July 2008 1 USAID/DRC Integrated Youth Development Activity (IYDA) Cohort I Midline Outcome Evaluation Report DO Number and Name: Cooperative Agreement 720066078CA00003 Activity Start and End Dates: April 6, 2018 to April 5, 2021 Total Estimated Cost: $22,150,000 AOR/COR/Activity Manager Name & Office: Pascal Tshimanga, USAID/DRC/Education Implemented by: Education Development Center, Inc. Partners: FHI 360, Catholic Relief Services (CRS), Souktel Prepared by: Education Development Center, Inc. ©2020 by Education Development Center, Inc. All Rights Reserved. This content may not be reproduced without the written permission of Education Development Center, Inc (EDC). This report is made possible by the generous support of the American people through the United States Agency for International Development (USAID). The contents are the responsibility of Education Development Center, Inc. and do not necessarily reflect the views of USAID or the United States Government. EDUCATION DEVELOPMENT CENTER 43 Foundry Ave. Waltham, MA 02453 idd.edc.org USAID/DRC IYDA COHORT I MIDLINE EVALUATION i REPORT CONTENT Contents TABLE OF CONTENT ....................................................................................................................................i EXECUTIVE SUMMARY.................................................................................................................................1 Methodology .................................................................................................................................................1 Evaluation Findings.......................................................................................................................................2 BACKGROUND ..............................................................................................................................................4 Purpose ...............................................................................................................................................................6 Evaluation Questions......................................................................................................................................6 METHODOLOGY ...........................................................................................................................................7 Evaluation Design ...........................................................................................................................................7 Sampling......................................................................................................................................................7 Tools............................................................................................................................................................8 Data collection and Analysis................................................................................................................... 10 Limitations .................................................................................................................................................... 11 Limitations Specific to Cohort I .............................................................................................................. 12 DEMOGRAPHICS ......................................................................................................................................... 13 YOUTH PERCEPTIONS.............................................................................................................................. 16 COMMUNITY ENGAGEMENT.................................................................................................................. 16 PERCEPTION OF COMMUNITY................................................................................................................ 18 Overall Results...............................................................................Error! Bookmark not defined. Community Perception Indicator Results............................................................................................... 21 PERCEPTION OF PROSOCIAL PARENTAL/CAREGIVER INVOLVEMENT AND SUPPORT ................. 22 Overall Results......................................................................................................................................... 22 Prosocial Parental/Caregiver Support Indicator Results....................................................................... 24 PERCEPTION OF SELF-EFFICACY.............................................................................................................. 24 Overall Results...............................................................................Error! Bookmark not defined. Self-Efficacy Indicator Results................................................................................................................. 27 PERCEPTION OF VIOLENCE...................................................................................................................... 27 ENGAGEMENT IN GANG AND ARMED GROUP VIOLENCE.................................................. 28 NON-VIOLENCE AND TOLERANCE............................................................................................... 30 PERCEPTION OF POSITIVE IDENTITY & OPTIMISM ....................................................................... 33 Overall Results......................................................................................................................................... 33 USAID/DRC IYDA COHORT I MIDLINE EVALUATION ii Positive Identity & Optimism Indicator Results..................................................................................... 35 PERCEPTION OF SCHOOL SAFETY...................................................................................................... 36 OVERALL PERCEPTION OF SCHOOL SAFETY............................................................................. 36 Belonging, Diversity, & Inclusion ............................................................................................................ 38 Child Protective Practices........................................................................................................................ 39 Discipline and Fairness ........................................................................................................................... 40 School Safety in and on route to school ......................................................................................... 41 Student–Student Relationships .............................................................................................................. 42 Student–Teacher Relationships.............................................................................................................. 43 School Safety Indicator Conclusions..................................................................................................... 44 WORK READINESS SKILLS ....................................................................................................................... 46 BIG FIVE INVENTORY OVERALL RESULTS.............................................................................................. 46 BFI Disaggregated ................................................................................................................................... 47 SITUATIONAL JUDGEMENT TEST OVERALL RESULTS ........................................................................ 49 Work Readiness Indicator Results and Conclusions................................................................................. 50 CONCLUSIONS AND RECOMMENDATIONS .................................................................................. 52 Work readiness and confidence.......................................................Error! Bookmark not defined. Resilience ............................................................................................Error! Bookmark not defined. Baseline-Midline Indicator Outcomes...............................................Error! Bookmark not defined. Recommendations .............................................................................Error! Bookmark not defined. ANNEXES ....................................................................................................................................................... 56 Resilience Spearman's rank‐order correlation .......................................................................................... 56 Sample School Demographics.................................................................................................................... 58 Average Scores School Climate Survey............................................Error! Bookmark not defined. Data Collection Tools................................................................................................................................... 61 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 1 EXECUTIVE SUMMARY The USAID/DRC Integrated Youth Development Activity (IYDA) is funded by USAID and implemented by Education Development Center (EDC) and its consortium of partners Family Health International 360 (FHI 360), Catholic Relief Services (CRS), and Souktel. IYDA works in the North and South Kivu regions of the Democratic Republic of Congo, specifically in the cities of Goma and Bukavu and in the South Kivu territories of Kalehe, Kabare, and Walungu. IYDA provides vulnerable youth with learning pathways and inclusive economic opportunities as well as a network of services to support these pathways. Based on the positive youth development (PYD) framework and its four elements of Assets, Agency, Contribution and Enabling Environment, the Theory of Change for the USAID/DRC Integrated Youth Development Activity (IYDA) is as follows: When youth gain relevant skills and have opportunities for social and economic engagement, and when community stakeholders are supportive of youth efforts, then youth will be more resilient to the impact of conflict and violence. IYDA works with existing structures to implement these activities – accelerated education and vocational training centers, governmental officials, the private sector, and local youth-serving organizations. In particular, the project works with accelerated learning programs (Centres de rattrapages Scolaires – CRSs), vocational training centers (Centres d’apprentissage professionnel – CAPs), and community literacy programs run by community-based organizations (CBOs). Methodology This report presents the findings of the evaluation examining the outcomes of the first cohort of youth completing IYDA’s interventions. The evaluation design is quasi-experimental repeated measure (without a comparison group). The evaluation aims to understand IYDA’s contribution to improving youth assets, agency, civic engagement (contribution), and alliance building (enabling environment). These outcomes are measured through nine IYDA outcome indicators that inform the questions that IYDA’s evaluation, seek to answer:  Q1. Do youth demonstrate improved resilience to conflict and violence (measured by improve perception of non-violent means to address conflict drivers and conflict management, self￾efficacy and optimism about the future) as a result of IDYA’s activities?  Q2. Do youth demonstrate improved perception of support and engagement in their community?  Q3. Do youth demonstrate improved perception of school safety or learning spaces?  Q4. Do youth demonstrate improved work readiness, as a result of IYDA’s activities (with “improved” defined as an increase of one level on a multi-leveled life and leadership skills test)?  Q5. Do youth participants demonstrate improved livelihoods as a result of IDYA’s activities (with “improved livelihoods” defined as new or better employment as compared to baseline measures (with “better”, in turn, defined as higher-paying, more stable, of higher quality, or in better working conditions)? USAID/DRC IYDA COHORT I MIDLINE EVALUATION 2 Evaluation Findings This evaluation suggest that the IYDA project may positively influence participating youth in key ways. While claims to causality cannot be made, these evaluations do highlight clear areas in which the IYDA project seems to be supporting youth in increasing their resilience, skills, and perception of community and school safety. Within the context of conflict-affected eastern DRC, generally, IYDA youth show positive changes in perceptions of violence – away from supporting violence toward supporting peaceful means of conflict resolution – are promising. YOUTH’S PERCEPTIONS OF COMMUNITY IYDA youth demonstrate significant improvements in their perceptions of their communities with greater engagement in their community. Findings also demonstrate improved perceptions period further corroborate the youth's engagement in their communities and prosocial community support. Overall, 54.4 percent of youth improved at least one level on their perception of their community. RESILIENCE SKILLS IYDA youth rated themselves significantly and considerably higher at midline than baseline. While it isn’t possible to speak to these gains conclusively without a comparison group, these findings do highlight the potential for further integration of resilience skills into Life Skills training and youth-led activities. Overall 61.9 percent of youth improved at least one level on their perception of self-efficacy; 52.5 percent of youth improved their perception of parental/caregiver support; 57.4 percent of youth improved their one level or more on their propensity to engage in gangs and armed group violence; 37.3 percent of youth improved at least one level on their perception of engaging in violent groups; and, 56.1 percent of youth improved at least one level on their perception of optimism and positive identity. School Safety IYDA youth showed a positive change in their perception of school safety at the midline compared to the baseline. Still, the overall rating remained as fair with 53.8 percent of youth showing improved perception of school safety. While the perception of the dimension of school safety like child protective practices, discipline, and student-student relationships improved, results overall are not equivalent to the IYDA’s indicator target on school safety. As IYDA is expanding its safe school activities these result provides an opportunity to expand youth-focused activities such as school civic engagement activities, extracurricular clubs, and teacher-student engagement discussion group —for future cohorts. LIFE, WORK READINESS, AND LEADERSHIP SKILLS IYDA youth improve their life skills, work readiness skills, and leadership skills over the course of their participation in the WRN training intervention. Overall, 60.3 percent of youth improved at least one level on the work readiness assessment at midline from the baseline. These results suggest that the IYDA activities that influence skill outcomes may have a positive effect on youth. This includes work ethic, teamwork, resilience, analytical thinking, emotional intelligence, and confidence. Consequently, the findings of this report suggest the following results for IYDA: USAID/DRC IYDA COHORT I MIDLINE EVALUATION 1 Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores1 Percent Youth increasing scores from overall mean2 Q1. Do youth demonstrate improved in resilience to conflict and violence as a result of IDYA’s activities? 1 Percent of youth with increased self￾efficacy at the conclusion of programming 75% 33.7 of 48 36.63 of 48 .000 54.3% 61.9% 2 Percent of youth showing positive attitudes of tolerance and non￾violent means to address conflict drivers and conflict management. 75% 42.1 of 54 40.6 of 54 .000 35.1% 37.3% 3 Percent of target youth reporting a decreased propensity to engage in violence 75% 37.2 of 48 37.0 of 48 .831 42.6% 57.4% 4 Percent of youth with increased optimism at the conclusion of programming 75% 26.3 of 36 26.7 .175 54.3% 56.1% 1 Estimated from a sample of 94 matched participants within the sample from baseline to midline 2 Tabulated by assigning the mean scores from the baseline to all participants and calculating the percent improve at the midline. 3 Revised from the previously reported as 33.9 due to a revised syntax error this is updated for the paired sample indicator from the previously reported 47.9%. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 2 Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores1 Percent Youth increasing scores from overall mean2 Q2. Do youth demonstrate improved perception support and engagement in their community? 2.2 Percent of youth who believe that they are of value in society and their positive contributions are recognized and rewarded at the conclusion of programming 75% 48 of 72 64.0 of 72 .000 53.2 % 4 54.4% 2.3.2 Percent of youth who report feeling comfortable asking their parents or guardians for help or advice during the past 30 days5 75% 46.2 of 60 45.9 of 60 .442 45.7% 52.5% Q3. Do youth demonstrate improved perception of school safety or learning spaces? 1.2.1 Percent of youth with improved perception of the safety of schools or learning spaces 70% 224 of 306 227 of 306 .001 48.9% 53.8% Q4. Do youth demonstrate improved work readiness, as a 3.1.2 Percent of individuals with improved skills following completion of 60% NA NA .000 61.3% 60.3% 4 While previously reported as 84.0% revised analysis of the paired sample revealed an error in syntax code which when revised resulted in a 53.2% outcome 5 While previously reported with two results this indicator measure was revised to report the entirety of prosocial parental/caregiver support. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 3 Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores1 Percent Youth increasing scores from overall mean2 result of IYDA’s activities? USG-assisted workforce development programs Q5. Do youth participants demonstrate improved livelihoods as a result of IDYA’s activities?6 3.1 Number of individuals with new or better employment following completion of USG￾assisted workforce development programs 4,000 (~40 percent of 10,325 completers) NA NA NA NA NA 6 This question was not evaluated as part of the midline at the completion of IYDA’s training. It will be measured as part of the endline evaluation scheduled six months post training in Quarter 2 of FY20. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 4 BACKGROUND The USAID/DRC Integrated Youth Development Activity (IYDA) is funded by USAID and implemented by Education Development Center (EDC) and its consortium of partners Family Health International 360 (FHI 360), Catholic Relief Services (CRS), and Souktel. IYDA works in the North and South Kivu regions of the DRC, specifically in the cities of Goma and Bukavu and in the South Kivu territories of Kalehe, Kabare, and Walungu. IYDA provides vulnerable youth with learning pathways and inclusive economic opportunities as well as a network of services to support these pathways. The project works with existing structures to implement these activities – education and vocational training centers, governmental officials, the private sector, and local youth-serving organizations. In particular, IYDA works with accelerated learning programs (Centres de rattrapages Scolaires – CRSs), vocational training centers (Centres d’apprentissage professionnel – CAPs), and community literacy programs run by community￾based organizations (CBO). Vulnerable youth receive one academic year of basic education followed by 6 months of livelihoods and employment support. In addition, IYDA works with youth who will not go through the ALP pathways below but who will receive employment or entrepreneurship support through project interventions. These youth participate in a Savings and Internal Lending Communities (SILC) group with a direct linkage to employment or Work￾Based Learning opportunities, accompanied by employability and entrepreneurship training and support. IYDA’s Goal is to increase the resilience of youth to conflict and violence in eastern DRC. IYDA’s goal is supported by three intermediate results.  IR1: Youth gain and apply relevant skills for improved educational and social outcomes. Under IR1, we will strengthen and complement the work of the principal institutions that provide non-formal education to youth in the Kivus: ALPs (Pathway 1); vocational training schools (Pathway 2); and community-based literacy programs (Pathway 3).  IR2: Youth positively engaged in community activities. Activities under IR2 will strengthen youth participants’ social-emotional and life skills by affording them safe spaces, opportunities for belonging and agency, social networks and supportive relationships with adults to help them become drivers of and actors in community￾building initiatives.  IR3: Youth engaged in economic activities. Activities under IR3 will focus on Livelihoods training to provide youth with basic resilience, adaptation, and market￾driven skills that can assist in times of stress, identify immediate opportunities, and provide a foundation for future prosperity. IYDA’s Theory of Change is as follows: IF youth gain relevant skills and have opportunities for social and economic engagement, and IF community stakeholders are supportive of youth efforts, THEN youth will be more resilient to the impact of conflict and violence. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 5 The components analyzes in this report are based on the Positive Youth Development framework tailored to IYDA that focus on youth assets, agency, civic engagement (contribution), and alliance building (enabling environment). Figure 1: IYDA Youth Resilience Framework USAID/DRC IYDA COHORT I MIDLINE EVALUATION 6 Purpose From its inception, IYDA devised this evaluation as a means to generate opportunities to learn from its holistic approach to create an enabling environment for positive youth development (PYD). While this evaluation aids in understanding IYDA’s contributions to improving youth assets, agency, civic engagement (contribution), and alliance building (enabling environment), it also informs decision-making tool during program review and refinement processes. Evaluation Questions As a quasi-experimental study, this evaluation examines the following questions related to nine of IYDA’s outcome indicators: Q1. Do youth demonstrate improved resilience to conflict and violence (measured by improved perception of non-violent means to address conflict drivers and conflict management, self-efficacy and optimism about the future) as a result of IDYA’s activities? Indicator1: Percent of youth with increased self-efficacy at the conclusion of programming Indicator 3: Percent of target youth reporting a decreased propensity to engage in violence Indicator 2: Percent of youth showing positive attitudes of tolerance and non-violent means to address conflict drivers and conflict management. Indicator 4: Percent of youth with increased optimism at the conclusion of programming Q2. Do youth demonstrate improved perception support and engagement in their community? Indicator 2.2: Percent of youth who believe that they are of value in society and their positive contributions are recognized and rewarded at the conclusion of programming Indicator 2.2.1: Number of youth who participate in civil society activities following social or leadership skills training or initiatives from USG assisted programs Indicator 2.3.2: Percent of youth who report feeling comfortable asking their parents or guardians for help or advice during the past 30 days Q3. Do youth demonstrate improved perception of school safety or learning spaces? Indicator 1.2.1: Percent of youth with improved perception of the safety of schools or learning spaces USAID/DRC IYDA COHORT I MIDLINE EVALUATION 7 Q4. Do youth demonstrate improved work readiness, as a result of IYDA’s activities (with “improved” defined as an increase of one level on a multi-leveled life and leadership skills test)? Indicator 3.1.2: Percent of individuals with improved skills following completion of USG-assisted workforce development programs Q5. Do youth participants demonstrate improved livelihoods as a result of IDYA’s activities (with “improved livelihoods” defined as new or better employment as compared to baseline measures (with “better”, in turn, defined as higher paying, more stable, of higher quality, or in better working conditions)? Indicator 3.1 Percent of individuals with new employment following completion of USG-assisted workforce development programs As sufficient time has not passed post-training completion, the midline report does not respond to the question of livelihood change. This question will be examined as part of the final evaluation for Cohort I planned in Q2 of FY2020. METHODOLOGY Evaluation Design The evaluation design is quasi-experimental assessing a sample group of participants (Cohort I) enrolled longitudinally at three points:  Baseline in the period of enrollment in December 2018,  Mid-line(the current phase of this evaluation)at the month of completion of IYDA training for the indicators on agency, civic engagement (contribution), and alliance building (enabling environment); and  End-line, at four to six months post-training for indicators on assets and employment (planned in Q2 of FY20). Sampling The Sampling frame uses a clustered approach to select Level 3CRS and CAP schools across IYDA’s intervention in the regions of Goma, Walungu, Bukavu, Kalehe, and Kabare. Given the demographic differences, CRS and CAP schools were treated as independent sampling frames stratified by gender. The sample frame both the Baseline and Midline was planned as follows; Table 1: Evaluation study sampling frame CRS Sample Size CAP Sample Size CI =99%, α = 0.5, N=2,500 Sampl e size 30% attrition rate Total Sample size CI =99%, α = 0.5, N=1000 Sampl e size 30% attrition rate Total Sample size Number of students 526 158 684 Number of students 400 120 520 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 8 Number of schools Number of schools Number of Students per school 30 Total schools 23 Number of Students per school 36 Total schools 14 Tools Data was collected from youth using four tools — the Youth Perceptions Survey (YPS), Surveys of Perceptions of School Climate (SPSC), the Youth Employment Survey (YES) and the Anchored BFI (ABFI). YPS was developed by EDC (specifically for the USAID/ Mindanao Youth for Development (MYDEV) project in the Philippines), measuring youth perceptions of violence, their position in the community, and government. The YPS was adapted to the context in Eastern DRC through the addition of the PYD International Youth Development Survey (IYDS)7 subscale on Optimism, Positive Identity, Support, and Prosocial Norms. Given the sensitivity of the implementing context. SPSC8 was adapted from the Education in Crisis and Conflict Network (ECCN) Safe Learning Environment Toolkit. The survey aims to collect information about the perceptions of students on the characteristics of school life, or “school climate.” A positive school climate is one where all aspects of school life come together to nurture, support, and protect students’ overall wellbeing. YES was developed and validated in previous EDC projects and collects data on the nuances that characterize youth employment and underemployment in both the formal and informal sectors: hours worked per day, days worked per week, number of jobs, type of job, earnings and savings and one’s ability to meet needs and achieve goals 7 The IYDS youth survey is largely based on the Communities That Care Youth Survey (CTCYS) that was extended and adapted for use in Australia. The PYD Measurement Toolkit Illustrative Indicator and Corresponding description of IYDS scale item: Positive Identity: Increased positive identity after training/programming (youth self￾report), Support: Improved family support after of training/programming, Prosocial norms: Increased prosocial norms at the conclusion of training/programming 8 The Surveys of Perceptions of School Climate instruments include measures of respondents’ perceptions of school climate along the following dimensions: students’ sense of belonging in the school, acceptance of diversity, and inclusion; child protective practices and structures; discipline and fairness; student–student relationships; student–teacher relationships; and teacher– staff relationships. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 9 ABFI was developed by EDC through Workforce Connections and in partnership with Professional Examination Services (ProExam) and Akilah Institute for Women, based on the Big Five Inventory (BFI-44; Benet-Martinez & John, 1998) popular 44-item, self￾report measurement instrument developed to measure an individual’s Big Five skills (sometimes referred to as the “CANOE” skills). The A-BFI includes 8 Workforce Situational Judgment Tests (SJTs) which present a hypothetical scenario and ask the respondent to indicate how he or she would react in the workplace and measure respondent’s industriousness, self￾control, and perseverance. The SJT is scored by taking the absolute difference between experts’ and respondents’ responses on each of the five response options for a situational judgment test. The SJT is the core tool used to report on the indicator measure of “Percent of individuals with improved soft skills following completion of USG-assisted workforce development programs.” The A-BFI also includes the Big Five Inventory that consists of 44 self-reported simple statements that the respondent agrees or disagrees with using a five‐point Likert scale developed to measure an individual’s Big Five broad personality traits or ‘CANOE’ skills (BFI; Benet-Martinez & John, 1998). Items consist of short phrases, starting with “I see myself as someone who…” and ending with a prototypical Big Five trait marker (e.g., “is relaxed, handles stress well”). As individuals may respond to these questions using different standards, EDC added 4 Anchoring Vignette questions to reduce the effects of bias introduced by different cultural expressions of CANOE skills. Due to some limitations, the analysis for this evaluation report does not include the anchoring vignettes (see the limitations section for more information). USAID/DRC IYDA COHORT I MIDLINE EVALUATION 10 Tool Dimension Cronbach’s Alpha9 How Reliable is it? Anchored BFI BIG Five Inventory 0.70 Acceptable Anchoring Vignettes 0.39 Poor (removed from analysis) Situational Judgement Test 0.70 Acceptable Youth Perceptions Survey Perceptions of Community 0.86 Excellent Perceptions of Violence 0.721 Acceptable Perceptions of Positive Identity 0.67 Acceptable Perception of Self-efficacy 0.72 Acceptable Perception of Support 0.76 Acceptable Youth Employment Survey Confidence 0.92 Excellent Gender and Employment 0.77 Good Surveys of Perceptions of School Climate School Climate 0.80 Good Table 2: Reliability Tests for YES and YPS Data Collection IYDA engaged a local consulting firm Research Initiatives for Social Development (RISD) for data collection. Following the training from September 9-11, 2019, under the direction and training of IYDA’s M&E Specialist, 35 RISD enumerators began data collection on September 12, 2019, and ended on September 30, 2019. The questionnaires were administered orally then entered on a tablet using the data collection tool Survey to Go (STG) by enumerators. There were 1,165 participants and 1,111 validated observations included in the analysis at the baseline. The midline included 1,160 participants and 1,143 validated observations for analysis. Due to the challenges encountered during the data collection process the number of schools visited included those in the original sample and new schools to complete the sample. Of the participants sampled at Baseline, only 94 were retraceable (see limitation section). The validated sample for analysis was: Baseline CRS Sample Size CAP Sample Size Number of schools 24 Number of students 669 Number of schools 15 Number of students 442 Midline CRS Sample Size CAP Sample Size Number of schools 46 Number of students 770 Number of schools 22 Number of students 373 9 Cronbach's alpha is a measure of internal consistency, that is, how closely related a set of items are as a group make-up a composite score of a survey measurement of a given concept consistently . It is considered to be a measure of scale reliability. High reliability means it measures the concept or variable, while low reliability means it measures something else (or possibly nothing at all). USAID/DRC IYDA COHORT I MIDLINE EVALUATION 11 Data Analysis Data analysis was conducted using the Statistical Package for Social Sciences (SPSS). Descriptive statistics were used to summarize the socio-demographic characteristics of subjects. Numerical data are presented as mean based on their normality distribution. Categorical data are presented as frequency (percentage). A comparison of means statistical tests was used to measure differences between demographics where appropriate. Bivariate statistical and multivariate analyses (e.g., correlations, regression) are used to examine the relationship between different variables. Pearson's chi-squared (x 2 ) test of independence was also used to determine whether a statistically significant difference exists between two categorical variables. The null hypothesis of the Chi-Square test is that no relationship exists on the categorical variables in the population. To assess the probability that the null hypothesis is true, the p-value was compared to the predetermined .05 significance level. 10 The null hypothesis is that there is no significance in scores between the baseline and the midline. As a result of data collection limitations, the baseline and midline results were compared as independent samples. The indicator outcomes reflect the results of the gain scores through two sets of analyses. The first utilizes the gains score on each indicator domain from the 94 matched pairs from the baseline to the Midline. As the small size of the paired sample limits inference to IYDA’s Cohort Population, the second analysis of gain—assuming all else equals—assigned the mean scores from the baseline of each indicator measured to each individual at the midline to compare the changes. Limitations  Claims to Causality. The quasi-experimental design selected for Cohort I is considered relatively rigorous. However, as this approach does not have a counterfactual to represent what youth’s outcomes might have been without the Activity there are limits to claims to causality. As such, factors other than the presence or absence of the IYDA’s intervention may influence youth’s outcomes such that the findings in this evaluation cannot be claimed to be caused by IYDA’s interventions. Instead, throughout the report, we point to findings that suggest the influence of IYDA’s interventions on IYDA youth without claims of causality.  Bias. This evaluation relies on self-report methods of data collection. Intentional misreporting—specifically, sensitive questions— for reasons of social desirability11 or a misunderstanding of concepts constitutes factors that may affect self-reported data. In 10 .A value greater than .05 shows no statistically significant difference between the two conditions and therefore any differences in the mean is likely due to chance. A value less than or equal to .05 is statistically significant and therefore the differences in the means is not likely due to chance. 11 Social desirability is the tendency of some respondents to report an answer in a way they deem to be more socially acceptable than would be their "true" answer. They do this to project a favorable image of themselves and to avoid receiving negative evaluations. The outcome of the strategy is overreporting of socially desirable behaviors or attitudes and underreporting of socially undesirable behaviors or attitudes.” Paul J. Lavrakas USAID/DRC IYDA COHORT I MIDLINE EVALUATION 12 addition, the malleability of emotional perceptions—participants' mood at the time of data collection, emotional experience, and sensitivity to questions could also affect their response. Further, the complexity of the situational judgment test of the work readiness tool was not always comparable to youth comprehension skills and required a significant amount of time for enumerators to explain the scenarios. As a result, there is a potential for interviewer bias. To address these risks, IYDA adapted the questionnaire to clarify concepts and reduce potential bias and enumerators received training on surveying techniques in order to standardize their interaction with participants.  Work Readiness tool. At the baseline, when reliability tests were conducted the Cronbach’s Alpha for the Anchors of the BFI was 0.39, which was substantially lower than the acceptable reliability (0.7). As this reliability of the anchors did not improve significantly at midline, it was omitted from the analysis as it was at baseline. Limitations Specific to Cohort I  Data collection. As a result of the US Government Trafficking In People (TIP) sanctions on the DRC, several schools suspended their collaboration with IYDA per official notice from the DIVAS. Efforts to relocate baseline youth were further hindered by the timing of the data collection that coincided with the end of the summer months when youth were already out of school and mobile. As a result, 85 percent of the youth sampled at baseline could not be retraced by the data collection team at the midline. To remediate these challenges, the data collections team sampled additional schools targeted in IYDA’s Cohort I intervention population. While 94 matched pairs were identified from the baseline to the midline their demographics were not proportional to the overall Baseline and Midline demographics when disaggregated by region and school type. (See demographics). USAID/DRC IYDA COHORT I MIDLINE EVALUATION 13 DEMOGRAPHICS There were 1,165 participants and 1,111 validated cases included in the analysis at the baseline. A total of 24 CRS schools and 15 CAP schools were included in the sample with 669 youth in the Level 3 of CRS schools and 442 youth in the CAP schools. The midline included 1,160 participants and 1,143 validated observations for analysis. A total of 770 youth in the Level 3 of CRS schools and 373 youth in the CAP schools were included in the sample from 42 CRS schools and 22 CAP schools. By school type, the midline sample was significantly disproportional to the baseline sample p<.00212. While at Baseline Level 3 CRS youth represented 60.2 percent to the 39.8 percent of CAP youth, at the Midline, Level 3 CRS youth represented 67.4 percent to the 32.6 percent of CAP youth By geography, the midline sample is close to parity with at the baseline. Goma retains a higher concentration of youth while Kabare has the lowest. Overall, the percent of youth from urban regions increased from 52 percent at baseline to 59 percent at the midline (p<.000). (A list of schools sampled per territory is included in the Annex.) 12 “The p-value reflects the confidence with which we can rely on finding these differences again, in a re-sampling of IYDA youth. P<.000 means that, were we to re-draw the sample over and over again, we would find the same results 100% of the time. Similarly, p<.05 means that, were we to draw the sample of youth over and over again, we would likely find the same results 95% of the time. Statistically” similar or different refers to the reliability of our statement. In other words, we can say with a strong level of certainty that these groups are either similar or different to one of these high confidence levels. Figure 3. Geographic distribution survey participants at the Baseline (inner n=1,111) and Midline (outer n=1,143) 34% 23% 12% 17% 14% 36% 22% 11% 14% 16% Goma Bukavu Walungu Kalehe Kabare Female, 58.1% Female, 60.6% Female, 58.6% Female, 64.3% Male, 63.1% Male, 39.4% Male, 41.4% Male, 35.7% C R S ( N = 6 6 9 ) C A P ( N = 4 4 2 ) C R S ( N = 7 7 0 ) C A P ( N = 3 7 3 ) B A S E L I N E ( N = 1 , 1 1 1 ) M I D L I N E ( N = Figure 2. Sex Distribution of the sample within CRS and CAP USAID/DRC IYDA COHORT I MIDLINE EVALUATION 14 Youth age differed from the baseline at the midline. While at baseline, the average age of surveyed participants was 16.6 years old, at the midline the average age was 17.4 years old (p<.000). Within Level 3 CRS, the youth average age was 13.9 at midline compared to 15.7 at the baseline (p<.000). Similar to the baseline, within CAP schools, youth averaged 20.7 years old. Level 3 The number of youth who identified as handicap also increased at the midline. While at baseline 1 percent of those surveyed were identified as having a handicap—predominantly physical (27.3 percent)—at the midline this increased to 2.8 percent, (p<.002). Of the 2.8 percent, 27.3 percent identified as blind, while 18.2 percent were physically handicap. While there is some variance by age group and region, overall the baseline and midline demographics resemble the proportions of the number of youth enrolled in IYDA’s Cohort I. 25.3% 57.0% 15.7% 0.0% 1.2% .6% .1%.1% 56.1% 35.2% 6.2% 1.8% .6% 0.0% L O W E S T T H R U 1 2 1 3 T H R U 1 5 1 6 T H R U 1 9 2 0 T H R U 2 4 2 5 T H R U 3 0 3 1 T H R U 35 3 6 T H R U H I G H E S T Baseline (n=669) Midline (n=770) Figure 4. CRS Baseline Midline Age Group Distribution .5% 9.0% 44.6% 30.1% 8.8% 3.4% 3.6% 0.0% 5.4% 41.0% 36.2% 12.1% 5.4% 0.0% L O W E S T T H R U 1 2 1 3 T H R U 1 5 1 6 T H R U 1 9 2 0 T H R U 2 4 2 5 T H R U 3 0 3 1 T H R U 35 3 6 T H R U H I G H E S T Baseline (n=442) Midline (n=373) Figure 5. CAP Baseline Midline Age Group Distribution USAID/DRC IYDA COHORT I MIDLINE EVALUATION 15 IYDA Cohort 1 Population Baseline Midline Overall 4485 Percent 1111 Percent 1143 Percent Gender Female 2734 61.0% 640 57.6% 691 60.5% Male 1749 39.0% 471 42.4% 452 39.5% School Type CRS 2459 54.8% 669 60.2% 770 67.4% CAP 2026 45.2% 442 39.8% 373 32.6% Region Goma 2187 48.8% 383 34.5% 417 36.5% Bukavu 1091 24.3% 250 22.5% 252 22.0% Walungu 408 9.1% 130 11.7% 124 10.8% Kabare 339 7.6% 155 14.0% 185 16.2% Kalehe 460 10.3% 193 17.4% 165 14.4% Age Group Under 13 242 5.4% 171 15.4% 0 0.0% 13 thru 15 1784 39.8% 421 37.9% 452 39.5% 16 thru 19 1498 33.4% 302 27.2% 424 37.1% 20 thru 24 655 14.6% 141 12.7% 183 16.0% 25 thru 30 191 4.3% 43 3.9% 59 5.2% 31 thru 5 105 2.3% 16 1.4% 25 2.2% Above 35 10 0.2% 17 1.5% 0 0.0% USAID/DRC IYDA COHORT I MIDLINE EVALUATION 16 YOUTH PERCEPTIONS IYDA’s approach focuses on building youth foundations for resilience. The interventions implemented across all three IR aims to build youth agency, assets, alliance building, and civic contribution and engagement towards increasing youth resilience to conflict. Overall, has a target of 75% of youth who demonstrate improved resilience to conflict and violence (measured by improved perception of non￾violent means to address conflict drivers and conflict management, self-efficacy and optimism about the future) and perception support and engagement in their community as a result of IDYA’s activities. To measure these outcomes, IYDA utilized the Youth Perception Survey (YPS). For the exception of community engagement, the YPS measure each dimension though a set of statements which is then used to create an aggregate overall score across all statements. This overall score is then ranked by percentile using a 5-point Likert scale from 1(“Very Poor” 59 percent and below) to 5 (“Excellent” 90 percent and above) as noted in the table below. For this evaluation, the reported indicator outcome measure reflects the gain scores from the paired sample of 94 participants and the gain scores comparing midline scores to the overall baseline means—assuming all else equals. Dimensions Number of Statements Total Possible Score How scores are ranked Community Perception 12 72 Excellent (90-100%) Prosocial Parental/Caregiver Involvement and Support 10 60 Good (80-89%) Self-Efficacy 8 48 Fair (70-79%) Engagement in violence 8 48 Poor (60-69%) Non-Violence and tolerance 9 54 Very Poor (0-59%) Positive identity 6 36 COMMUNITY ENGAGEMENT While the full scope of IYDA programming on community service and civic engagement did not fully launch with Cohort 1, through the work readiness training, IYDA aimed to increase youth community engagement, support, and competencies linked to civic engagement. The community engagement component of the YPS measures youth’s attitudes toward their communities and their role in it as youth At baseline, an average of 22.1 percent of youth surveyed (246 out of 1,111) participated in a civil society association in their community. At the midline, this increased to an average of 36.3 percent of youth severed, a gain of 14.2 percent from the baseline (p<.000). USAID/DRC IYDA COHORT I MIDLINE EVALUATION 17 As it was at the baseline, youth in CAP schools (51.2 percent) were more involved in associations in their community compared to Level 3 CRS students (29.1 percent), (p<.000). While no statistical differences exist by gender from the baseline to the midline, by territory, strong differences existed. While at baseline, youth in Bukavu were the least engaged in community associations at 14.4 percent compared to the average of 22.1 percent, at the midline youth they were significantly more engaged than the average and all other regions at 40.9 percent, (p<.003). By age, as it was at the baseline, youth in the extended core age of 25 to 35 (68.0 percent) showed a greater rate of involvement in community civil society organization, than those in other age range, (p<.000). This may due in part to the more concerted effort to adhere to the age demographics targeted in CAP and CRS . At the midline, 82.4 percent of those involved in civic community associations were involved in at least one association similar to the baseline. However, with an increase from14.3 percent at the baseline to 16.6 percent at the midline, a greater number more youth were involved in two to three associations at the midline. The majority of youth were involved in youth associations which speaks well of IYDA’s focus on building the capacity of the YSOs to provide greater avenues for youth involvement in youth￾focused associations. For those who remain disengaged in community associations, the reasons remain the same as the baseline. The majority of youth were either not interested (29.4 percent) or lacked time with school obligations (17.1 percent). 23.2% 14.4% 23.8% 30.1% 20.6% 22.1% 76.8% 85.6% 76.2% 69.9% 79.4% 77.9% 35.3% 40.9% 37.9% 35.2% 32.4% 36.3% 64.7% 59.1% 62.1% 64.8% 67.6% 63.7% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% Goma Bukavu Walungu Kalehe Kabare Total Baseline Involved Baseline Not Involved Midline Involved Midline Not Involved Figure 6. Perception of Community Engagement by territory Baseline (n=1,111)- Midline (n=1,143) 35.8% 15.4% 21.5% 4.1% 39.8% 53.3% 8.7% 20.2% 5.3% 30.8% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% Youth Association Sports Association Volunteer Association / Community Service Civic/Political Engagement Association Other Baseline (n=246) Midline (n=415) Figure 7. Community Engagement through a number of associations USAID/DRC IYDA COHORT I MIDLINE EVALUATION 18 PERCEPTION OF COMMUNITY IYDA’s interventions seek to improve youth engagement with their community. The community perception component of this study consisted of a series of 12 statements13 measuring youth’s attitudes toward their communities and their role in it. These attitudes are measured over time through gain score, indicating the change in youth’s perceptions of their communities from baseline to the midline. The answer options are on a Likert scale and are scored on a range of 1 to 6. These findings, explored further below, suggest that, by encouraging youth to be engaged in community service activities, IYDA may help in improving youth’s perceptions of their communities. Youth’s perception of their role in the community Specifically, across the four statement statements focused on youth’s perception of their role in the community. The average responses in the agreement or strong agreement increased from an average of 45.2 percent at the baseline to 48.5 percent at the midline (p<.000). Disaggregation by gender and school type revealed no significant variations on this outcome. The most notable change was to the statement “In my community, young people understand their roles as important partners of the leaders solving the problems of the community” which increased by 14 percent from the baseline. This higher awareness of youth’s roles at the midline indicates the importance of IYDA’s systems-based stakeholders’ engagement, service-learning activities, and the civic engagement modules of IYDA’s Work Ready Now! (WRN!) Training. These activities support youth to become more aware of their role in building an enabling environment within their community through avenues of collaboration and learning between youth and their communities. 13 These statements are a composite of EDC’s YPS tool from USAID’s MYDEV project and includes the subscale on prosocial community involvement from The Communities That Care Youth Survey (CTCYS) Developed by The Social Development Research Group (SDRG) at the University of Washington (United States)13 the CTCYS is noted at tool for prosocial community support in USAID/Youth Power Positive Youth Development Toolkit USAID/DRC IYDA COHORT I MIDLINE EVALUATION 19 Youth’s perception of prosocial community support Compared to the statements of youth’s role in the community, changes in the four statements on community pro-social support of youth were more subtle. Overall, the average responses in the agreement or strong agreement decreased from an average of 68.7 percent at the baseline to 66.6 percent at the midline (p<.000). This result is due in part to the average decrease of 4 percent across the three statements specific to positive community support. Disaggregation by gender and school type revealed no significant variations on this outcome. One could posit that as the youth became more aware of their role within their community, they also became more aware of the type of support they require that is unavailable within their community. To address this in the future, IYDA will need to amplify the engagement of community leaders -including parent associations and other community groups- in its interventions through Youth Service Organizations (YSOs) and Youth Development Alliances. A paradigm shift of community awareness of youth’s needs and potential from within their community would support a drive towards a more enabling environment for youth. 3% 2% 3% 2% 3% 2% 7% 3% 20% 16% 15% 12% 5% 6% 5% 2% 24% 20% 25% 20% 19% 10% 20% 7% 5% 4% 5% 6% 5% 6% 6% 3% 7% 17% 16% 20% 12% 23% 15% 23% 27% 24% 25% 23% 33% 29% 29% 35% 14% 17% 11% 16% 24% 24% 18% 26% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Adults in my community create organized activities for people my age. (Χ2(6) = 61.284, p < .000) Adults in my community provide ways for people my age to develop new skills. (Χ2(6) = 32.061, p < .000) There are a lot of chances in my community for Youth to participate in community service. (Χ2(6) = 81.600 p < .000) In my community, young people understand their roles as important partners of the leaders solving the problems of the community. (Χ2(6) = 143.920, p < .000) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 8. Perception of Community and Role Statements from Baseline (n=1,111) to Midline (n=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 20 Youth’s perception of Gender and Community Access In assessing youth perceptions of their community, it was also important for IYDA to employ a gender lens. IYDA continues to refine and develop a gender transformative approach with a focus on themes such as inclusive leadership, positive masculinities, S/GBV prevention, and female empowerment. At the midline, perceptions of community gender roles and access to community leaders saw some positive changes. While at baseline, on average 50.7 percent of all youth surveyed perceived a positive role of young men and women community at the midline this increased to 56.5 percent, (p<.003). While this is a positive change, it is important to note that responses in the agreement or high agreement to the two statements on women’s roles in the community were lower to those focused on males’ roles and engagement. As it was at the baseline, with no significant difference between the way female and male youth responded to these questions, we can assume that both male and female youth perceived that men have greater opportunity to assume useful roles and access to leaders in their community compared to women. The RERA/DNH and the Gender Analysis confirm the need for gender-inclusive governance, equitable access to community leaders, and decision making processes (women in decision-making and civic engagement, etc.) within the implementing context. As the midline results show, a larger societal paradigm shift is needed in order to arrive at more equitable concepts of gender roles. While there are many factors outside of IYDA’s control, as today’s youth will become tomorrow’s leaders, they certainly have an opportunity and a responsibility to advance more equitable gender outcomes, partially through the support of IYDA’s interventions. 6% 4% 2% 2% 2% 1% 2% 1% 6% 3% 4% 2% 3% 1% 3% 2% 12% 7% 6% 5% 6% 4% 7% 5% 5% 4% 3% 3% 1% 2% 3% 3% 15% 24% 14% 21% 14% 21% 13% 20% 35% 32% 45% 37% 45% 37% 44% 36% 21% 27% 27% 31% 29% 33% 28% 33% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline In my community, youth can have a positive future. (Χ2(6) = 69.509, p < .000) There are people in my neighborhoo d who are proud of me when I do something well. (Χ2(6) = 35.105, p < .000) There are people in my neighborhoo d who encourage me to do my best. (Χ2(6) = 45.989, p < .000) My neighbors notice when I do a good job and let me know about it. (Χ2(6) = 40.663, p < .000) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 9. Perception of Community Prosocial Support from Baseline (N=1,110) to Midline (n=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 21 Overall Results With a possible score of 72 across the twelve statements of positive community perception, at baseline youth averaged a score of 48.0 (66.7% of the total score). The midline saw an improvement demonstrated by an increase of the average score to 52.0 (72.2% of the total score) (p<.003). This variance is largely due to more youth responding as “Agree” and ‘Strongly agree” across the statements on community perceptions, which is statistically significant. Disaggregation by gender and school type revealed no significant variations on this outcome. Across the twelve statements, 57.2 percent of all youth perceived their community and their role within the community as “good” or “excellent” with 33.6 percent reported at the ranking highest level. This is a positive increase from 54.9 percent of all youth at the baseline, (p<.000). While disaggregation by gender, school type and age band revealed no statistical difference from the baseline to the midline, results by region were statistically significant as increased by regions were not proportional, (p<.000). While Kalehe, Kabare, and Goma saw a larger increase from the baseline, Bukavu and Walungu increased only slightly from the baseline. Community Perception Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on feeling valued in society and recognition and reward of their positive contributions to their community at the conclusion of programming (Indicator 2.2). Based on the paired sample of surveyed youth, 53.2 percent of youth perceptions of the community improved at least one level from baseline to baseline. Assuming all else equals, in comparing the individual midline, scores to the baseline mean score of 48.0 across the twelve statements, 54.4 percent of youth improved at least one level on their perception of their community. These results are noted in the following table. 3% 3% 2% 2% 4% 3% 4% 3% 19% 10% 8% 3% 0% 4% 4% 2% 28% 17% 17% 10% 20% 10% 13% 5% 6% 7% 5% 4% 5% 5% 3% 3% 12% 24% 14% 21% 12% 21% 13% 19% 23% 27% 34% 34% 32% 31% 39% 36% 10% 13% 20% 26% 18% 26% 25% 33% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline In my community, young women are given equal opportunity to assume useful roles in the community. (Χ2(6) = 120.208, p < .000) In my community, young men are given equal opportunity to assume useful roles in the community. (Χ2(6) = 74.863, p < .000) Young women have access to community leaders to voice their concerns and opinions (Χ2(6) = 106.173, p < .000) Young men have access to community leaders to voice their concerns and opinions (Χ2(6) = 80.390 p < .000). I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 10. Perception of Community Gender Roles from Baseline (n=1,110) to Midline (n=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 22 Community Perception Indicator Results from Midline-Baseline Means gains Gender (Χ 2 (1)> = .240, p = .628) School Type (Χ 2 (1)> = .581, p = .448) Region (Χ 2 (4) = 66.401, p < .000) Age band (Χ 2 (4) = 5.076, p <=.280) Overall Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progress 319 202 357 164 167 163 70 52 69 212 194 83 26 6 521 46.2% 44.7% 46.4% 44.0% 40.0% 64.7% 56.5% 31.5% 37.3% 46.9% 45.8% 45.4% 44.1% 24.0% 45.6% Improved 372 250 413 209 250 89 54 113 116 240 230 100 33 19 622 53.8% 55.3% 53.6% 56.0% 60.0% 35.3% 43.5% 68.5% 62.7% 53.1% 54.2% 54.6% 55.9% 76.0% 54.4% N 691 452 770 373 417 252 124 165 185 452 424 183 59 25 1143 PERCEPTION OF PROSOCIAL PARENTAL/CAREGIVER SUPPORT Pro-social parental/caregiver support is measured through 10 self-reported statements assessing the level of presence, reliability, and support to youth by parents/caregivers. During data collection, youth are asked to reflect on their experience of parental/caregiver support in the last 30 days prior to the date of the survey as they respond to each of these statements. This scale combines the prosocial components of EDC’s YPS tool with the Perceived Social Support Questionnaire subscale of family support.14 Overall Results Based on the responses in agreement or strong agreement across the nine positive statements, on average, 71.4 percent of all youth at midline demonstrated a positive perception of parental involvement and support in their lives compared to 64.8 percent at the baseline. Across all statements, youth showed positive changes in their perceptions. This is particularly interesting as IYDA’s interventions involving parents including discussion on safe parenting, and psychosocial support did not implement during Cohort I activities. However, with IYDA Work Ready Now! Curriculum which focuses on communication, leadership and self-management youth are being given the tools to become more effective communicators. As a result, youth develop higher self-esteem which fosters more prosocial behavior and feelings about one’s familial relationships. This may have indirectly contributed to improved communications with parents and caregivers to more effectively convey their needs and ask for support. 14 Developer: Procidano, M.E. & Heller, K Scales were developed to assess an individual’s perceived level of social support from family and friends. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 23 0% .3% 0% .2% 0% .1% 0% .2% 0% .3% 1% 1.5% 1% 1.5% 5% 4.7% 9% 5.1% 4% 3.6% 4% 3.1% 3% 3.3% 7% 6.4% 14% 8.6% 5% 5.6% 2% 2.5% 2% 2.0% 4% 4.6% 5% 6.5% 1% 3.6% 5% 13.4% 9% 16.4% 17% 20.4% 19% 20.6% 11% 15.3% 44% 35.2% 44% 32.9% 38% 29.5% 30% 30.6% 44% 33.5% 44% 44.0% 40% 43.7% 30% 34.3% 23% 28.5% 34% 38.1% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline My parent or caregiver are good at talking with me about things. (Χ2(6) = 64.857, p < .000) I If I had a personal problem, I could ask my parent or caregiver for help or advice. (Χ2(6) = 41.932 p < .000) My parent or caregiver give me lots of chances to do fun things with them. (Χ2(6) = 22.583, p < .001) My parent or caregiver ask me what I think before most family decisions affecting me are made. (Χ2(6) = 34.568, p < .000) I can rely on my parent or caregiver for emotional support. (Χ2(6) = 38.924, p < .000) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 11. Perception of Prosocial Parent Support Baseline (N=1,110) to Midline (N=1,143) 0% 0% 0% 0% 0% 0% 0% 0% 1% 1% 4% 3% 1% 1% 2% 2% 2% 3% 11% 4.5% 8% 5% 2% 2% 2% 3% 5% 6% 9% 9.2% 3% 4% 2% 3% 2% 1% 3% 4% 2% 3.7% 11% 21% 11% 16% 11% 15% 12% 17% 4% 11.0% 0% 32% 47% 40% 47% 40% 44% 34% 28% 27.3% 30% 34% 38% 38% 37% 39% 34% 36% 45% 43% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline My parent or caregiver were sensitive to my personal needs (Χ2(6) = 70.229, p < .000) My parents or caregiver notice when I am doing a good job and let me know about it. (Χ2(6) = 25.221 p < .000) My parents or caregiver tells me they’re proud of me for something I did. (Χ2(6) = 17.251 p < .008) I enjoyed spending time with my parents or caregiver. (Χ2(6) = 30.664, p < .000) I wish my parents were different.(r) (Χ2(6) = 71.586, p < .000) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 12. Continued Perception of Prosocial Parent Support Baseline (N=1,110) to Midline (1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 24 With a possible score of 60 across the10 statements measuring positive perception of parental involvement, at baseline, the average youth scored a 46.2 (77 percent of the total score). At the Midline, the average youth’s score was 45.9 (76.5 percent) which is statistically similar to the baseline (p=.454). As Level 3CRS youth average a higher score (46.6) compare to CAP youth (44.7), by school type the results on parental/prosocial support were statistically significant at the midline (p<.000). Given the age difference between youth in CAP schools compared to youth in Level 3 CRS schools, it can be inferred that naturally, parents would have a higher presence in their lives are they are still minors. Interestingly, disaggregation by gender and regions showed no statistical variance from the baseline to the midline. Prosocial Parental/Caregiver Support Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their level prosocial parental/caregiver support. Based on the paired sample of surveyed youth, 45.7 percent of youth’s perceptions of prosocial parental/caregiver support improved at least one level from the baseline. Assuming all else equals, in comparing the midline scores to the baseline mean score of 46.2, overall, 52.5 percent of youth improved at least one level. Prosocial Parental/Caregiver Support Indicator Results from Midline-Baseline Means gains Gender School Type Region Age band Overall (Χ 2 (1)> = 2.303, p = .130 (Χ 2 (1)> = 6.046, p < .016) (Χ 2 (4) =21.019, p < .000) (Χ 2 (4) = 15.152, p < .004) Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progress 340 202 346 196 185 149 68 59 81 212 194 83 26 6 542 49.3% 44.7% 44.9% 52.7% 44.5% 59.1% 54.8% 35.8% 43.8% 46.9% 45.8% 45.4% 44.1% 24.0% 47.5% Improved 350 250 424 176 231 103 56 106 104 240 230 100 33 19 600 50.7% 55.3% 55.1% 47.3% 55.5% 40.9% 45.2% 64.2% 56.2% 53.1% 54.2% 54.6% 55.9% 76.0% 52.5% N 690 452 770 372 416 252 124 165 185 452 424 183 59 25 1142 The role of parents and caregivers are pertinent to youth prosocial behaviors and resilience. Links between parental styles and practice are important to youth perceptions of parental, empathy, support, and overall prosocial behavior. As the parents of Cohort I youth did not benefit from IYDA parental support activities (specifically the intergenerational discussions series that is slated to begin in year two), these results are important for IYDA’s learning. While there is a multitude of factors that affect youth’s perception of prosocial support, parents/caregivers' roles helps in building an enabling environment for youth. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 25 PERCEPTION OF SELF-EFFICACY Self-efficacy is defined as one’s belief in their abilities to do many different things well. Building strong self-efficacy is at the core of agency within the framework for positive youth development (PYD). Agency is where “youth perceive and can employ their assets and aspirations to make or influence their own decisions about their lives and set their own goals, as well as to act upon those decisions to achieve desired outcomes”15 . Youth self-efficacy is assessed through a subscale of eight statements (4 positive and 4 negative) as recommended by the PYD Indicator reference sheet for this indicator16 . Self-efficacy measured The four positive statements of the self-efficacy scale have a strong focus on youth’s confidence in their capabilities to face life’s challenges. Across these statements at baseline, 49.2 percent of all youth’s responses were in agreement or strong agreement of their ability to meet the challenges ahead. The midline saw a positive change to 55.8 percent of youth surveyed. This is a difference of 6.6 percent more youth who felt strong enough to overcome life’s struggles, deal effectively real-world situations; or proactively think through their actions towards others and themselves (p<.000). Desegregations by gender and school types revealed no statistical difference in these outcomes. 15 PYD Framework of the PYD toolkit 16 The measure includes the items from the Generalized Self-Efficacy or GSE (Judge, Locke, Durham, & Kluger, 1998*). 1% 1% 1% 1% 1% 1% 1% 0% 8% 4% 0% 2% 8% 4% 3% 3% 18% 8% 16% 9% 17% 8% 9% 6% 8% 7% 8% 7% 7% 7% 3% 3% 18% 29% 25% 30% 22% 29% 21% 18% 34% 32% 33% 34% 31% 32% 40% 34% 13% 19% 10% 17% 13% 19% 22% 35% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline I am strong enough to overcome life’s struggles (Χ2(6) = 57.5228, p < .000) I feel competent to deal effectively with the real world. (Χ2(6) = 75.461 p < .000) I can handle the situations that life brings. (Χ2(6) = 75.128, p < .000) I can think of the consequences of my actions to other persons and to myself (Χ2(6) = 47.818, p < .000) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 13. Positive Perception of Self-efficacy from Baseline (N=1,110) to Midline (N=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 26 On the opposite spectrum of the self-efficacy subscale are the four negative statements related to youth’s overall resilience. While at the baseline, on average 17 percent of youth answered in agreement or strong agreement to perceptions of low self-efficacy at the midline this decreased to 9.4 percent (p<.000). Overall more youth more in agreement or strong agreement of perceived themselves with a higher sense of self-efficacy at the midline then at the baseline. Desegregations by gender and school types revealed no statistical difference in these outcomes. We can deduct from these results that youth overall at the midline was more likely to feel like they can bounce back and recover from failure and negative thoughts about themselves showing not only improved self-efficacy but also improved self-esteem. Overall Results From the overall score of 48 across the eight statements, at baseline, the average score was 33.7 (70.2 percent). Results at the midline showed positive changes with an average score of 36.6 (76.3 percent). On average youth surveyed at the Midline scored 2.9 points above the baseline which is a significant improvement (p<.000). While results by gender were statistically different at the baseline as males ranked their level of self-efficacy higher compared to female youth, at the midline scores by gender were statistically similar, and female scores were on par with males. Gender 1% 1% 1% % 2% 1% 1% 1% 30% 36% 25% 36% 24% 35% 25% 39% 39% 32% 33% 33% 35% 34% 41% 32% 10% 12% 10% 12% 11% 10% 12% 12% 7% 10% 8% 9% 8% 9% 9% 9% 11% 8% 18% 7% 14% 8% 9% 5% 3% 2% 5% 2% 5% 4% 3% 3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline I often feel that there is nothing that I can do well. (r) (Χ2(6) = 30.871, p < .000) At root, I am a weak person. (r) (Χ2(6) = 96.144, p < .000) I usually feel that I am an unsuccessful person. (Χ2(6) = 50.401, p < .000) I often feel like a failure. (r) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 14. Negative Perception of Self-efficacy (N=1,110) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 27 equitable outcomes such as these may speak to IYDA’s strong focus on gender transformative approaches throughout its interventions especially through the WRN! Training, Self-Efficacy Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their level self￾efficacy at the conclusion of programming (Indicator 1). Based on the paired survey youth, 54.3 percent of youth improved at least one level on their perception of self-efficacy. Assuming all else equals, in comparing the midline scores to the baseline mean, 61.9 percent of youth improved at least one level on their perception of self-efficacy. Self-Efficacy Indicator Results from Midline-Baseline Means gains Gender School Type Region Age band Overall (Χ 2 (1)> = 2.694, p = .105 (Χ 2 (1)> = 10.314, p < .001) (Χ 2 (4) =9.541, p < .049) (Χ 2 (4) = 29.675, p < .000) Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progress 276 159 318 117 147 108 57 63 60 214 145 54 16 6 435 40.0% 35.2% 41.3% 31.5% 35.3% 42.9% 46.0% 38.2% 32.4% 47.3% 34.2% 29.5% 27.6% 24.0% 38.1% Improved 414 293 452 255 269 144 67 102 125 238 279 129 42 19 707 60.0% 64.8% 58.7% 68.5% 64.7% 57.1% 54.0% 61.8% 67.6% 52.7% 65.8% 70.5% 72.4% 76.0% 61.9% N 690 452 770 372 416 252 124 165 185 452 424 183 58 25 1142 PERCEPTION OF VIOLENCE General violence, gangs, and armed group recruitment are tangible risks within the implementing context. Note that the RERA/DNH and the initial baseline determined that 20.9 percent of youth surveyed reported the presence of a gang or members of armed groups in proximity to their school. Specifically, Level 3 CRS school students (25.6 percent) reported a higher presence of a gangs or armed group members than CAP students (13.8 percent) surveyed (p<.000). By territory, Kalehe ranked the highest in reports of gang and armed group presence (35.8 percent), followed by Kabare (29.0 percent), Bukavu (23.6 percent), Goma (13.1 percent), and Walungu (5.9 percent) p<.000). In order for IYDA to understand youth’s perception of engagement in and tolerance of violence, the youth perception of violence was adapted. Perception of violence is measure through two sets of statements. The first set of statements assesses the acceptability and engagement in violent actions and violent groups as a means of conflict resolution. The second set of statements assesses the acceptability of non-violence and peaceful means of conflict resolution. Youth expressed their agreement to each statement using the same Likert Scale of the YPS and scores were assigned ranging from 0 to 6, with higher scores reflecting agreement to peaceful means of conflict resolution. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 28 ENGAGEMENT IN GANG AND ARMED GROUP VIOLENCE The propensity to engage in gangs and armed group violence and violent groups was measured through three main questions comprised of six sub-statements justifying engagement, a statement on the likelihood to join a violent group, and a statement on the capacity to resist recruitment. Overall Results At the midline, IYDA youth demonstrated positive changes in their perception of engaging with gangs and armed groups. While at baseline 43.6 percent of youth disagreed at the highest levels to the six reasons justifying involvement in gangs and armed groups at the midline this increased to 49.3% (p<.000). IYDA’s male and female youth demonstrated on average, similar positive change scores in their perceptions of violence from baseline to the midline. Interestingly, however, at both baseline and Midline, IYDA female youth scored significantly higher than male youth in disagreement with the reasons justifying joining an armed group (p<.000). Overall, more youth reported confidence in their skills to resist recruitment into armed groups at 55.4 percent at the midline to 47.4 percent at the baseline (p<.000). However, fewer youth reported that they were unlikely to join an armed group or gang at the midline at 86.4 percent compared to 90 percent of youth at the baseline, (p<.004). While this undesired change could be the byproduct of several variables including the environment, IYDA will need to continue to increase its efforts in building youth resilience to recruitment. This may also include adding training on the foundations to resilience for youth similar to that provided to teachers. 3% 1% 3% 1% 4% 1% 3% 1% 3.4% 2.1% 3.6% 1.9% 4% 6% 8% 13% 6% 11% 3% 4% 3.4% 1.6% 1.1% 2.0% 6% 4% 7% 6% 10% 6% 9% 8% 7.5% 4.4% 6.1% 3.0% 3% 8% 5% 8% 4% 5% 5% 8% 3.2% 5.6% 2.3% 5.2% 4% 6% 4% 6% 5% 6% 5% 9% 6.0% 7.7% 5.9% 6.5% 32% 24% 31% 22% 32% 25% 32% 24% 35.1% 24.9% 34.6% 26.1% 47% 51% 43% 44% 40% 45% 43% 46% 41.3% 53.7% 46.5% 55.3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline a) If they are being discriminat ed by others in the community /country (Χ2(6) = 64.430, p < .000) b) If they are suffering from poverty because of unemploy ment and lack of income (Χ2(6) = 68.133, p < .000) c) If they wanted to belong to a group and do, what the members are doing. (Χ2(6) = 50.806 p < .000) d) If they are being pressured by friends/fa mily members (Χ2(6) = 42.830 p < .000) e) If they want to protect themselves (Χ2(6) = 67.743 p < .000) f) If they are being disrespect ed (Χ2(6) = 67.743 p < .000) I do not know Strongly agree Agree Somewhat agree Somewhat disagree Disagree Strongly disagree Figure 15. Negative Perception Engagement in Violence (N=1,110) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 29 Across the eight statements, from the overall score of 48, on average, youth surveyed scored 37.2 on the likelihood of refraining from engagement in violence (77.5 percent) at the midline which is similar to the baseline with an average score of 37.1 (77.3 percent). While disaggregation by gender and school type did not show any variance, by geography some difference were noted. While disaggregation by gender and school type did not show any variance, by geography some differences were noted. While average scores for Goma, Kalehe, and Kabare increased (+3.4 points net gains), average scores for Bukavu and Walungu decreased (-6.8 net loss), (p<.000). From a high score rank of good or excellent, while 57.4 percent of youth show high marks at midline compared to 58.7 percent baseline. There was not a correlation between these outcomes and youth reporting of the presence of gangs and armed group members near their schools. Engagement in gang and Armed Groups Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their propensity to engage in gangs and armed group violence at the conclusion of programming (Indicator 3). Based on the paired survey youth, 42.6 percent of youth improved at least one level of demonstrated positive changes in their perception of engaging in violent groups. Assuming all else equals, in comparing the midline results to the overall baseline mean, 57.4 percent of youth improved at least one level on their perception of engaging in violent groups. Engagement In Violent Group Indicator Results from Midline-Baseline Means gains Gender School Type Region Age band Overall (Χ 2 (1)> = 1.119, p = .299 (Χ 2 (1)> = .222, p < .655) (Χ 2 (4) =26.256, p < .000) (Χ 2 (4) = 9.283, p < .054) Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progress 285 201 324 162 167 126 71 59 63 171 193 82 31 9 486 41.3% 44.5% 42.1% 43.5% 40.1% 50.0% 57.3% 35.8% 34.1% 37.8% 45.5% 44.8% 53.4% 36.0% 42.6% Improved 405 251 446 210 249 126 53 106 122 281 231 101 27 16 656 2% 1% 3% 1% 1% 3% 19% 40% 2% 3% 28% 15% 1% 1% 10% 7% 4% 6% 3% 6% 26% 22% 13% 15% 64% 64% 25% 16% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline It is likely that I will join a gang or violent groups in my community (Χ2(6) = 19.049 p < .004) I feel that I have the skills to resist being recruited by armed groups. (Χ2(6) = 178.569 p < .000) I do not know Strongly agree Agree Somewhat agree Somewhat disagree Disagree Strongly disagree Figure 16. Negative Perception Engagement in Violence (N=1,110) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 30 58.7% 55.5% 57.9% 56.5% 59.9% 50.0% 42.7% 64.2% 65.9% 62.2% 54.5% 55.2% 46.6% 64.0% 57.4% N 690 452 770 372 416 252 124 165 185 452 424 183 58 25 1142 PERCEPTION OF NON-VIOLENCE AND TOLERANCE Attitudes on tolerance and violence are measure through three sets of statements. The first set of (four) statements promotes the acceptability of violent actions as a means of conflict resolution. The second set of (three) statements promotes the acceptability of non-violence and peaceful means of conflict resolution. Two additional statements were included to understand youth’s perception of gender roles and violence. Overall Results IYDA Youth at the midline shows positive changes in their perception of non-violence. While at the baseline on average, 82.8 percent of youth agreed or were in strong agreement in supporting non-violent actions at the midline this increase to 85.8 percent. This means that IYDA youth have moved significantly towards supporting non-violent or peaceful conflict resolution, (p<.000). While differences school types existed as CAP students responded at a greater degree of agreement to non-violence compared to Level 3 students, results by gender were similar across these statements. Consistent with the results on non-violence, IYDA Youth surveyed also showed positive changes in attitudes of tolerance at the midline. This is such that while youth overall agreed with avoiding, walking away, or talking out a problem instead of resorting to violence consistently from the baseline (82.8 .5% .5% .1% .7% .4% .9% .8% 1.6% 2.8% 2.7% 3.3% 2.4% 3.4% 1.0% 3.8% 1.1% 2.0% 3.5% 8.9% 3.6% 6.7% 3.2% 5.4% 3.2% 1.3% .9% 2.3% 1.8% 5.9% 2.4% 11.9% 3.5% 2.5% 2.5% 7.5% 4.8% 4.5% 7.9% 5.2% 8.6% 32.0% 21.3% 32.1% 28.0% 39.6% 30.4% 39.4% 33.4% 59.0% 68.5% 45.8% 58.6% 39.5% 54.2% 33.5% 48.6% 0% 20% 40% 60% 80% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Violence is an acceptable means to resolve conflict in the community. (Χ2(6)> = 37.488 p < .000) If I walk away from a fight, I’d be a coward ® (Χ2(6)> = 60.657 p < .000) Sometimes violence is the only way to express your feelings. (Χ2(6)> = 97.012 p < .000) Sometimes a person doesn’t have any choice but to use violence. (Χ2(6)> 122.573= p < .000) I do not know Strongly agree Agree Somewhat Agree Somewhat disagree Disgree Strongly disagree Figure 17. Perception of Non-Violent Actions Baseline (N=1,110) to Midline (N=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 31 percent) to the midline (82.7 percent) more youth strongly agreed with these statements at the midline than at the baseline. These results could be due in part to exposure to IYDA activities that focus on leadership, communication, and active listening that are incorporated within the WRN! modules. These activities indirectly teach youth conflict resolution which includes ways to analyze a situation, think critically, and decide on more favorable outcomes. Conventional notions of violence generally associate men as the perpetrators and women as the victims. Across the youth surveyed, this convention held. Compare to 43.0 percent at the baseline, 48 percent of youth surveyed agreed that men are always the perpetrator of violence. Meanwhile compared to 37.1 percent at the baseline 42.6 percent of youth surveyed at the midline agree that women are always the victim. Interestingly overall responses across both statements from both males and females were statistically similar. However, by school type compared to youth in Level 3 CRS, youth in CAP schools were more likely to agree that men are always the perpetrators (p<.016) and women are always the victim of violence (p<.002). A spearman Rank Correlation shows as positive association between school type (r=.091, n=1,143, p<.002), age (r=132, n=1,143, p<.000) and the likelihood that one believes that men are always the perpetrators of violence. Given the differences between CRS and CAP school environment especially in age range and the history of violence within the implementing context we can assume that those who are older have been more exposed to violence where men are most frequently the perpetrators and women and children are most frequently the victims. 3% 1.9% 4% 6% 1.5% 2.0% 5% 2.7% 6% 6% 2.3% 2.9% 1% 2.1% 1% 1% 1.5% 2.0% 7% 7.1% 6% 6% 11.0% 11.2% 38% 27.2% 34% 28% 42.0% 33.9% 44% 58.3% 48% 53% 41.5% 47.6% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Baseline Midline There are non- violent ways to resolve conflict in the community. (Χ2(6)> = 53.737p < .000) Avoiding or walking away from someone who wants to fight you is an effective way to avoid violence (Χ2(6)> = 13.2438 p < .039) I try to talk out a problem instead of using violence (Χ2(6)> = 18.025 p < .006) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Agree Strongly agree Figure 18. Perception of Non-Violent Response Baseline (N=1,110) to Midline (N=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 32 Overall the average score across the non-violence and tolerance subscale decreased slightly from 42.1 (78.0 percent of the total score of 54) at the baseline to 40.6 (75.2 percent) at midline, (p<.001). While results by gender and school type on the overall score were similar at the midline, results by regions showed some variance. While the scores for Kalehe and Kabare were above the mean, average scores for Bukavu, Walungu, and Goma were at the mean (p<.015). Attitudes of Tolerance and Non-Violence Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their attitudes of tolerance and non-violent means to address conflict drivers at the conclusion of programming (Indicator 2). Based on the paired survey youth, 35.1 percent of youth improved at least one level of demonstrated positive changes in their perception of engaging in violent groups. Assuming all else equals, in comparing the midline results to the overall baseline mean, 37.3 percent of youth improved at least one level on their perception of engaging in violent groups. Attitudes of Tolerance and Non-Violence Indicator Results from Midline-Baseline Means gains Gender School Type Region Age band Over (Χ all 2 (1)> = .855, p = .381 (Χ 2 (1)> = .467 p = .514) (Χ 2 (4) =16.017, p < .003) (Χ 2 (4) =5.920, p = .205) Fema le Mal e CRS CA P Gom a Buka vu Walun gu Kale he Kaba re 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progres s 440 276 488 228 233 171 74 109 129 302 253 113 33 15 716 63.8% 61.1 % 63.4 % 61.3 % 56.0 % 67.9% 59.7% 66.1% 69.7% 66.8 % 59.7 % 61.7 % 56.9 % 60.0 % 62.7% 1% 1% 1% 2% 15% 18% 15% 15% 19% 11% 24% 15% 9% 10% 14% 13% 12% 11% 8% 13% 27% 23% 23% 22% 16% 25% 14% 21% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Men are always the perpetrators of violence. (Χ2(6)> = 37.488 p < .000) Women are always the victims of violence. (Χ2(6)> = 47.480 p < .000) I do not know Strongly agree Agree Somewhat Agree Somewhat disagree Disgree Strongly disagree Figure 19. Gendered Perception of Violence Baseline (N=1,110) to Midline (N=1,143) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 33 Improv ed 250 176 282 144 183 81 50 56 56 150 171 70 25 10 426 36.2% 38.9 % 36.6 % 38.7 % 44.0 % 32.1% 40.3% 33.9 % 30.3% 33.2 % 40.3 % 38.3 % 43.1 % 40.0 % 37.3% N 690 452 770 372 416 252 124 165 185 452 424 183 58 25 1142 PERCEPTION OF POSITIVE IDENTITY & OPTIMISM Within the PYD framework, positive identity is defined as positive and coherent attitudes, beliefs, and values that one holds about him/herself and his/her future. The YPS measures positive identity through six self-reported statements adapted from the International Youth Development Survey from the PYD toolkit. Overall Results From the overall score of 36 across the six statements, at baseline, the average score of positive identity score was 26.3 (73.1% of the total score). At the Midline, this measure remained statistically similar as youth scored on an average of 26.7 (74.2 percent of the total score), (p=.177). As it was at the baseline, youth scores remained, were statistically similar by gender, and statistically different by school type as CAP youth average score was slightly higher (27.2) than the average of level CRS youth (26.4), (p<.005). Interestingly, there was no correlation between youth age and positive identity scores. This leads to the conclusion that the difference in score between CAP and CRS youth is not the result of age but rather the difference in the environment. This Disaggregation by territory also showed statistical differences between regions from the baseline to the midline. While the average score decreased by one point for Goma and remained the same for Bukavu, Walungu, score in Kalehe, and Kabare increased by one point (p<.009). Overall response in the agreement or strong agreement to positive identity remained relatively similar to the baseline. While at baseline 83.7 percent of youth answering in the agreement or strong agreement to the first three statements regarding belief in/knowledge self and optimism of the future, at the midline this amounts to 80.5 percent of respondents. However, less youth response in the agreement or strong agreement on three statements positive view of life decreasing from 39.7 percent at the baseline to 37.5 at the midline. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 34 Based on the responses in the agreement or strong agreement across all six statements, at baseline on average, 35.2 percent of youth perceived a high level of positive identity and optimism (good or excellent). At midline, the results show an average of 36.2 percent. While this result is statistically significant, that is largely due to a decrease of the lowest rankings (very poor and poor) from 32.8 percent at the baseline to 26.9 percent at the midline (p<.004). Disaggregation by school type, gender, region, and age group did not reveal any significant variance. It was hypothesized that positive identity—as a result of IYDA’s social and life skills training and other youth-specific interventions—would yield a higher increase at the midline. However, these results may mean that positive identity is a layered process of forming self-image, self-esteem, self-efficacy, and overall emotional growth over time during general development. While changes were not observed in the short term, IYDA is building youth strengths and skills needed to use throughout their lives to prepare themselves for success and cope with the challenges to positive identity. Given the limited scope of the positive identity subscale, further research, through a mixed-method approach is recommended to IYDA in order to investigate further the effects of its life skills training on youth self￾esteem and self-image. 3% 3% 1% 1% 0% 0% 1% 0% 1% 1% 2% 1% 2% 1% 1% 1% 1% 3% 6% 6% 20% 14% 7% 8% 3% 3% 5% 3% 2% 2% 17% 10% 27% 23% 16% 15% 1% 2% 2% 1% 2% 2% 6% 7% 10% 18% 6% 8% 7% 11% 11% 13% 7% 11% 20% 28% 21% 27% 21% 22% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline Baseline Midline When I think about the future, I feel optimistic. (Χ2(6) = 18.576, p < .005) I know what kind of person I am. (Χ2(6) = 13.079, p < .042) I really believe in myself. (Χ2(6) = 36.359, p < .000) Things usually turn out well for me. (Χ2(6) = 41.465, p < .000) I think the world and people in it are basically good. (Χ2(6) = 52.840, p < .000) I feel okay about the way I’ve handled myself so far. (Χ2(6) = 19.147, p < .004) I do not know Strongly disagree Disagree Somewhat disagree Somewhat agree Figure 20. Perception of Positive Identity from Baseline (N=1,110) to Midline (N=1,143) ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 35 Positive Identity & Optimism Indicator Results IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their level of positive identity and optimism at the conclusion of programming (Indicator 4). Based on the paired survey youth, 54.3 percent of youth perceptions improved at least one level on their perception of Positive Identity & Optimism. Assuming all else equals, in comparing the midline results to the overall baseline mean, 56.1 percent of youth improved at least one level on their perception of optimism and positive identity. Positive Identity & Optimism Indicator Results from Midline-Baseline Means gains Gender School Type Region Age band (Χ Overall 2 (1)> = .078, p = .807 (Χ 2 (1)> = 4.789, p < .031) (Χ 2 (4) =5.120, p = .275) (Χ 2 (4) = 1.891, p = .756) Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of Progress 276 159 355 146 165 117 58 73 88 208 178 81 23 11 501 40.0% 35.2% 46.1% 39.2% 39.7% 46.4% 46.8% 44.2% 47.6% 46.0% 42.0% 44.3% 39.7% 44.0% 43.9% Improved 414 293 415 226 251 135 66 92 97 244 246 102 35 14 641 60.0% 64.8% 53.9% 60.8% 60.3% 53.6% 53.2% 55.8% 52.4% 54.0% 58.0% 55.7% 60.3% 56.0% 56.1% N 690 452 770 372 416 252 124 165 185 452 424 183 58 25 1142 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 36 PERCEPTION OF SCHOOL SAFETY The Survey of Perceptions of School Climate (SPSC) instrument measures student’s perceptions of a safe learning environment through six dimensions of a holistic school climate. A holistic school climate is define as one “where social, emotional, and academic development are nurtured and supported; where the interpersonal relationships between and among teachers, students, and staff are open and positive; and where teachers and students of diverse sexual orientations and gender identities are both respected and have a voice in school affairs… When students’ social and emotional development is nurtured in school, they become empowered to make decisions about their own safety and to report and challenge violence when they encounter it personally or as a witness”17 . Each dimension of the SPSC is scored using a 0-6 Likert scale of "Strongly Disagree" to 1 "Strongly Agree" with reverse coding of negative statements which is then utilized to create an aggregate overall score across all statements. This overall score is then ranked by percentile using a 5-point Likert scale from 1(“Very Poor” 59 percent and below) to 5 (“Excellent” 90 percent and above) as noted in the table below. The indicator outcomes reflect the results of the gain scores from the paired sample of 94 youth and the gain scores using a comparison of the individual midline scores to the overall baseline means— assuming all else equals. OVERALL PERCEPTION OF SCHOOL SAFETY The perception of school safety saw some positive changes at the midline. While at Baseline, only 18.1 percent of youth perceived the level of their school safety as good or excellent, this increased to 26.0 percent at the midline. Overall, from a total score of 306 across the statements of the SPSC, perception scores of learning environment increased on average by 4 points significantly from 224 (72.9 percent) at the baseline, to 228 (74.5 percent) at the midline, (p<.001). Interestingly while at baseline, youth from CRS schools scored their schools below the 17 USAID/RTI International. 2016. Conceptual Framework for Measuring School-Related Gender-Based Violence. Washington, DC: U.S. Agency for International Development. Dimensions Number of Statements Total Possible Score How scores are ranked Students’ Sense of Belonging, Acceptance of Diversity, and Inclusion in school 10 60 Excellent (90-100 percentile) Child Protective Practices and Structures, Psychosocial support 7 42 Good (80-89 percentile) Discipline and Fairness 9 54 Fair (70-79 percentile) School Safety in and on route to school 14 84 Poor (60-69 percentile) Student–Student Relationships 4 24 Very Poor (0-59 percentile) Student–Teacher Relationships 7 42 Overall Perception Score 51 306 Table 5. Dimensions of the Surveys of Perception of School Climate Score and Ranking USAID/DRC IYDA COHORT I MIDLINE EVALUATION 37 average with a score of 221 (72.2 percent) at the Midline score increased to above the average at 226, (p<.000). However, while the average scores for CAP schools increased from 229 (74.8 percent) to 231, no statistical differences were shown (p=.173). Between school types, statistical differences still exist at the midline similar to the baseline as less youth from Level 3 CRS rated their learning enviroment as ‘Good or Excellent’ compared to CAP youth, and overall rating remaining as fair. From IYDA’s RERA/DNH and Gender Analysis, we know that while male students face risks in and out of schools, female students face considerably more risks specifically School-related Gender-Based Violence (SRGBV). In school, boys are more likely to receive corporal punishment, face bullying, on route they are more at risk to banditry, gangs, and armed group recruitment. Whereas girls in school are more likely to face denigration, insults, and verbal abuse, and often find themselves unable to endure this trauma in what they consider humiliation and shame in addition to sexual harassment and other forms of gender violence both in and out of school. At the baseline, this was shown as males rated their schools statistically higher level of school safety (226 of 306) compared to female survey youth (222 of 306) (at p<.013). While there were some improvements, scores on school safety were similar at the midline between males (228 of 306) and females (227 of 306). By School, type overall the average score of school safety did not increase significantly from the baseline. Overall, 24.2 percent of youth in Level 3 CRS schools and 29.7 percent of youth in CAP schools ranked their schools as ‘good’ or ‘excellent.’ Overall perception of School Safety by gender School Safety Ranks Baseline (n=1,111) Midline (n=1,143) Female Male Total Female Male Total CRS Very Poor 37.6% 33.0% 35.6% 31.5% 26.6% 29.5% Poor 1.9% 3.4% 2.5% .7% .6% .6% Fair 44.9% 47.5% 46.0% 43.0% 49.5% 45.7% Good 13.2% 12.5% 12.9% 21.1% 19.1% 20.3% Excellent 2.4% 3.7% 3.0% 3.8% 4.1% 3.9% 35.6% 22.4% 30.4% 29.5% 25.2% 28.1% 2.5% 1.8% 2.3% .6% .8% .7% 46.0% 54.2% 49.3% 45.7% 44.2% 45.2% 12.9% 17.9% 14.9% 20.3% 20.9% 20.5% 3.0% 3.6% 3.2% 3.9% 8.8% 5.5% C R S C A P T O T A L C R S C A P T O T A L B A S E L I N E ( N = 1 , 1 1 1 ) M I D L I N E ( N = 1 , 1 4 3 ) Very Poor Poor Fair Good Excellent Figure 21. Overall perception of School Safety by school type ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 38 Top Rank 15.6% 16.2% 15.9% 24.9% 23.2% 24.2% CAP Very Poor 25.0% 18.5% 22.4% 26.3% 23.3% 25.2% Poor 2.2% 1.2% 1.8% .8% .8% .8% Fair 53.4% 55.5% 54.2% 42.5% 47.4% 44.2% Good 17.9% 17.9% 17.9% 20.4% 21.8% 20.9% Excellent 1.5% 6.9% 3.6% 10.0% 6.8% 8.8% Top Rank 19.4% 24.8% 21.5% 30.4% 28.6% 29.7% At baseline, by territory, Bukavu and Kabare were the lowest-rated in terms of safety, as only 11.6 percent of youth in Bukavu and 12.3 percent of youth in Kabare rated the level of safety of their learning environment as ‘good’ or ‘excellent’ at baseline. At the midline, Kalele and Kabare saw the most significant increases from the baseline. In Kalehe 27.9 percent of survey youth rated their school safety as high (p<.000) compared to 17.1 percent at baseline and while in Kabare 18.9 percent rated their school high compared to 11.6 at baseline (p<.000). Belonging, Diversity, & Inclusion From a score of 60, at baseline on average youth rated their learning environment as fair with a score of 43.9 (73.2 percent), at the midline the results were similar with a score of 43.9 (p=.972). While no statistical differences were observed by gender, school type of region, disaggregation by territory was 23.8% 27.8% 32.4% 34.5% 15.4% 18.5% 34.2% 21.8% 51.0% 31.9% 1.6% .7% 4.0% .8% 0% 1.6% 2.6% 0% 2.6% .5% 51.6% 37.4% 52.0% 50.4% 60.0% 49.2% 46.1% 50.3% 34.2% 48.6% 14.4% 20.4% 11.6% 13.1% 23.1% 28.2% 17.1% 27.9% 11.6% 18.9% 8.6% 13.7% 1.2% 1.5% 2.4% 0% 0% .6% 0% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% Baseline (n=382) Midline (n=417) Baseline (n=250) Midline (n=252) Baseline (n=130) Midline (n=124) Baseline(n=193) Midline ((n=165) Baseline (n=155) Midline (n=185) Goma Bukavu Walungu Kalehe Kabare Very Poor Poor Fair Good Excellent Figure 22. Overall perception of School Safety by region ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 39 statistically different as Bukavu, Kalehe and Kabare remain at the lowest on this domain at p<.016.. While, 30.1 percent of youth surveyed perceived a high sense of belonging, acceptance of diversity, and inclusion in school at baseline, at the midline this number increased to 34.5 percent. Child Protective Practices From a score of 42 across the 7 statements, at baseline, youth on average rated their learning environment as fair with a score of a 29.2 (69.5 percent of the total score) at the midline this remained statistically similar at to 30.6 (72.9 percent), at p<.000. As it was at the baseline, disaggregation by school type was statistically different as CRS Level 3 students rated this domain the lowest compared to CAP students surveyed (p<.040). While disaggregation by gender showed no variance, disaggregation by territory was statistically different (p<.000). This is such as Bukavu rated the lowest on this domain compared to other regions as it was at the baseline. While at baseline, 19.4 percent perceived a high level of child protective practices and structures within their schools, at the midline this increased to 28.1 percent. 11.5% 11.1% 11.4% 14.2% 16.1% 14.8% 23.3% 16.3% 20.5% 22.7% 20.9% 22.1% 40.4% 34.5% 38.0% 28.7% 27.9% 28.4% 20.0% 30.6% 24.2% 20.3% 20.4% 20.3% 4.8% 7.5% 5.9% 14.2% 14.7% 14.3% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent Figure 23. Overall perception of Belonging and Inclusion ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 40 Overall, 19.4 percent of the youth surveyed perceived their learning environments as having increased child protective practices and structures and psychosocial support. Disaggregation by school type was statistically different at p<.010, as CRS Level 3 students rated this domain the lowest compared to CAP students surveyed. While results were statistically similar by gender in CRS schools, within CAP these results were statistically different. Discipline and Fairness Of the youth surveyed at baseline, only 30.1 percent rated their learning environment as having ‘good or excellent’ discipline and fairness practices. At the Midline, this increased to 42.3 percent. From a score of 54 across the 7 statements on average youth rated their learning environment as fair with a score of a 40.8 which mirrors the baseline score (75.4 percent of the total score). Disaggregation by school type was statistically different (p<.000). As it was at the baseline, CRS level 3 youth-rated this domain the lowest compared to CAP students surveyed. While results were statistically similar by gender statistical variance were shown by region compared to other results on the SPSC, Bukavu rated the lowest on this domain compared to other regions at p<.000. Learning from the RERA, discipline, and fairness is a major concern for students and parents alike. While all of IYDA’s school safety activities did not occur during the interventions for Cohort I, teacher training and reviews of the school code of conduct were among the means through which IYDA began to address these concerns. Still, these results show that there is still more ground to cover. 12.6% 5.5% 9.7% 10.8% 4.1% 8.6% 17.1% 14.4% 16.0% 21.6% 13.5% 19.0% 29.8% 33.1% 31.1% 31.5% 27.3% 30.2% 30.1% 34.7% 32.0% 25.5% 34.4% 28.4% 10.3% 12.3% 11.1% 10.7% 20.7% 13.9% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent USAID/DRC IYDA COHORT I MIDLINE EVALUATION 41 School Safety in and on route to school At baseline, of the youth surveyed, 15.7 percent18 viewed their learning environment as having a high level of safety on route to and from school. At the midline, this increased significantly to 28.1 percent (p<.000). From a score of 84 across the 14 statements measuring this domain, the average score increased to 60.5 percent at the midline from 57.4 (68.3 percent). No differences by school type or gender were shown in this domain as similarly to the CRS Level 3 youth and female students rated this domain the lowest. Disaggregation by territory was statistically significant as Bukavu and Walungu rated the lowest on this domain compared to other regions, (p<.000). As it was at the baseline however, on average youth ranked their learning environment as poor on this domain. These results are especially interesting as the project is still in the process of adapting it’s school safety measures. Given the nature of the implementing context, many variables are outside of IYDA’s scope, still, we are making progress through activities such as teacher training activities in safe school measures and development of school safety implementation plans. 18 Previously reported at 30.1 percent erroneously on the baseline Figure 25. Overall perception of Discipline and fairness ti 17.3% 12.0% 15.2% 13.2% 9.1% 11.9% 26.5% 24.7% 25.8% 18.2% 22.3% 19.5% 39.8% 39.2% 39.5% 41.7% 38.1% 40.5% 13.2% 18.6% 15.3% 19.5% 19.8% 19.6% 3.3% 5.4% 4.1% 7.4% 10.7% 8.5% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent Figure 24. Overall perception of Child Protective Practices (N=1,110) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 42 Student–Student Relationships While 30.1 percent of the youth surveyed perceived student-to-student relationships as ‘good’ or ‘excellent’ at baseline this increased to 43.4 at the midline. From a score of 24 across the 4 statements measuring this domain, while youth rated their learning environment as fair similarly at baseline (18.0 of 24) and midline (18.3 of 24), the results overall were statistically different (p<.026). While no variance was noted by gender and school type, disaggregation by territory was statistically different as Kabare rated the lowest on this domain compared to other regions, (p<.000). IYDA’s WRN training reinforces youth ability to lead, communicate effectively, cooperate, and foster community through civic engagement. These skills are pertinent in the student-student interaction as the aid students to address conflict in a positive matter thus, overall improving student-student relationships. 23.9% 13.6% 19.8% 14.2% 13.1% 13.8% 38.6% 35.4% 37.3% 30.4% 31.4% 30.7% 23.2% 33.3% 27.2% 28.7% 24.7% 27.4% 10.3% 13.2% 11.4% 19.0% 19.0% 19.0% 4.0% 4.5% 4.2% 7.8% 11.8% 9.1% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent Figure 26. Overall perception of Safety on route ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 43 Student–Teacher Relationships Of the youth surveyed at baseline, 41.519 percent perceived student to teacher relationship as highly positive. At the midline, 43.4 percent perceived the same outcome which is statistically different at p<.000. From a score of 42 across the seven statements measuring this domain, youth average score decreased slightly to 33.5 (79.8 Percent) at midline from 34.8 (82.9 percent) at baseline. Disaggregation by gender and school type was statistically similar to the baseline, as CRS Level 3 students rated this domain similarly to CAP students surveyed. These results may be due to a myriad of factors including the results of increasingly informed youth holding teachers to higher standards, or a negative impact of teachers feeling stress or overwhelmed due to the new material. It is recommended that IYDA look more closely at these factors for future implementation. Overall, no differences were shown through a disaggregation by gender. However, as Bukavu rated the lowest on this domain compared to other regions, disaggregation by region was different, (p<.000). Overall youth rated their learning environment as good on this domain. 19 Previously reported as 30.1 15.1% 9.0% 12.8% 8.0% 7.8% 7.9% 20.5% 18.8% 19.8% 21.2% 21.0% 21.2% 23.3% 29.8% 25.9% 28.3% 25.7% 27.5% 20.0% 20.3% 20.1% 24.4% 26.3% 25.0% 21.0% 22.0% 21.4% 18.1% 19.2% 18.4% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent Figure 27. Overall perception of Student-Student Relationship ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 44 School Safety Indicator Conclusions IYDA project has a target of 75 percent of youth that ‘improve’ one level or more on their level of perception of school safety or learning spaces at the conclusion of programming (Indicator 1.2.1). Based on the paired youth surveyed, 48.9 percent of youth perceptions of school safety improved at least one level from the baseline to the midline. Assuming all else equals, in comparing the midline individual scores to the baseline mean score, 53.8 percent of youth improved at least one level on their perception of school safety. While there was not a significant association between gender, school type, and age band on youth scores of school safety, the outcome there was a significant association by region. Differences between regions are mostly due to the low percentage of youth with improved school safety perceptions in Bukavu and Kabare compared to the means. N-1,143 Gender (Χ 2 (1)> = 2.803, p = .094) School Type (Χ 2 (1)> = 2.834, p = 0.100) Region (Χ 2 (4) = 33.795, p < .000) Age Band (Χ 2 (4)> = 6.698, p = .153) Total Female Male CRS CAP Goma Bukavu Walungu Kalehe Kabare 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of progress 333 195 369 159 174 151 44 64 95 227 190 79 24 8 528 48.2% 43.1% 47.9% 42.6% 41.7% 59.9% 35.5% 38.8% 51.4% 50.2% 44.8% 43.2% 40.7% 32.0% 46.2% Improved 358 257 401 214 243 101 80 101 90 225 234 104 35 17 615 51.8% 56.9% 52.1% 57.4% 58.3% 40.1% 64.5% 61.2% 48.6% 49.8% 55.2% 56.8% 59.3% 68.0% 53.8% N 691 452 770 373 417 252 124 165 185 452 424 183 59 25 1143 Baseline, overall IYDA youth rated their learning environment safety as fair at the midline similar to the baseline. While safe learning environment interventions such as school improvement plans were delayed for Cohort I, findings on school safety suggest that IYDA is making some improvements for youth though teacher training, which has translated to 2.8% 2.0% 2.5% 4.9% 4.0% 4.6% 7.0% 4.8% 6.1% 9.2% 9.1% 9.2% 29.9% 25.9% 28.3% 33.5% 26.0% 30.9% 31.1% 37.9% 33.7% 37.1% 37.0% 37.1% 29.3% 29.5% 29.4% 15.2% 23.9% 18.0% 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% CRS CAP Total CRS CAP Total Baseline (n=1,111) Midline (n=1,143) Very Poor Poor Fair Good Excellent Figure 28. Overall perception of Student-Teacher Relationship (N=1,110) ti USAID/DRC IYDA COHORT I MIDLINE EVALUATION 45 improved perceptions of youth’s feelings of safety, security and wellbeing in a school environment. IYDA’s use of a competency-informed curriculum for teachers coupled with efforts to improve teachers wellbeing and build teachers’ psychosocial support skills via inclusive teaching delivery strategies and specific resilience-building modules appears to have a positive effect on student’s perceptions of wellbeing. IYDA’s resilience-building teacher curriculum is helping them to learn not only how to best support students and address unhealthy coping strategies and behaviors, but also how to build a holistic school climate. While these results are not equivalent to the IYDA’s indicator target, it may speak to the impact of these activities. While the project cannot change all the context of risks raised through the RERA/DNH, the focus on building teachers' resiliency and sense of wellbeing is pertinent to IYDA’s interventions toward building an enabling environment for positive youth development—specifically safe learning environments (SLE). While IYDA should scale up its SLE interventions—specifically via youth-focused activities such as school civic engagement activities, extracurricular clubs, and teacher-student engagement discussion group —for future cohorts, preliminary results on the SPSC provides a positive trajectory for growth. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 46 WORK READINESS SKILLS At the core of IYDA’s interventions is its intensive Life Skills training that develops a youth’s foundational life skills, work readiness skills, and leadership skills. Coupled with practical technical skills training, community service projects, participation in community-based youth networks, and involvement in civic activities, this intensive skills training is intended to prepare youth for meaningful and productive engagement in their communities, with their families, and at their jobs. BIG FIVE INVENTORY OVERALL RESULTS The Big Five Inventory (BFI) measures the five broadest dimensions of personality in an organizing framework that has emerged to guide research on soft skills. The BFI consists of 44 simple statements that the respondent agrees or disagrees with using a five‐point Likert scale20(Gosling, Rentfrow & Swann Jr., 2003). Since the BFI-44 is a diagnostic without a set benchmark, at baseline IYDA’s youth BFI scores were compared to the findings from Schmitt et al.’s data collection in seven African nations in 2007. Comparable to Schmitt et al findings, IYDA’s youth scored highest on agreeableness and conscientiousness, and the lowest on emotional stability and openness at the baseline. Findings at the midline, suggest that IYDA’s youth improved their skills in the BFI dimensions for extraversion (+0.09), agreeableness (+0.10), conscientiousness (+0.15), and openness (+0.26) at the midline compared to the baseline, (p<.000). Specifically for the dimension on openness—which showed the highest gains—at the completion of IYDA’s intervention’s youth scored themselves as having a higher propensity to adopt analytical and divergent thinking strategies. Why does this matter to the workplace? Openness involves higher thinking skills. A person who is high in openness tends to be intellectually curious. In the workplace, this person is more reflective, creative, analytical, which results in more innovation. . Second, to Openness, youth scored themselves on the dimension of conscientiousness at the end of IYDA’s interventions. This matter for the workplace as a highly conscientious person will tend to behave higher work ethic, dependability, and grit. However, on the dimension for emotional stability, youth scored themselves statistically similar from the baseline to the midline (p<.048). Specifically for the transition to the workforce, emotional stability is important. A person who is low in emotional stability tends to be moody, depressed, and susceptible to stress. 20 John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory‐‐Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 47 BFI Disaggregated At the school level, youth consistently improved their scores on the BFI domains. While Level 3 CRS youth are not on the continuum of transition to employment—but rather transition to further education—their level of improvement was similar to youth in CAP schools. However, on the dimension of emotional stability, CRS Level 3 youth scored themselves lower at the midline compared to the baseline which may be due to the age group of these youth. While youth show a higher, sense of awareness of their emotional stability it also shows the need for continued support especially within CRS schools towards building safer learning environments. Baseline and Midline BFI Averages by School Type Baseline (N=1,111) to Midline (n=1,143) Average Score Extraversion Agreeableness Conscientiousness Emotional Stability Openness CAP Youth Baseline (n=442) 3.47 4.27 4.32 2.20 3.78 Midline (n=373) 3.52 4.33 4.44 2.20 3.97 Difference 0.05 0.07 0.12 0.01 0.19 CRS Youth Baseline (n=669) 3.35 4.20 4.17 2.32 3.37 Midline (n=770) 3.47 4.34 4.36 2.23 3.71 Difference 0.12 0.13 0.18 -0.10 0.34 While overall, males statistically scored themselves higher on all BFI dimensions compared to female youth at baseline. At the midline, females showed the highest gains. For the exception of the dimension on emotional stability, female youth showed higher progressions 3.38 3.87 3.72 2.76 3.61 3.39 4.23 4.23 2.27 3.48 3.54 4.33 4.38 2.22 3.80 0.00 1.00 2.00 3.00 4.00 5.00 Extraversion Agreeableness Conscientiousness Emotional Stability Openness Africa Baseline(n=1111) Midline (n=1,143) Figure 29. DRC IYDA Baseline and Midline BFI Averages versus Africa Regional Averages (N=1,111) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 48 on the dimensions for extraversion (+0.09 points), agreeableness (+0.11 points), conscientiousness (+0.18 net points), and openness (+0.30 points) at the midline compared to the baseline. While mean scores at the baseline were statistically different by gender, at the midline only the score on extraversion, emotional stability, and openness was statistically different. These results suggest that IYDA’s gender transformative approach to building agency specifically for vulnerable youth, within the implementing context may be leveling the playing field between the sexes. Average BFI Score(* denotes Significant Difference) Extraversion Agreeableness Conscientiousness Emotional Stability Openness Baseline (n=1,111) Female (n=640) 3.35 4.20 4.19 2.33 3.43 Male (n=471) 3.45 4.26 4.29 2.20 3.67 Difference 0.11 0.06 0.10 -0.12 0.24 P-value .000* .049* .003* .001* .000* Midline (n=1,143) Female (n=691) 3.45 4.31 4.37 2.27 3.73 Male (n=452) 3.53 4.36 4.41 2.15 3.89 Difference 0.07 0.05 0.04 -0.12 0.16 P-value .011* .087 .169 .004* .000* Overall Difference Baseline to Midline Female 0.10 0.11 0.18 -0.06 0.30 Male 0.07 0.10 0.12 -0.05 0.22 P-value Disaggregation by zone showed consistency with the overall midline results. Except for on the agreeableness scale, youth in rural and urban areas on average scored themselves similarly. Desegregation by region however showed some variances between the baseline and midline. While results on all five dimensions were statistically different for Kalehe and Kabare, and Goma, the statistical difference only exists on emotional stability for Bukavu and on Extraversion and Openness for Walungu. Average BFI Score (* denotes Significant Difference) Extraversion Agreeableness Conscientiousness Emotional Stability Openness Goma Baseline 3.39 4.19 4.29 2.24 3.78 Midline 3.52 4.39 4.47 2.13 4.09 Difference 0.13 0.20 0.18 -0.12 0.31 P-value .000* .000* .000* .011* .000* Bukavu Baseline 3.45 4.25 4.21 2.17 3.46 Midline 3.53 4.16 4.20 2.35 3.54 Difference 0.07 -0.08 -0.01 0.18 0.08 P-value .175 .055 .802 .001* .174 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 49 Walungu Baseline 3.50 4.42 4.27 2.03 3.37 Midline 3.35 4.32 4.27 2.17 3.59 Difference -0.16 -0.10 -0.01 0.14 0.22 P-value .016* .090 .927 .073 .010* Kalehe Baseline 3.31 4.21 4.23 2.44 3.47 Midline 3.51 4.39 4.43 2.26 3.76 Difference 0.19 0.18 0.20 -0.18 0.29 P-value .000* .000* .000* .007* .000* Kabare Baseline 3.26 4.14 4.11 2.51 3.28 Midline 3.55 4.40 4.48 2.26 3.66 Difference 0.29 0.25 0.37 -0.25 0.38 P-value .000* .000* .000* .000* .000* SITUATIONAL JUDGEMENT TEST OVERALL RESULTS The Situational Judgement Test (SJT) focuses on one factor of workforce readiness, conscientiousness. Consciousness is the tendency to be organized, responsible, and hardworking. Employees with low levels of this trait are less able to manage goals and time, work independently, or take responsibility and ownership of their work and decisions21 . Why does consciousness matter in the workplace? An employee with high levels of this trait are more able to self-manage, they have a strong work ethic, plan and maintain focus on a task. Overall, they are growth-minded and goal striving which is a benefit to any enterprise From a mean score22 of 40 across the eight SJT items, on average youth scored statistically different at the midline with a mean score of 76.6 percent correct compared to the baseline with a mean score of 74.8 percent correct (p<.000). While at the baseline a statistical difference existed by gender as female survey youth had a higher average score (75.3 percent) compared to males (74.6 percent), at the midline the average scores between males (76.4 percent) and females (76.7) was statistically similar. The same results could be observed by school 21 . A Pearson’s r product-moment correlation coefficient was computed to assess the relationship between the SJT and the Consciousness trait measure by the BFI. Pearson's correlation coefficient (r) is a measure of the strength of the association between the two variables. There was a positive correlation between these two variables both at the baseline (r=102, n=1,111, p<.001) and midline baseline (r=246, n=1,143, p<.000) As such, increases in the BFI consciousness correlate to increases in the SJT measure of workforce consciousness. 22 The SJT scoring is based on the absolute value of the sample in comparison to the ratings of a group of 10 personality experts, who scored the SJT items response options based on the effectiveness of each response to the situation and on the item’s ability to represent the facet. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 50 type. Whereas at baseline youth from CAP (75.8 percent) schools had a higher mean score to the Level 3 CRS youth (74.4 percent), at the midline average score increase to a statistically similar level for the Level 3 CRS youth (76.5 percent) and CAP youth (76.9 percent). Desegregation by regions also showed some variations. While the difference at midline for youth in Goma (p<.000), Bukavu (p<.008), Kalehe (p<.000) and Kabare (p<.000) were significantly different, changes in Walungu was statistically similar (p<.573). This is such as while on average gained 1-3 points difference from the baseline, Walungu scores were nearly unchanged. Work Readiness Indicator Results and Conclusions IYDA project has a target of 60 percent of youth that ‘improve’ one level or more on their level of work readiness at conclusion of programming (Indicator 3.1.2). Based on the paired survey youth, on average youth gained 2.6 points at the midline, with an average of 76.3 percent correct at the midline, compared to 73.7 percent at the baseline. Overall, the results of the matched pairs were slightly proportional to the overall independent sample results. As a measure of improved skills, 61.7 percent of youth improved at least one level on the work readiness assessment from baseline to endline. Assuming all else equals, comparison of the survey youth’ midline average score (76.6 percent) to the overall baseline score (74.8 percent), shows that on average youth gained 1.8 points. As a measure of improved skills, 60.3 percent of youth improved at least one level on the work readiness assessment at midline from the overall baseline means. While there was not a significant association between gender, school type, and age-band on the work readiness outcome 75.8% 77.1% 75.0% 76.1% 75.7% 75.4% 73.8% 76.6% 73.7% 76.8% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% Baseline (n=383) Midline (n=417) Baseline (n=250) Midline (n=252) Baseline (n=130) Midline (n=124) Baseline (n=193) Midline (n=165) Baseline (n=155) Midline (n=185) Goma Bukavu Walungu Kalehe Kabare Figure 30. DRC IYDA SJT Baseline and Midline Average Scores (N=1,111) USAID/DRC IYDA COHORT I MIDLINE EVALUATION 51 there was a significant association between region and work readiness outcome. This is the result of the variance in the average scores by region, which shows that Walungu had the lowest rate of improvement. n=1,143 Gender (Χ 2 (1)> = 2.375, p = 0.123) School Type (Χ 2 (1)> = 442, p = 0.520) Region (Χ 2 (4) = 18.246, p < .001) Age Band (Χ 2 (4)> = 6.396, p = 0.171). Tot al Fem me Hom me CRS CA P Gom a Buka vu Walun gu Kale he Kaba re 13 thru 15 16 thru 19 20 thru 24 25 thru 30 31 thru 35 Lack of progres s 262 192 311 143 147 110 67 67 63 185 174 58 25 12 454 37.9% 42.5% 40.4 % 38.3 % 35.3 % 43.7% 54.0% 40.6% 34.1% 40.9 % 41.0 % 31.7 % 42.4 % 48.0 % 39.7 % Improv ed 429 260 459 230 270 142 57 98 122 267 250 125 34 13 689 62.1% 57.5% 59.6 % 61.7 % 64.7 % 56.3% 46.0% 59.4 % 65.9% 59.1 % 59.0 % 68.3 % 57.6 % 52.0 % 60.3 % Total N 691 452 770 373 417 252 124 165 185 452 424 183 59 25 1143 Across the BFI and the SJT IYDA’s youth show significant gains. Inspection of the 95% confidence intervals around each mean indicated that there was a significant increase in average gain scores for youth overall on the BFI and the SJTs. While these results of the BFI are subject to self￾score bias—based on research showing the BFI factors are malleable throughout young adulthood23— and given the limitations of the SJTs, overall results from the baseline to the midline may suggest youth work ethic, teamwork, resilience, analytical thinking, emotional intelligence, and confidence have improved from the baseline. Still, there is room for progress especially given the variance of the results by region. 23 Roberts, B. W., Walton, K. E. & Viechtbauer, W. (2006). Patterns of mean‐level change in personality traits across the life course: A meta‐analysis of longitudinal studies. Psychological Bulletin, 132, 1–25. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 52 CONCLUSIONS This evaluation suggest that the IYDA project may positively influence participating youth in key ways. While claims to causality cannot be made, these evaluations do highlight clear areas in which the IYDA project seems to be supporting youth in increasing their resilience, skills, and perception of community and school safety. Within the context of conflict-affected eastern DRC, generally, IYDA youth show positive changes in perceptions of violence – away from supporting violence toward supporting peaceful means of conflict resolution – are promising. YOUTH’S PERCEPTIONS OF COMMUNITY IYDA youth demonstrate significant improvements in their perceptions of their communities with greater engagement in their community. Findings also demonstrate improved perceptions period further corroborate the youth's engagement in their communities and prosocial community support. Overall, 54.4 percent of youth improved at least one level on their perception of their community. RESILIENCE SKILLS IYDA youth rated themselves significantly and considerably higher at midline than baseline. While it isn’t possible to speak to these gains conclusively without a comparison group, these findings do highlight the potential for further integration of resilience skills into Life Skills training and youth-led activities. Overall 61.9 percent of youth improved at least one level on their perception of self￾efficacy; 52.5 percent of youth improved their perception of parental/caregiver support; 57.4 percent of youth improved their one level or more on their propensity to engage in gangs and armed group violence; 37.3 percent of youth improved at least one level on their perception of engaging in violent groups; and, 56.1 percent of youth improved at least one level on their perception of optimism and positive identity. School Safety IYDA youth showed a positive change in their perception of school safety at the midline compared to the endline. Still, the overall rating remained as fair with 53.8 percent of youth showing improved perception of school safety. While the perception of the dimension of school safety like child protective practices, discipline, and student-student relationships improved, results overall are not equivalent to the IYDA’s indicator target on school safety. As IYDA is expanding its safe school activities these result provides an opportunity to expand youth-focused activities such as school civic engagement activities, extracurricular clubs, and teacher￾student engagement discussion group —for future cohorts. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 53 LIFE, WORK READINESS, AND LEADERSHIP SKILLS IYDA youth improve their life skills, work readiness skills, and leadership skills over the course of their participation in the WRN training intervention. Overall, 60.3 percent of youth improved at least one level on the work readiness assessment at midline from the baseline. These results suggest that the IYDA activities that influence skill outcomes may have a positive effect on youth. This includes work ethic, teamwork, resilience, analytical thinking, emotional intelligence, and confidence. The findings of this report suggest the following baseline measures for IYDA’s performance monitoring indicators. Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores24 Percent Youth increasing scores from overall mean25 Q1. Do youth demonstrate improved in resilience to conflict and violence as a result of IDYA’s activities? 1 Percent of youth with increased self￾efficacy at the conclusion of programming 75% 33.7 of 48 36.626 of 48 .000 54.3% 61.9% 2 Percent of youth showing positive attitudes of tolerance and non￾violent means to address conflict drivers and conflict management. 75% 42.1 of 54 40.6 of 54 .000 35.1% 37.3% 3 Percent of target youth reporting a decreased propensity to engage in violence 75% 37.2 of 48 37.0 of 48 .831 42.6% 57.4% 24 Estimated from a sample of 94 matched participants within the sample from baseline to midline 25 tabulated by assigning the mean scores from the baseline to all participants and calculating the percent improve at the midline. 26 Revised from the previously reported as 33.9 due to a revised syntax error this is updated for the paired sample indicator from the previously reported 47.9%. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 54 Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores24 Percent Youth increasing scores from overall mean25 4 Percent of youth with increased optimism at the conclusion of programming 75% 26.3 of 36 26.7 .175 54.3% 56.1% Q2. Do youth demonstrate improved perception support and engagement in their community? 2.2 Percent of youth who believe that they are of value in society and their positive contributions are recognized and rewarded at the conclusion of programming 75% 48 of 72 64.0 of 72 .000 53.2 %27 54.4% 2.3.2 Percent of youth who report feeling comfortable asking their parents or guardians for help or advice during the past 30 days28 75% 46.2 of 60 45.9 of 60 .442 45.7% 52.5% Q3. Do youth demonstrate improved perception of school safety or learning spaces? 1.2.1 Percent of youth with improved perception of the safety of schools or learning spaces 70% 224 of 306 227 of 306 .001 48.9% 53.8% 27 While previously reported as 84.0% revised analysis of the paired sample revealed an error in syntax code which when revised resulted in a 53.2% outcome 28 While previously reported with two results this indicator measure was revised to report the entirety of prosocial parental/caregiver support. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 55 Evaluation Questions Indicator # Indicator Target Baseline Average Score Midline Average Score p-value of Significance of difference Percent Paired Youth increasing scores24 Percent Youth increasing scores from overall mean25 Q4. Do youth demonstrate improved work readiness, as a result of IYDA’s activities? 3.1.2 Percent of individuals with improved skills following completion of USG-assisted workforce development programs 60% NA NA .000 61.3% 60.3% Q5. Do youth participants demonstrate improved livelihoods as a result of IDYA’s activities?29 3.1 Number of individuals with new or better employment following completion of USG￾assisted workforce development programs 4,000 (~40 percent of 10,325 completers) NA NA NA NA NA 29 This question was not evaluated as part of the midline at the completion of IYDA’s training. It will be measured as part of the endline evaluation scheduled six months post training in Quarter 2 of FY20. USAID/DRC IYDA COHORT I MIDLINE EVALUATION 56 ANNEXES Resilience Spearman's rank‐order correlation N=1,110 Community perceptions Self￾Efficacy Pro￾social family support Engagement in Violence Tolerance Positive identity Extraversion Scale Score Agreeableness Scale Score conscientiousness Scale Score Emotional Stability Scale Score Openness Scale Score SJT Overall Community perceptions Spearman’s rho (rs) 1.000 .356** .344** -.032 .197** .383** .179** .424** .390** -.262** .320** .106** Sig. (1- tailed) .000 .000 .142 .000 .000 .000 .000 .000 .000 .000 .000 Self-Efficacy Spearman’s rho (rs) .356** 1.000 .314** .133** .304** .438** .265** .352** .368** -.300** .416** .221** Sig. (1- tailed) .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 Pro-social family support Spearman’s rho (rs) .344** .314** 1.000 .086** .174** .366** .195** .357** .311** -.225** .241** .028 Sig. (1- tailed) .000 .000 .002 .000 .000 .000 .000 .000 .000 .000 .179 Engagement in Violence Spearman’s rho (rs) -.032 .133** .086** 1.000 .255** .135** .087** .031 -.022 -.073** -.086** .107** Sig. (1- tailed) .142 .000 .002 .000 .000 .002 .150 .234 .007 .002 .000 Tolerance Spearman’s rho (rs) .197** .304** .174** .255** 1.000 .143** .116** .301** .294** -.241** .189** .218** Sig. (1- tailed) .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 57 Positive identity Spearman’s rho (rs) .383** .438** .366** .135** .143** 1.000 .198** .324** .330** -.277** .230** .055* Sig. (1- tailed) .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .034 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 58 Sample School Demographics GOMA Sample Baseline Midline Centres de rattrapages Scolaires (CRS) CRS ANNA MICHELI 28 20 CRS APROFIM 30 26 CRS BADIRIKA 30 0 CRS BETHEL 31 37 CRS CARE 30 15 CRS CEFAJED 30 3 CRS CENED 30 0 CRS KABA 0 15 CRS KANYARUCHINYA 0 10 CRS NAZAREHEEN 0 20 CRS REHEMA 0 22 CRS RUDO 0 11 CRS SHALOM 0 39 CAP ATELIER NAZARETH 0 5 Total 209 223 Centres d’apprentissage professionnel (CAP) CRS COV 36 0 CRS KANYARUCHINYA 36 0 CRS LA VICTOIRE 36 0 CRS REHEMA 28 0 CRS RUDO 38 4 CAP APROJED 0 35 CAP ATELIER NAZARETH 0 28 CAP ETN 0 35 CAP HOLD 0 31 CAP KARISIMBI 0 17 CAP MUHUNGANO 0 11 CAP PEACE LIVES 0 32 Total 174 194 BUKAVU Sample Baseline Midline Centres de rattrapages Scolaires (CRS) CRS AMALDEFEA 30 19 CRS AMUKENI 28 9 CRS ANNONCIATA 29 36 CRS BAGIRA 1 30 21 CRS BAGIRA 2 31 24 CRS CAHI 0 8 CRS CREAV 0 21 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 59 CRS MWANGA 0 12 CRS NYOTA 0 9 CRS UMOJA 0 22 CAP BAGIRA DIVAS 0 3 CFPO BAGIRA 0 3 Total 148 187 Centres d’apprentissage professionnel (CAP) CRS CAHI 40 0 CRS CREAV 25 0 CRS KADUTU FOYER 11 0 CRS MUSHEKERE 26 0 CAP ANNONCIATA 0 13 CAP BAGIRA DIVAS 0 21 CAP KADUTU FOYER 0 5 CAP KAWA 1 0 7 CAP UMOJA 0 5 CFPO BAGIRA 0 14 Total 102 65 WALUNGU Baselinne Midline Baseline Midline Centres de rattrapages Scolaires (CRS) CRS IKOMA 13 10 CRS KAHEMBARI 30 12 CRS LURHALA 0 11 CRS MADAKA 0 14 CRS MUSHINGA 17 8 Total 60 55 CRS LURHALA 34 2 CRS MADAKA 36 0 CAP IKOMA 0 45 CFPO WALUNGU 0 22 Total 70 69 KALEHE Baseline Midline Baseline Midline Centres de rattrapages Scolaires (CRS) CRS AMKENI MINOVA 16 23 CRS BUSHUSHU 37 26 CRS CIBANDA 20 23 CRS KASHEKE 17 7 CRS MINOVA 23 12 CRS MUHONGOZA 0 20 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 60 CRS MURHAMBI 0 10 CRS RAMBIRA 16 6 CRS TCHOFI 0 15 CRS UMOJA MINOVA 0 9 Total 129 151 CRS MUHONGOZA 28 0 CRS MURHAMBI 36 0 CAP CFPO KASHEKE 0 14 Total 64 14 KABARE Baseline Midline Baseline Midline Centres de rattrapages Scolaires (CRS) CRS AMANI 32 38 CRS BARAKA 19 9 CRS BENEMUDAKA 42 14 CRS KABINDI 19 11 CRS KATANA CENTRE 0 10 CRS LA MERVEILLE 11 26 CRS SAINT JEAN PAUL 2 0 33 CRS UJIJUWE 0 13 Total 123 154 Centres d’apprentissage professionnel (CAP) CRS AMANI 0 6 CRS KAHAYA 17 0 CRS KATANA CENTRE 15 0 CRS SAINT JEAN PAUL 2 0 8 CAP LA MERVEILLE 0 6 CAP SAINT JEAN PAUL 2 0 11 Total 32 31 USAID/DRC IYDA COHORT I MIDLINE EVALUATION 61 Data Collection Tools RDC ADIJ Etude Baseline Nov-Dec '18_2019.doc