Abstract
Purpose
This paper examines gender differences in attitudes towards sexual risk-taking behaviors of AIDS-orphaned youth participating in a randomized control trial testing an economic empowerment intervention in rural Uganda.
Methods
Adolescents (average age 13.7 years) who had lost one or both parents to AIDS from fifteen comparable schools were randomly assigned to either an experimental (n=135) or control condition (n=142). Adolescents in the experimental condition, in addition to usual care, also received support and incentives to save money toward secondary education.
Results
Findings indicate that although adolescent boys and girls within the experimental condition saved comparable amounts, the intervention appears to have benefited girls, in regards to the attitudes towards sexual risk-taking behavior, in a different way and to a lesser extent than boys.
Conclusions
Future research should investigate the possibility that adolescent girls might be able to develop equally large improvements in protective attitudes towards sexual risk-taking through additional components that address gendered social norms.
Keywords: Gender, sexual risk taking, microfinance, economic empowerment intervention, orphan, adolescent, sub-Saharan Africa
Introduction
Worldwide, gender and poverty are two of the most important influences on health [1]. This is particularly true in sub-Saharan Africa (SSA), the poorest region of the world, and the region most affected by HIV/AIDS, a highly gender-discriminatory epidemic [2-6]. This paper considers the role of gender in a health-promoting economic empowerment intervention study in Uganda. Specifically, the paper compares the behavioral health functioning of adolescent girls with boys who are AIDS-orphaned participating in a randomized clinical trial testing the effect of an economic intervention on adolescents’ health outcomes.
The need for the present study is rooted in three contexts. First, in SSA,women are disproportionately affected by HIV [1,2]. Women's vulnerability to infection is exacerbated by poverty, early marriage, and cultural attitudes that limit women's power in relationships, including the power to negotiate safe sex [2-6]. Among young adults ages 15-24, females are particularly affected, representing 74 percent of new HIV infections in SSA [6]. In Uganda, where the first HIV case in East Africa was discovered in the 1980s [7], females aged 15-19 are four times more likely to be HIV-infected than their male peers, with higher rates in more heavily affected areas (most recent data is from the period 2004-2005, [8]).
Second, HIV has also contributed to vast increases in the numbers of AIDS-orphaned children and adolescents defined by the UN as pers Cons of age 0-18 years who have lost at least one parent to HIV/AIDS. By this definition, Uganda is home to over 1 million AIDS-orphaned children, and in the heavily-impacted southern district of Rakai, approximately two in five children have been orphaned due to AIDS [9]. Studies have repeatedly linked orphan status – particularly for females – to heightened HIV-risk [10].
Third, the need for the present inquiry is reflected in the increasing recognition of the ways in which poverty, health disparities, and gender discrimination reinforce one another [11-13]. The Lancet's series on the science of HIV prevention, for example, has recently emphasized structural determinants of health, calling for the development and testing of “combined” interventions to address multiple dimensions of preventive health [11]. Indeed, studies have begun integrating social and economic empowerment, such as the IMAGE study in South Africa which combines microfinance with women's empowerment [14]. In this vein, the current paper considers the relation between gender and success in a health-promoting economic empowerment intervention for AIDS-orphaned children in Uganda. Specifically, this paper is guided by the questions: Do female and male AIDS-orphaned youth benefit equally from an economic empowerment intervention in rural Uganda? If not, what are the policy and program implications of these findings?
Background and Theory: The SUUBI Research Program
Data examined here were obtained from the SUUBI Research Program (hereafter, SUUBI, which is a Luganda word for “hope”). SUUBI (2005-2008) was funded by the National Institute of Mental Health. The study took place in Rakai District of Uganda the site of the first HIV infection detected in East Africa during the 1980s [7]. The study received IRB approval by Columbia University and Uganda National Council of Science and Technology. The overall aim of SUUBI was to develop and test the feasibility of an intervention program for improving the health outcomes and life chances of AIDS-orphaned adolescents through the use of microfinance—in the form of Child Savings Accounts (CSAs).
The SUUBI Research Program was guided by asset theory [15, 16]. Asset theory posits that when people have assets, such as education, savings accounts, or productive assets, such as small businesses, it can change not only their economic lives, but also their behavior, attitude, and hope for the future. In addition, asset theory suggests that asset accumulation creates “asset effects,” in which more assets in the present influence expectations for more resources in the future, and also engender greater social trust (e.g., feelings of safety and security), and lengthened time horizons (e.g. greater future planning), [17]. Overall, asset theory posits that economic opportunities influence the formation of attitudes and beliefs, as well as intentions and behaviors [18]. Asset theory is consistent with behavioral and psychosocial theories that endorse the idea that opportunity influences the formation of attitudes, intentions, beliefs and behaviors [19-21].
Accordingly, SUUBI provided poor AIDS-orphaned adolescents with asset-ownership opportunities through the use of CSAs. Modeled on U.S. Individual Development Accounts, a CSA is a microfinance product that facilitates the development of savings and their attendant “asset effects” [17]. CSA programs, including SUUBI, involve youth in basic financial education, introduce them to formal financial institutions, and incentivize them to save small amounts by offering a subsidized match-rate. In SUUBI, the match rate was 2:1. The matched savings had to be used either for the youth's post-primary education, or for starting a family small-business. All participants choose post-primary education as their saving goal.
Methods
Participant selection and assignment
SUUBI included 286 AIDS-orphaned youth (average age 13.7 years) in primary school—just prior to the transition to secondary school. AIDS orphan status was defined as having lost either one or both parents to AIDS, according to self-report. Due to attrition, the final sample included 277 participants. Participants were selected from fifteen primary schools with similar socio-economic characteristics, including overall performance on the standardized national primary leaving examinations (PLE) used as a qualification for admission to secondary school. In addition, the selected schools attracted students from similar socio-economic backgrounds.
Each of the fifteen schools was randomly assigned to either experimental or control condition at the school level, so that all students within one school received the same condition assignment. Due to geographic distance between schools, participants assigned to different conditions did not know each other.
Participants assigned to the control condition received the usual care for orphaned children, consisting of counseling and educational related supplies (including text-books). And by virtue of being in school, all participants received health education (including AIDS-focused education) provided through a nationwide school-based curriculum. Participants assigned to the experimental condition (SUUBI-program) received the usual care plus an economic empowerment intervention consisting of three components: 1) twelve 1-2 hour workshops over a 10-month period focused on assets-building and financial planning. The workshops introduced the participants to asset-building strategies, including saving, education and small business development; 2) a monthly mentorship program for adolescents with peer mentors on future planning and life options; and 3) a matched CSA, dedicated to paying for post-primary schooling. (See details below).
Child Savings Account (CSA) Details
SUUBI worked with Centenary Rural Development Bank, one of the foremost financial institutions working with low-income families in Uganda. The minimum base deposit for each participant was provided by SUUBI to facilitate access to a no-fee account. The adolescent's family members and relatives were then allowed to make deposits directly into the CSA, and were encouraged to do so by the participants. Although matching funds were kept in a separate account—distinct from participants’ CSAs—participants received monthly savings account statements that indicated both the participant's own accumulated savings, as well as the current match amount [22]. These statements were intended to act as a morale booster, and to facilitate trust-building. When participants were ready to use their savings to pay for post-primary education, they would produce their one-third of the total due, and the SUUBI project staff would pay the remaining two-thirds in matching funds directly to the school. Thus, although they were kept abreast of their accumulated matching fund value, at no time did CSA group participants have direct access to matching funds. The match rate was set at 2:1—based on earlier studies on IDAs that found that participants who received a match rate of 2:1 saved at a significantly higher rate than participants who received a lower match rate [23-24].
Additionally, a match cap, or maximum amount of savings to be matched by SUUBI, was set at an equivalent of $10 (US dollars) per CSA group participant, per month (with a cap of $120 per participant for a twelve-month period). Thus, if a participant deposited $10 each month in their CSA for twelve months without taking it out of the account, they would accumulate $120. These funds would then be matched by SUUBI with an additional $240, paid directly to the secondary school. The total amount saved by the child, including matching funds, would therefore be $360, in this case. In rural Uganda $360 is sufficient to pay for at least 2 years of secondary education, including tuition and school uniform.
It is important to note that although there are studies focused on saving performance of poor families in SSA [for example, see 25-27], none of these studies were the actual accounts held in the names of the youth. Nor were the youth surveyed in these studies.
Measurement and Analysis
Data were obtained through a 90-minute individual assessment at baseline and at the 10-month follow-up interviews. The interviews were conducted by local research assistants blind to study assignment. We matched the interviewers with the interviewees by gender. Assessment items, adapted for Ugandan adolescents from previously tested scales in Uganda [9] and South Africa [28] were translated into the local language and checked for wording and meaning by Ugandan research staff. Specifically, adaptation of the instruments involved the Ugandan PI researching relevant measures used in similar studies throughout SSA, drawing items from these scales and revising to fit the local culture. The measures were reviewed by three community leaders in the study area, and pretested by the local project coordinator in Uganda.
This paper primarily focuses on two outcomes: (1) savings performance; and (2) attitudes toward sexual risk-taking behaviors.
Saving performance
We used Average Monthly Net Deposit (AMND) as the measure of saving performance [24, 29]. AMND is defined as net deposits per month of participation. This measure controls for the length of time that a participant had had to save. Higher AMND implies higher savings, with a theoretical range of zero to infinity.
Attitudes toward sexual risk-taking behaviors
It is important to note upfront that measuring sexual risk taking among adolescents presents a problem in most public health studies due to lack of reliable measurement instruments [30-32]. For instance, although actual sexual risk taking behavior is one of the outcomes we tried to capture in the SUUBI study, the number of study participants reporting sexual risk taking behavior was too small (N=7) to permit statistical analysis. These low reports of sexual activity among participants may be attributed to age (median age of participants was 13.15 years) and school enrollment status. In rural Uganda the median age at first intercourse is 16.5 years among women and 18 years for males [33]. Given this finding, and since shifts in attitude may be the next best option to capture sexual risk taking behaviors among adolescents [32, 34, 35], plus the fact that such measures have been found to be predictive of future sexual behavior [36], in this paper we focus on estimating the effects of the SUUBI intervention on adolescents’ attitudes toward sexual risk-taking behaviors.
Before we delve into the attitudes toward sexual risk-taking behavior question, however, we need to point out that since all school-going children are exposed to health education (including AIDS-focused education) through a national curriculum, one would expect no statistical difference in HIV-knowledge between the two study groups. Indeed our baseline HIV Knowledge measure consisting of 13-items about HIV infection and prevention mechanisms, with sample items such as: “HIV can be passed by an infected person even though that person is not feeling sick”, “Using a condom will lessen the chance of getting AIDS”, “If you are really healthy, then exercising daily can prevent getting HIV”, revealed no significant differences between the two group means. In addition, gender disaggregated analysis demonstrated that girls in the control group did not statistically differ in their HIV knowledge (mean score 9.4) from girls in the treatment group (mean score 9.8), and boys in the control group (mean score 11.0) had a similar HIV knowledge score as boys in the treatment group (mean score11.3).
In regards to attitudes toward sexual risk-taking behavior, students were asked to report their opinion on 6 items (e.g., “I believe it's OK for people my age to have sex with someone they've just met”, “I believe it's OK for people my age to have sex with someone they love”, “I believe it's OK to have sex without protection with someone you know”). For each item, a statement was read to the participant, who was then asked to rate the degree to which they agreed with or disagreed with the statement on a scale of 1-5, respectively. The total score on this particular measure ranged from 6 to 30. A higher score indicated a more favorable attitude toward sexual risk-taking behaviors, and a low score indicated more protective attitudes.
Data Analysis
Tram and Cole [37] have argued that the analysis of potential moderators is crucial for correct interpretation of intervention outcomes. Accordingly, we examine gender as a moderator of treatment effects of SUUBI intervention. Specifically, we test for differences in treatment outcomes based on child gender that should be considered when examining which adolescents within the experimental group benefited the most from their participation in SUUBI. We construct a 2×2×2 ANOVA [Gender (boy vs. girl) by Treatment group (control vs. experimental) by Time (wave 1 vs. wave 2)] model, with time of measurement being treated as a repeated measures factor. Due to randomization at school level, participants observations from the same school could be correlated, which would violate the assumption of independence of observations [38]. Therefore, to control for school effects, we treat the school ID variable as a random effect. We fit the three-way interactions model to the attitudes toward sexual risk-taking outcome variable using the mixed model procedure in SAS 9.2. The model was diagnosed to meet the assumptions of normality of residuals and a constant variance across levels of the predicted values. We control for socio-demographic characteristics that demonstrated significant relationships to attitudes toward sexual risk-taking behavior, namely age, gender, and employment status of the primary caregiver. The highest order statistically significant interaction was further decomposed using tests of effect sizes. We examined marginal mean values to determine the direction of the relationship.
Results
Socio-demographic characteristics
As indicated in Table 1, with the exception of employment status of caregivers, there were no observable significant gender differences within the control group. Within the experimental group, boys and girls differed significantly only on the type of female caregiver.
TABLE 1.
Gender differences in socio-demographic characteristics of treatment and control conditions
| Control Mean (SD) or % |
t/χ2 |
Treatment Mean (SD) or % |
t/χ2 |
|||
|---|---|---|---|---|---|---|
| Key Characteristics |
Boys (n=70) |
Girls (n=78) |
|
Boys (n=55) |
Girls (n=83) |
|
| Mean child age, years |
13.77 (1.6) |
13.50 (1.36) |
−.91 |
14.07 (1.52) |
13.70 (1.2) |
−1.24 |
| Average # of people in household |
6 (SD=2) |
6 (SD=2) |
.21 |
7 (SD=3) |
6 (SD=3) |
−1.96 |
| Average # of children in household |
3 (2) |
3 (1) |
−.21 |
4 (2) |
3 (2) |
−.85 |
| Father not living |
55 (79%) |
68 (88%) |
2.08 |
43 (78%) |
64 (78%) |
.05 |
| Mother not living |
42 (62%) |
42 (54%) |
.67 |
37 (67%) |
45 (54%) |
2.03 |
| Both parents not living |
45 (40%) |
32 (41%) |
.09 |
25 (46%) |
26 (31%) |
1.9 |
| Caregiver self-employed |
55 (80%) |
45 (58%) |
7.68** |
29 (53%) |
50 (60%) |
.63 |
| Female caregiver | 7.51 | 13.5* | ||||
| Biological mother | 25 (36%) | 29 (37%) | 13 (24%) | 45 (54%) | ||
| Stepmother | 2 (3%) | 3 (4%) | 1 (2%) | 3 (4%) | ||
| Grandmother | 25 (36%) | 34 (44%) | 23 (42%) | 21 (25%) | ||
| Aunt | 10 (14%) | 10 (13%) | 12 (22%) | 11 (13%) | ||
| Sister | 2 (3%) | 2 (3%) | 2 (4%) | 1 (1%) | ||
| No woman |
6 (9%) |
0 (0%) |
|
4 (7%) |
2 (2%) |
|
| Male caregiver | 3.85 | 9.13 | ||||
| Father | 16 (23%) | 19 (24%) | 16 (29%) | 30 (36%) | ||
| Stepfather | 1 (1%) | 1 (1.%) | 0 (0%) | 0 (0%) | ||
| Grandfather | 7 (10%) | 11 (14%) | 8 (15%) | 13 (16%) | ||
| Uncle | 3 (4%) | 8 (10%) | 13 (24%) | 7 (8%) | ||
| Brother | 3 (4%) | 2 (3%) | 3 (6%) | 2 (2%) | ||
| Other | 1 (1%) | 0 (0%) | 1 (2%) | 0 (0%) | ||
| No Male Present | 39 (56%) | 37 (47%) | 14 (26%) | 31 (37%) | ||
p < .05
p < .01
*** p < .001
Gender differences in Savings Outcomes
Average monthly net deposit (AMND) for boys was an equivalent of $7.26 (SD=$2.56), while girls saved $6.72 (SD=$1.92). The median AMND for girls was $5.01 vs. $5.70 for boys. Three girls (out of 83) and nine boys (out of 55) did not save. There were no statistically significant differences by gender in AMND. Girls and boys saved at comparable rates.
Attitudes toward engaging in sexual risk-taking behaviors
Overall, the intervention demonstrated a beneficial effect on attitudes toward sexual risk-taking behaviors [Time by Treatment interaction is statistically significant F(1, 266)=40.36, p<.05, partial eta-squared =.13]. Furthermore, results presented in Figure 1 indicate that gender moderates the effect of the SUUBI intervention on attitudes toward sexual risk-taking behaviors [Time × Treatment × Gender interaction is statistically significant, F(1, 266)=16.1, p<.001, partial eta-squared=.06]. Both girls and boys in the control condition demonstrated an increased “approval” of risky sexual behaviors. Specifically, the score for girls in the control condition increased from 7.73 (at pre-test) to 9.19 (at 10-month follow-up) and increased for boys in the control group from 8.9 (at pre-test) to 13.43 (at 10-month follow-up). Within the experimental group different trends were observed: Whereas the score for girls in the experimental group was unchanged from 8.28 (at pre-test) vs. 8.25 (at 10-month follow-up), the boys in the experimental group reported a significant decrease in approval of risky sexual behaviors from 12.44 (at pre-test) to 10.29 (at 10-month follow-up). This finding suggests that SUUBI intervention may have increased boys’ protective attitudes towards sexual risk-taking behaviors but probably did not benefit girls in an equivalent way. Given the fact that girls in the control group actually became more accepting of risky sexual behaviors at the 10-month follow-up, yet girls in SUUBI intervention group maintained their original degree of protective attitudes, the intervention appears to have benefited girls in SUUBI intervention, but probably in a different way and to a lesser extent than for their male counterparts.
Figure 1.
Gender differences in attitudes toward sexual risk-taking behaviors
Discussion and Implications
Findings partially support the hypothesis that gender has a moderating effect on SUUBI study outcomes. SUUBI, as a combined intervention, worked better for boys than girls in improving attitudes towards sexual risk-taking behaviors. That said, both boys and girls in the SUUBI intervention group benefited relative to their same-sex counterparts assigned to the control group, although the intervention—overall—seems to have had a modest or smaller effect for girls.
Gender equality in savings outcomes
Adolescent girls and boys saved comparable amounts. This probably means, among other things, that caregivers who contributed most of the funds in CSAs, deemed it a good idea to support female children in trying to save money and attain further education. This may indicate that caregivers perceive both saving money and pursuit of secondary education to be as desirable for girls as boys. This finding is consistent with the fact that, in Uganda, gender inequality in education is relatively low compared to several African countries. Moreover, the government encourages all children, regardless of gender, to attend primary school through universal primary education policy.
Gender differences in attitudes toward sexual risk-taking behaviors
Given the premises of asset theory, one would predict that the economic opportunities offered by SUUBI would increase orphaned adolescents’ capacity to envision the future with optimism, lead to increases in educational aspirations and protective attitudes. Results from this study show that although girls saved comparable amounts to boys, they did not experience equally protective improvements in attitudes toward sexual risk-taking behaviors: Girls in the control condition became less protective and more lenient towards sexual risk taking attitudes during the study period than their counterparts in the experimental condition. Thus, although gender disparity is evident in the study results, the program is nevertheless associated with a modest impact for girls’ attitudes towards risk-taking behavior when one compares them to their same-sex counterparts in the control condition.
With this in mind, and considering the open forms of gender discrimination in local cultures that put girls at higher HIV-risk than their male peers, particularly early marriage to older men in the face of economic hardship [39], results from this study indicate that future research should consider testing the effect of a gender-specific empowerment component.
Finally, given the fact that attitudes towards sexual risk-taking behavior improved considerably for male participants in the experimental group, future research may consider testing strategies that work in groups to build consensus across genders about what behavior is acceptable, in hopes that male participants who develop greater protective attitudes might “show” their female counterparts that such protective behavior is acceptable to – at least some – of their male peers.
Conclusion
The findings of this study provide essential knowledge regarding the benefits of an economic empowerment approach in the care and support for orphaned adolescents within a family setting. In this particular case, adolescent girls saved as much as boys, and maintained slightly more protective attitudes towards sexual risk taking than their female peers in the control condition. Interestingly, the largest disparity in outcomes was manifest between boys and girls in the experimental condition. In spite of comparable levels of savings, boys developed much greater improvements in protective health attitudes than girls over the course of the intervention. Future research should investigate the possibility that adolescent girls might be able to develop equally large improvements in protective attitudes towards sexual risk taking through additional components that address gendered social norms, particularly with regard to sexual decision making. Overall, results indicate that a combined, health-promoting economic empowerment intervention has the potential to yield important benefits for both adolescent boys and girls.
Acknowledgments
The SUUBI-project was funded by a grant from the National Institute of Health (R21 MH076475-01). We are grateful to the SUUBI-Uganda Research Staff and volunteers for monitoring the study implementation process, especially, Ms. Proscovia Nabunya, and Reverend Fr. Kato Bakulu. We thank Professors Jane Waldfogel, Nabila El-Bassel, Michael Sherraden, and Torsten Neilands for helpful comments on the study intervention design, implementation and/or data collection methods. Our thanks also go to all the children and their caregiving families who agreed to participate in the SUUBI study.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Fred Ssewamala, Columbia University.
Leyla Ismayilova, Columbia University
Mary McKay, Mount Sinai School of Medicine
Elizabeth Sperber, Columbia University
William Bannon, Jr., Mount Sinai School of Medicine
Stacey Alicea, Mount Sinai School of Medicine
References
- 1.Chen LC. Paper presented to Swedish International Development Agency. Stockholm: 2002. Engendering International Health: The Challenge of Equity. [Google Scholar]
- 2.Foster G, Williamson J. A review of current literature on the impact of HIV/AIDS on children in sub-Saharan Africa. AIDS. 2000;14:S275–284. [PubMed] [Google Scholar]
- 3.Gregson S, Garnett GP. Contrasting gender differentials in HIV-prevalence and associated mortality increase in eastern and southern Africa: artifact Rof data or natural course of epidemics. AIDS. 2000;14:S85–99. [PubMed] [Google Scholar]
- 4.Gregson S, Nyamukapa C, Garnett G, et al. HIV infection and reproductive health in teenage women orphaned and made vulnerable by AIDS in Zimbabwe. AIDS Care. 2005;17:785–794. doi: 10.1080/09540120500258029. [DOI] [PubMed] [Google Scholar]
- 5.Glynn JR, Carael M, Auvert B, et al. Why do young women have a much higher prevalence of HIV than young men? A study in Kisumu, Kenya and Ndola, Zambia. AIDS. 2001;15:S51–60. doi: 10.1097/00002030-200108004-00006. [DOI] [PubMed] [Google Scholar]
- 6.Hargreaves J, Boler T. Girl power: the impact of girls’ education on HIV and sexual behaviour. ActionAid International; Johannesburg, South Africa: 2006. [Google Scholar]
- 7.Serwadda D, Mugerwa RD, Sewankambo NK, Lwegaba A, Carswell JW, Kirya GB, et al. Slim disease: a new disease in Uganda and its association with HTLV-III infection. Lancet. 1985;2(8460):849–52. doi: 10.1016/S0140-6736(85)90122-9. [DOI] [PubMed] [Google Scholar]
- 8.Ugandan AIDS Commission. Uganda HIV/AIDS Sero-Behavioural Survey 2004-2005. Uganda AIDS Commission; Uganda: 2008. [Google Scholar]
- 9.Ssewamala FM, Alicea S, Bannon W, Ismayilova L. A Novel Economic Intervention to Reduce HIV Risks among AIDS Orphans in Rural Uganda. Journal of Adolescent Health. 2008;42:102–104. doi: 10.1016/j.jadohealth.2007.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Thurman T, Brown L, Richter L, Maharaj P, Magnani R. Sexual risk behavior among South African adolescents: is orphan status a factor? AIDS Behav. 2006;10:627–635. doi: 10.1007/s10461-006-9104-8. [DOI] [PubMed] [Google Scholar]
- 11.Marmot M. Social determinants of health inequalities. Lancet. 2005;365:1099–1104. doi: 10.1016/S0140-6736(05)71146-6. [DOI] [PubMed] [Google Scholar]
- 12.Evans G. The environment of childhood poverty. Am Psych. 2004;59:77–92. doi: 10.1037/0003-066X.59.2.77. [DOI] [PubMed] [Google Scholar]
- 13.Horton R, Das P. Putting prevention at the forefront of HIV/AIDS. Lancet. 2008;372:421–422. doi: 10.1016/S0140-6736(08)60882-X. [DOI] [PubMed] [Google Scholar]
- 14.Kim J, Watts C, Hargreaves J, et al. Understanding the impact of a microfinance-based intervention on women's empowerment and the reduction of intimate partner violence in South Africa. American Journal of Public Health. 2007;97:1794–1802. doi: 10.2105/AJPH.2006.095521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sherraden M. Stake-holding: Notes in a theory of welfare based on assets. Social Service Review. 1990;64:580–601. [Google Scholar]
- 16.Sherraden M. Assets and the Poor: A new American welfare policy. M.E. Sharpe; New York: 1991. [Google Scholar]
- 17.Schreiner M, Sherraden M. Can the Poor Save? Saving and Asset Building in Individual Development Accounts. Transaction Publishers; New Brunswick, USA: 2007. [Google Scholar]
- 18.Sherraden M. School Dropouts in Perspective. Educational Forum. 1986;51:15–31. [Google Scholar]
- 19.Ajzen I, Fischbein M. Understanding Attitudes and Predicting Social Behavior. Prentice Hall; Englewood Cliffs, NJ: 1980. [Google Scholar]
- 20.Jemmott JB, Jemmott LS, Hacker CI. Predicting intentions to use condoms among African-American adolescents: The theory of planned behavior as a model of HIV-associated behavior. Ethnicity and Disease. 1992;2:371–380. [PubMed] [Google Scholar]
- 21.Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Prentice Hall; Englewood Cliffs, NJ: 1986. [Google Scholar]
- 22.Ssewamala FM, Ismayilova L. Integrating Children Savings Accounts in the Care and Support of Orphaned Adolescents in Rural Uganda. Social Service Review. doi: 10.1086/605941. (in press) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ssewamala FM, Sherraden M. Integrating Savings into Microenterprise Programs for the Poor: Do Institutions Matter? Social Service Review. 2004;78:404–428. [Google Scholar]
- 24.Schreiner M, Sherraden M, Clancy M, et al. Savings and Asset Accumulation in Individual Development Accounts: Down payments on the American Dream Policy Demonstration, A National Demonstration of Individual Development Accounts. Research report. Washington University, Center for Social Development; St. Louis, MO: 2001. [Google Scholar]
- 25.Rutherford S. The poor and their money. Oxford University Press; New Delhi: 2000. [Google Scholar]
- 26.Ssewamala FM. Expanding Women's Opportunities: the Potential of Heifer Projects in sub-Saharan Africa. Development in Practice. 2004;14:550–559. [Google Scholar]
- 27.Dupas P, Robinson J. Savings Constraints and Microenterprise Development: Evidence from a Field Experiment in Kenya. 2009. NBER Working Paper.
- 28.Bhana A, Petersen I, Mason A, et al. Children and youth at risk: Adaptation and pilot study of the CHAMP (Amaqhawe) programme in South Africa. African Journal of AIDS Research. 2004;3:33–41. doi: 10.2989/16085900409490316. [DOI] [PubMed] [Google Scholar]
- 29.Sherraden M, Johnson L, Clancy M, et al. Saving Patterns in IDA Programs. Research report. Washington University, Center for Social Development; St. Louis, MO: 2000. [Google Scholar]
- 30.MacPhail C, Campbell C. ‘I think condoms are good but, aai, I hate those things’: condom use among adolescents and young people in a southern African township. Social science and medicine. 2001;52:1613–1627. doi: 10.1016/s0277-9536(00)00272-0. [DOI] [PubMed] [Google Scholar]
- 31.Mensch BS, Hewett PC, Erulkar AS. The reporting of sensitive behavior by adolescents: A methodological experiment in Kenya. Demography. 2003;40:247–268. doi: 10.1353/dem.2003.0017. [DOI] [PubMed] [Google Scholar]
- 32.Mensch BS, Hewett PC, Gregory R, et al. Sexual Behavior and STI/HIV Status among Adolescents in Rural Malawi: An Evaluation of the Effect of Interview Mode on Reporting. Working Paper No. 8. Population Council; New York: 2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Uganda Bureau of Statistics (UBOS) and Macro International Inc. Uganda Demographic and Health Survey 2006. UBOS and Macro International Inc; Calverton, Maryland: 2007. [Google Scholar]
- 34.White KM, Terry DJ, Hogg MA. Safer Sex Behavior: The Role of Attitudes, Norms, and Control Factors. Journal of Applied Social Psychology. 1994;24:2164–2192. [Google Scholar]
- 35.Raj A. Identification of Social Cognitive Variables as Predictors of Safer Sex Behavior and Intent in Heterosexual College Students. Journal of Sex and Marital Therapy. 1996;22:247–258. doi: 10.1080/00926239608404403. [DOI] [PubMed] [Google Scholar]
- 36.Gillmore MR, Archibald ME, Morrison DM, et al. Teen Sexual Behavior: Applicability of the Theory of Reasoned Action. Journal of Marriage and Family. 2002;64:885–897. [Google Scholar]
- 37.Tram JM, Cole DA. Self-perceived confidence and the relation between life events and depressive symptoms in adolescence: mediator or moderator? Journal of Abnormal Psychology. 2000;109:753–760. doi: 10.1037//0021-843x.109.4.753. [DOI] [PubMed] [Google Scholar]
- 38.Raudenbush S, Bryk AS. A Hierarchical Model for Studying School Effects. Sociology of Education. 1986;59:1–17. [Google Scholar]
- 39.Unicef . Annual Report 2008. United Nations Children's Fund; New York: 2008. [Google Scholar]