specifically, when cost is an issue in a family and the choice is between sending a boy or a girl to school, the boy is usually given the opportunity (Kasente, 2003; Ssewamala et al., 2010, 2011).

Other reasons that lead to gender disparities include early pregnancies and early marriages that increase the school dropout rates among girls; and the prevalence of socio-cultural norms such as the expectation that girls will eventually become mothers and housewives, hence girls’ access to education is not a priority at the family level, particularly in developing regions (Atekyereza, 2001). Related to that, parents or caregivers generally prefer sending boys to school since the education benefits for a girl are enjoyed by the family into which she marries and not her family of origin (Atekyereza, 2001). Additionally, the HIV/AIDS pandemic in sub-Saharan Africa and the resulting illnesses and deaths of the adults in the household, create the need for girls to stay home from school to help take care of sick relatives and smaller siblings, and perform overall household duties in the context of ill or deceased caregivers. Safety concerns for girls who have to walk long distances are an issue in a country where gender-based violence is common (Kasente, 2003). Add these barriers to the constant financial inability to pay school-related fees, girls are left with few resources that enable them to attend school and improve their lives.

Recognizing a need to decrease the hardship of paying school fees at the secondary level and to address the increasing number of children who successfully complete primary education and are prepared to continue to the next level, the Government of Uganda introduced the policy of Universal Post-Primary Education and Training (UPPET) which took effect in 2007. The policy includes a small number of selected government aided secondary and private schools and business, technical, vocational education and training institutes. Similar to UPE, the government pays tuition to qualifying students and provides selected instructional materials to schools. The policy was introduced starting with UPE students who completed Primary Leaving Examinations (PLE) in 2006, so most of the children who had completed primary school before 2006 did not qualify. Moreover, at the time of this writing, the policy is being implemented in select (and willing to participate) secondary schools. Current secondary school enrollment is estimated at over 1.5 million children compared to only 900,000 in 2005/2006 before the introduction of UPPET (UBOS, 2010). The policy has also helped reduce the gender disparity in secondary education from the ratio of 0.79 in 2000 to 0.84 in 2009 (The Republic of Uganda, 2010). However, because UPPET is only in a few selected schools, the majority of parents must find ways of financing secondary education. Thus, many eligible youth in Uganda are still not attending secondary school due to the high cost – the majority of whom are girls (see Ssewamala et al., 2010). Given the low secondary school enrollment rate among girls in Uganda and the delay in prioritizing universal secondary education or secondary education subsidies at the national level, alternate programs are needed, such as public-private partnerships, in order to mitigate the low post-primary education levels among girls and young women in Uganda.

The remainder of this article describes the intervention program that provided, among other components, matched savings accounts (child development accounts) to HIV/AIDS-orphaned primary school children in Rakai District of southern Uganda (The Rakai District Local Government, 2010). Specifically, this article reports on the effects of child development accounts on educational outcomes for girls who participated in the study, Suubi-Uganda. The article will also demonstrate how CDAs, over and above the current educational policies in the country such as UPE and UPPET, may be an effective strategy aimed at increasing the attendance of girls in secondary school. When girls are educated, an entire country’s economy is strengthened (Chaaban and Cunningham, 2011). At the individual and family level, educated girls will see increased access to resources which may help reroute their trajectory from a life of poverty and dependence, to a life of opportunity and self-reliance.

Asset theory

Asset theory (Sherraden, 1991) suggests that assets have important psychological, social and economic benefits for individuals and families. Assets can be used in times of financial hardships and can be passed on to future generations. Owning assets gives people a sense of stability and allows them to widen their vision on possible opportunities. It provides personal efficacy and self-esteem and gives people a feeling of social connectedness. These feelings are internalized and help form a person’s individual behavior (Sherraden, 1991). Research shows that the advantages of asset ownership are numerous and multi-dimensional, including fighting poverty, promoting positive social, health, and economic behavior, and increasing future orientation (Celia, 1994; Page-Adams and Sherraden, 1997; Ssewamala and Ismayilova, 2008, 2009; Yadama and Sherraden, 1996; Zhan and Sherraden, 2003).

Child development accounts (CDAs): Educational opportunities for Ugandan girls

CDAs can provide children with some financial resources by which they can begin to realistically plan for their post-primary education or job/skills training. If children do not see that they have the financial means with which to begin to pursue long-term educational or vocational aspirations, psychosocial interventions such as counseling or simply food aid, which is the current reactive program from the government and several non-government organizations to orphaned and vulnerable children, may have little effect (Ssewamala and Curley, 2006; Ssewamala et al., 2008).

To illustrate the possible effects of CDAs, asset theory would predict a child in primary school with the belief that he/she has the economic means to afford post-primary education is less likely to drop out of school. Hence, provided with the economic means or a stake in the formal financial system, this child may think and behave with a positive future orientation. Envisioning a concrete possibility for his/her future, this child may act as if he/she will have a future worth living (Sherraden, 1986). This child would be more likely to stay in school, strive to earn good grades, avoid health-related risks, and may eventually become a productive member of society. An intervention aimed at promoting asset-ownership and economic empowerment would more likely create a reciprocal cycle in which asset accumulation and positive social behaviors will be mutually reinforcing (Ssewamala, 2005b).

The Suubi Project

The Suubi Project, which stands for hope in Luganda (one of the widely spoken languages in Uganda), provides orphaned children with a three component program consisting of: 1) workshops that focus on financial education, asset building, and career planning; 2) mentorship from near-peers to reinforce learning; and 3) a joint CDA in both the child’s and caregiver’s name. The combination of these three components is referred to as a family-based economic empowerment intervention. The strategy behind this multi-feature intervention is to create and broaden asset-ownership opportunities for orphaned children and their caregiving families. The expectation is that by creating asset-ownership opportunities for children and their caregiving families, before they are forced to separate from one another, family breakdown, especially related to gender, can be reduced, and several child-related negative outcomes, including school dropouts, can be minimized. This approach recognizes that keeping children within their community, within their family network, and within a household – is preferable compared to other options in terms of providing effective and sustainable care for orphaned children. The option provides children with continuity, emotional support, and an opportunity for them to develop within their culture and traditions. This approach is also supported by the United Nations Convention on the Rights of the Child, which asserts that a child, for the full and harmonious development of his or her personality, should grow up in a family environment (United Nations, 1989). Asset building within the family or caregiver household can help achieve this goal.

Three types of community institutions worked in collaboration with the researchers and were key to the implementation and maintenance of the CDA program in Uganda: a local faith-based institution (Matale Parish within the Catholic Diocese of Masaka), 15 rural and semi-urban primary schools (throughout Rakai District), and two financial institutions: Centenary Rural Development Bank (hereafter, Centenary Bank), and Development Finance Company of Uganda (hereafter, DFCU). The faith-based institution’s role was to help locate schools that might be willing to participate in the program; they helped mobilize the community to receive and support the program; and they acted as mediators with people of authority such as local and central government leaders (including several church leaders and district leaders).

The schools helped identify the orphaned children; helped locate the villages where the children were residing and provided other relevant information about the children and their caregiving families. Schools also provided the participant’s academic performance and attendance records for the data collectors, plus a safe space for meetings between the participants and the project staff, including the mentorship meetings. Finally, the banks provided the financial education to the participants and their caregivers, including money management techniques; opened the savings accounts for the participants; and provided quarterly statements to the participants at no cost.

Project design

A total of 286 AIDS-orphaned children were chosen from 15 pre-selected schools to participate in the study. This analysis uses a subset of that study population, specifically only girls. The sample for this analysis consists of 157 girls with an average age of 14 years. All participants were in the first academic term of the last two years of primary school, 6th and 7th grades (an equivalent of 7th and 8th grade in the US system). There are three school terms in an academic year.

All the schools from which the girls were selected were matched on the basis of academic performance and were all rural UPE schools located in the study area: Rakai District of southern Uganda. None of the schools were part of the UPPET program. Each of the schools were randomly assigned to the treatment (n = 10 schools) or control group (n = 5 schools). All children from a particular school were placed in the same study group in an effort to guard against contamination. Schools were on average approximately 13 kilometers apart. All children who self-identified as AIDS-orphans, having lost one or both parents to AIDS were included in the program. Participants’ caregivers/family members were invited to the schools for an informational meeting using flyers and letters that were sent to the selected students’ homes. The informational meetings took place in the children’s schools. The meetings were attended by the selected children, their caregivers, the school headteachers, and the parish priests (the study community collaborators). The caregivers who attended the informational meetings and expressed interest in having their children participate in the study were asked to sign a consent form authorizing their children to participate in the study. In addition, each child (individually) had to express an interest in participating. Each child was consented separately from his/her caregivers.

The children in the control group received existing orphan care services (also known as usual care) consisting of support and counseling from faith-based organizations in the target community plus school supplies, including exercise books and textbooks from the intervention. The children in the treatment group received, in addition to the usual care and the school supplies from the intervention, an asset-based family intervention – which, as mentioned earlier, included a CDA; workshops that focused on asset building, including how to save money, career planning, in addition to a monthly mentorship session with peer mentors on future planning and life options.

CDAs were held in a joint account with both the child’s and caregiver’s name in two well established and recognized banks in the country: Centenary Bank and DFCU Bank. Specifically, Centenary Bank is one of the most recognized financial institutions working with low-income families in Uganda. The initial account opening deposit was made by the program and the child was expected to make deposits each month in order to receive the match. Any of the child’s family members, relatives, or friends were allowed, and indeed encouraged, to make deposits into the CDAs. The account was then matched with money from the program. The match cap (the maximum amount of family contribution matched by the intervention program) was set at an equivalent of US$10 a month per family, or US$120 for each year during the study period. The match rate was 2:1 – meaning that if a child saved an equivalent of $10, he/she would be matched by $20. The exchange rate for the US dollar was approximately 2500 Uganda shillings at the time of the study. After one year, the amount of savings accumulated would be enough to pay for at least two years of secondary school education in an average rural school, and pay for a school uniform.

Methods

Data and sample

This study uses longitudinal data from two points in time – data collected at baseline/pre-Suubi intervention (herein referred to as Wave 1) and 10–12 months post-Suubi intervention initiation (herein referred to as Wave 2) – to examine how the Suubi Project influenced educational outcomes among the girls participating in the program. Specifically, educational planning and confidence of plan were examined – over time – to determine any changes and whether girls in the treatment group showed different patterns of change/growth in their plans and confidence compared to girls in the control group following the intervention. The experimental group has 83 participants and the control group has 74.

Study limitations

As part of the inclusion criteria, recruited orphaned girls had to be enrolled in primary school at the time of recruitment in order to participate in the study. Girls enrolled in school may have different characteristics from orphaned girls who were not enrolled in school. As a result of these inclusion criteria, we do not know how the Suubi Project would have affected out-of-school orphaned girls. Second, Suubi focused on orphaned children enrolled in rural schools, and the findings may have differed if urban schools had been included in the cluster randomization. Third, the study results are based on two data points. If data were analyzed from a relatively longer period with additional data points, the results may be different. And finally, due to the nature of the questions asked in the interviews, more specific data were not available to assess a more direct connection between the program and the girls’ aspirations to stay in school.

Measures

Key outcome variables of this study are educational planning and confidence of achieving the educational plan. First, at both waves, the survey asked ‘What are your educational plans after A level (high school)?’ Participants selected from four responses: ‘I have no plans’, ‘Vocational/technical or job training’, ‘Go to a college which awards diplomas’, and ‘Go to the university for a degree’. Change scores in the educational planning between Wave 1 and 2 were calculated by subtracting the educational plan at Wave 1 from the measure at Wave 2. A positive score indicates a positive desired change in the educational plan at Wave 2.

Second, following the specific educational plans question, the children were asked how sure they were that they would achieve their plan. The children had a three-point scale response choice with 1 being the least confident and 3 being the most confident. Similar to the educational plan measure, a change score for the confidence was calculated (Confidence at Wave 2 – Confidence at Wave 1).

Results

Demographic characteristics

The ages for girl participants ranged from 11 to 17 years with an average age of 14 years. The average family size is 6.5 people with 3.5 children per family. Forty-six percent of the girls have only their mother living, while 17 percent report only their father living. Thirty-seven percent have neither their mother nor father living – hence they are defined as double orphans. In regards to financial support, biological parents support 33 percent of the families in the sample; grandparents support 29 percent; and uncles, aunts, and other relatives support 38 percent of the families. Homeownership (referring to very modest homes compared to Western or middle-income country standards) is reported by approximately 87 percent of families (see Table 1).

Table 1.

Descriptive and bivariate results of sample characteristics.

Variables	Total sample (n = 157)	Treatment group (n = 83)	Control group (n = 74)	t-test or χ2
Age	13.6	13.7	13.5	2.59
	(1.28)	(1.19)	(1.36)
Type of orphan %
Only father living	17	21	12	2.16
Only mother living	46	46	45	0.01
Both parents deceased	37	33	41	0.86
Number of adults in household
Household	6.5	6.6	6.3	3.36
	(2.82)	(2.98)	(2.56)
Number of children in household	3.45	3.11	3.77	8.83
	(2.47)	(2.92)	(1.78)
Source of financial support %
Parent	33	42	24	7.08**
Grandparents	29	27	31	0.78
Others	38	31	45	3.08
Home ownership %	87	82	93	4.63*

^*p ≤ .05;

^**p ≤ .01.

Comparing demographic characteristics between the girls in the treatment group and the girls in the control group, analyses locate no significant differences in age, household size, number of children in a household, and orphan status. However, girls in the treatment group were significantly more likely to be financially supported by one of their biological parents (χ² [1, n = 157] = 7.08, p = .001). On the other hand, girls in the control group were more likely to live in families that owned their own homes (χ² [1, n = 157] = 4.63, p = .03) (see Table 1). We are not sure how these baseline differences influenced the overall participation of the girls in each of the two groups. Given the relative importance of homeownership, one would assume that girls in the control group may have been slightly better off on that specific (homeownership) variable.

Savings outcomes

Eighty-three girls in the program saved a combined amount of US$3729 (not including the 2:1 match) during the study period. The total amount with the match was US$11,187. This averages to about 53 percent of the available match – the maximum amount that the program would match. Breaking the amounts down by participants, the average quarterly deposit (AQD), which is equal to the net deposit per quarter of participation for a participant is equal to US$19 per quarter or US$6.33 a month. The mean monthly deposit with a match (provided by the Suubi Project) was US$19 a month. This amount was equivalent to the savings amount for the program as a whole.

Educational outcomes

Educational plans

Table 2 represents the changes in the girls’ educational plans between Wave 1 and Wave 2 in the treatment group. For the treatment group, there is a 33 percent change to a more positive response in educational plans and a 13 percent change to a more negative response. Fifty-five percent of girls in the treatment group reported no changes to their plans over the study period. The control group, in comparison, had a 27 percent change to a more positive response and a 21 percent change to a more negative response. Fifty-one percent of the control group made no changes. The difference in the percentage change to a positive response is moderately higher for the treatment group compared to the control group (33% vs 27%, respectively). The same holds true for the percentage change to a negative response between the treatment and control group (13% vs 21%, respectively).

Table 2.

Educational plans of orphan girls.
	Educational plans W1	Educational plans at W2				Total
		No plans	Voc/tech training	Attend college for diploma	Attend univ. for degree
Treatment group	No plans	0	1	1	6	8
	Voc/tech training	1	0	0	6	7
	Attend college for diploma	1	0	2	12	15
	Attend univ. for degree	2	4	2	42	50
	Total	4	5	5	66	80
Control group	No plans	5	6	1	5	17
	Voc/tech training	1	1	0	2	4
	Attend college for diploma	1	0	1	6	8
	Attend univ. for degree	4	4	6	31	45
	Total	11	11	8	44	74

Educational plans of orphan boys.
	Plans W1	Educational plans at W2				Total
		No plans	Voc/tech training	Attend college for diploma	Attend univ. for degree
Control group	No plans	6	1	1	3	11
	Voc/tech training	2	1	1	1	5
	Attend college for diploma	1	2	2	10	15
	Attend univ. for degree	8	4	7	16	35
	Total	17	8	11	30	66
Treatment group	No plans	0	1	0	3	4
	Voc/tech training	1	3	1	3	8
	Attend college for diploma	1	0	0	0	1
	Attend univ. for degree	8	1	4	16	29
	Total	10	5	5	22	42

Note: Bold indicates negative changes in educational plans at Wave 2; italic indicates positive changes.

Level of confidence in achieving educational plans

Results regarding changes in levels of confidence among study participants achieving their educational plans – between Wave 1 and Wave 2 – indicate the following: 1) there was a 31 percent change in the number of girls in the treatment group reporting a more optimistic and higher level of confidence in achieving their educational plans. The control group, in comparison, had only a 10 percent change in the number of girls reporting high level of confidence (or optimism) in achieving their educational plans. The 21 percent point difference between the two groups at Wave 2 (in favor of the treatment condition) is worth noting; 2) there was only a 1 percent change in the number of girls in the treatment group reporting less optimism in achieving their educational plans by Wave 2. On the other hand, the percentage change for girls in the control group reporting a negative change–toward being less confident and pessimistic in achieving their educational plans was 35 percent. This implies that whereas by Wave 2 only 1 percent of the girls in the treatment group had lost confidence in achieving their educational plans reported at Wave 1, a significant percentage of their peers (35%) in the control group had lost confidence in achieving their educational plans during the same period. The 34 percent point difference in favor of the treatment group is noteworthy (see Table 3).

Table 3.

Confidence in educational plans for girls.
	Confidence in plans W1	Confidence inl plans at W2			Total	Total
		Not very sure	Fairly sure	Very sure
	Not very sure	0	1	4	5	10
Treatment	Fairly sure	0	2	17	19	38
Group	Very sure	0	1	47	48	96
	Total	0	4	68	72	144
	Not very sure	2	0	2	4	8
Control	Fairly sure	0	2	3	5	10
Group	Very sure	1	17	25	43	86
	Total	3	19	30	52	104

Confidence in educational plans for boys.
	Confidence in plans W1	Confidence inl plans at W2			Total
		Not very sure	Fairly sure	Very sure
	Not very sure	5	14	10	29
Control	Fairly sure	0	0	4	4
Group	Very sure	0	1	9	10
	Total	5	15	23	43
	Not very sure	0	3	6	9
Treatment	Fairly sure	0	2	9	11
Group	Very sure	2	3	14	19
	Total	2	8	29	39

Note: Bold indicates negative changes in educational plan at Wave 2; italic indicates positive changes.

Bivariate analysis

Independent and paired-sample t-tests were conducted on the two educational outcome measures, educational plans and level of confidence in achieving educational plans, to determine if there were any differences between groups and/or between waves. For the educational plans variable, independent t-test concluded no significant differences in Wave 1 between the treatment and control groups. However, results between the treatment and control groups in Wave 2 were significant (t(154) = 2.94, p < .01), indicating that the treatment group’s positive response to their educational plans was significantly higher than the control groups. However, although the mean change score for the treatment group (M = .40) was higher than the control group (M = .04) for the same variable, it was not significant (t(152) = 1.26, p > .05) (see Table 4).

Table 4.

Descriptive and bivariate results of educational outcomes for girls.
	Total sample	Treatment group	Control group	t-test
Educational plan
Wave 1	2.99	3.16	2.81	1.4
	(1.54)	(1.36)	(1.70)
Wave 2	3.19	3.51	2.85	2.94**
	(1.42)	(1.17)	(1.58)
Difference in plan	0.22	0.40	0.04	1.26
	(1.77)	(1.74)	(1.79)
Confidence in education plan
Wave 1	2.65	2.56	2.78	–2.09*
	(.61)	(.64)	(.56)
Wave 2	2.75	2.92	2.55	4.78**
	(.50)	(.32)	(.59)
Difference in confidence	0.1	0.35	−.23	4.70**
	(73)	(.60)	(.76)

Descriptive and bivariate results of educational outcomes for boys.
	Total sample	Treatment group	Control group	t-test or χ2
Educational plan
Wave 1	3.24	3.42	3.1	1.58
	(1.12)	(1.05)	(1.17)
Wave 2	2.97	3.15	2.83	1.38
	(1.25)	(1.22)	(1.27)
Difference in plan	–.27	–.30	–.26	–.16
	(1.52)	(1.53)	(1.52)
Confidence in education plan
Wave 1	1.87	2.24	1.54	4.24**
	(.91)	(.85)	(.85)
Wave 2	2.49	2.67	2.33	2.55*
	(.69)	(57)	(.75)
Difference in confidence	0.66	0.44	0.86	−2.10*
	(.93)	(1.02)	(80)

^*p ≤ .05;

^**p ≤ .01.

Examining the level of confidence scores with independent and paired-sampled t-tests, results show several significant changes. In Wave 1, confidence scores were significantly higher for the control group compared to the treatment group (t(133) = −2.09, p < .05).

However, in Wave 2, the results changed. The treatment group’s scores were significantly higher than the control group’s scores (t(138) = 4.78, p < .01). Finally, the computed change score for the level of confidence variable was significant between the treatment and control groups (t(122) = 4.70, p < .00) (see Table 4).

Discussion

Results from this study, show a difference in outcomes between the treatment and control groups. The educational outcomes reveal that although there were small changes among the two groups in educational plans between the waves, the confidence levels for the girls in the treatment group increased significantly between Wave 1 and Wave 2 while the control group’s levels decreased, suggesting that the girls in the treatment group felt more secure about their current education and were able to project that confidence in their future aspirations. Given more time in the program, the girls may feel stronger about their educational plans as well.

Saving outcomes showed that girls were saving at approximately 53 percent of their match eligibility. Knowing that financial barriers are a major issue, adding other incentives for girls to increase their savings and stay in school, such as assisting teachers or tutoring other students as income generating activities, might be options. Lessening the financial burden for families to send girls to school may help prioritize its importance within the family and community. These findings suggest that an asset-based family intervention program might be one tool that may help increase opportunities for girls to stay in school and continue their education.

Empowering girls through education is a powerful way to help them change their views about themselves and their future. The family is essential in these types of programs, because when economic empowerment interventions target the family, household-level financial barriers decrease, which can help facilitate girls’ regular school attendance and pursuit of post-primary education. In other words, programs and policies will need to invest financial resources at the family/household level – similar to the programs being implemented by Suubi Project – if we are to reach the girl child and address the existing gender gaps and inequalities in education and resource ownership. Moreover, these interventions need to start early, at least at the primary school level before the girl child is disengaged from school or taken out of school because of limited family resources.

Conclusion

The establishment of UPE in Uganda made considerable headway in trying to narrow the gender inequality gap in education. However, to ensure that the initiative is successful, girls must be facilitated in advancing their education to the secondary school level and beyond, without the limitations of their own and their family’s inability to fund their education. The government has acknowledged these needs by instituting the UPPET secondary school pilot programs, but unfortunately the programs, thus far, reach very few children. The results from the first two waves of the Suubi Project indicate that CDAs have the potential to increase financial and income-generating assets among girls and their caregiving families – which can mitigate the effect of gender inequality, particularly with regard to education. Girls in the treatment condition – who received CDAs – developed a future orientation and experienced improved levels of confidence compared to their counterparts in the control condition.

Findings from this study provide support for the implementation of broader CDA programs in Uganda and other developing regions, in which free and universal access to secondary education are yet to be realized. It is important to emphasize, however, that no single program will solve the problem, but if it is approached through a multi-dimensional development plan that helps offset the cost of comprehensive education and increases educational opportunities for girls and other economically disadvantaged populations, an asset-based family intervention program that includes CDAs could add a strong component to an overall plan.

The Suubi Project provides important data that supports policies and programs that have the potential to increase current and future opportunities for girls as well as other vulnerable populations, particularly in developing countries and low-resource settings. Results contribute to the growing body of knowledge on asset theory, suggesting that the accumulation of productive assets for girls and other vulnerable populations has several positive impacts.

More research is needed to determine the best fit of program features with specific populations, but the Suubi Project has laid the foundation and opened the doors to answer and ask more questions about asset-building interventions and to move the agenda forward.

Biographies

Jami Curley, PhD, MSW, is an Associate Professor at the School of Social Work, Saint Louis University, USA.

Fred M Ssewamala, PhD, is an Associate Professor at the School of Social Service Administration, University of Chicago, USA.

Proscovia Nabunya, MSW, is a doctoral student at the University of Chicago School of Social Service Administration, USA.

Vilma Ilic is a Program Manager at Columbia University School of Social Work, USA.

Han Chang Keun, PhD, is an Assistant Professor at the Department of Social Welfare, Sungkyunkwan University, Republic of Korea.