Study design
This study analyzes baseline data from Discover Learning collected in August and October 2020 from non-orphans in the peri urban Temeke District and data collected from orphans in three municipals districts in Dar es Salaam, Tanzania in April and May 2020. The study protocol for Discover Learning and these baseline (pre-intervention) data have been described in detail elsewhere, including recruitment, eligibility criteria, data collection procedures and human research protections (19). In 2020, data were collected from adolescent orphans ages 10 to 15. Participants for this study were recruited in collaboration with our local Tanzania partner, Health for a Prosperous Nation (H-PON) in Tanzania and youth-serving orphanages. The city of Dar es Salaam has five districts: Kinondoni, Ubungo, Ilala, Temeke and Kigamboni. H-PON identified youth orphanages in three peri-urban districts: Ubungo, Ilala and Temeke. Within the city, Ubungo is in the northwest, Ilala in the center, and Temeke in the southeast. Orphanages were introduced to the study and objectives, including consent/assent procedures. Orphanages that agreed to participate were included in the study and research staff obtained permission from the Ministry of Health and Social Welfare responsible for the orphanage to obtain approval to complete research activities. All adolescents ages 10 to 15 agreed to participate and provided assent were included in the study. Data was initially collected to support several objectives including an understanding and measurement of Utu. In this article, we examine the culturally grounded construct of Utu and associated relationships with mental health and well-being measures within these two samples. Comparing orphan and non-orphan groups broadens the understanding of Utu within different populations.
Survey measures
Utu. The development of the Utu scale measure was completed in close collaboration with local research partners including Health for a Prosperous Nation, a Tanzanian NGO that administered surveys to adolescents and Ubongo Kids, a Tanzania-based organization that develops engaging and locally relevant digital content for children in Africa. Ubongo Kids’ edutainment content is disseminated via TV and radio episodes to over 6.4 million East African households weekly and aired on national television via the Tanzanian Broadcasting Corporation (TBC) every Saturday. Ubongo Kids’ edutainment (a combination of educational and entertainment content) video content target areas of adaptive social emotional mindsets and skills designed to be culturally relevant to children and adolescents in Tanzania.
First, to generate a culturally relevant pool of items for the Utu scale, the team conducted formative, qualitative research that explored how youth, adults and community members defined the concept of Utu. Ubongo Kids in partnership with researchers at University of California Berkeley collected data in May 2019. Qualitative research included 42 in-depth interviews and 4 focus groups, resulting in a total qualitative sample of 92 participants. Two focus groups were conducted with children and adolescents, one with female adults and one with male adults. Adult participants included church leaders, teachers, parents, and older siblings.
The formative research explored essential dimensions of Utu described by participants. These descriptions were then compared to those found in in-depth literature reviews. Researchers found the results of the formative research aligned with the literature reviews in describing concepts of Utu. Results indicated five core dimensions comprising the construct of Utu in Tanzania. These constructs included Resource Sharing, Respect and Dignity, Solidarity, Collectivism and Compassion. To measure these constructs, researchers adapted items from previously validated scales of Ubuntu used in South Africa (20, 21). They also added the compassion dimension using select items from the validated compassionate engagement scale (22, 23). Items were slightly adapted to ensure cultural relevancy and appropriateness for adolescent populations. Response categories were measured using a 4-point Likert scale ranging from one (strongly disagree) to four (strongly agree).
Adverse Life Experiences. This study used the Childhood Trauma Questionnaire to measure adverse life experiences or risk in this population. This instrument is a self-report measure that assesses emotional abuse, emotional neglect, and physical neglect. Responses were recorded on a 5-point Likert-type scale, with 1="never true" and 5="very often true”. This scale demonstrates a Cronbach's alpha of 0.95 (24).
Coping. The KidCope was used to assess coping strategies in response to concentrated stress. The KidCope is a measure used in many different global contexts and includes 15-questions to measure ten cognitive and behavioral coping strategies in children and adolescents (Spirito et al., 1988). The original checklist of 15 questions was adapted for this study to include an additional 16th question, “I prayed to feel better” based on qualitative research conducted in Tanzania and the Democratic Republic of Congo (25). The four-factor structure of the KidCope was validated for these data in a previous study and includes, distraction, resignation, problem-focused and social support coping strategies (Cherewick, 2022, under review). Administration of the KidCope first asks adolescents to think of something stressful they have experienced or ongoing stressors and to rate how often they utilized each of the 16 items on a Likert scale (“not at all” =0 to “all the time” =1).
Resilience. The Child and Youth Resilience Measure (CYRM) is a 28-item self-report measure of resilience among young people that has been widely used in a diversity of contexts and translated to more than 20 languages (18). The 28 item CYRM included 11 items representing a contextual resilience subscale. Understandably, this subscale has shown inconsistencies in factor structure in different global contexts (26, 27). Findings have attributed variation in the CYRM factor structure because of differences in individualistic vs. collectivistic cultures (28, 29). The Child and Youth Resilience Measure Revised (CYRM-R) is a 17-item, 2-dimensional scale of intra/interpersonal resilience and caregiver resilience subscales that have been validated for use in diverse cultures and contexts (30).
Demographic Characteristics. Social and demographic characteristics of participants were collected including age, sex, report of general health, pubertal development status and orphan status.
Data Analysis
First, descriptive statistics were calculated for each item measure included in the Utu scale for orphans and non-orphans. Differences between the samples were compared. Second, confirmatory factor analysis (CFA) was executed to test our hypothesis that a five-factor model would best fit these data on the orphan and non-orphan samples, and the full analytic sample. The following conventional criteria used to evaluate goodness of fit of the CFA (31) included chi-square test (model vs. a baseline p<=0.05); the Comparative Fit Index (CFI) values >=0.95 (32); the Tucker Lewis Index (TLI) values >=0.95 (33); the Standardized Root Mean Square Residual (SRMR) values <=0.08 (34); the Root Mean Square Error of Approximation (RMSEA) values <=0.06 (35); Akaike’s Information Criterion (AIC) (36); and Bayesian Information Criterion (BIC)(37). To estimate the internal consistency of scales, researchers calculated Cronbach’s alpha, McDonald’s omega coefficient and item-rest correlations for the entire analytical sample.