USA Group 2 - Case Study 1 (Assessment)
Leah Bellow-Handelman lrb2144
Christine Heckman ceh2161

Unit of measurement
Since three weeks have passed since the disaster, our assumption is that immediate needs such as food and water have been addressed. Therefore, it seems like a sensible time to opt for a deeper assessment/analysis, which would best be achieved by using a household level of measurement.

Due to limited experience/training, we would send the social workers out in teams of two (ideally one male and one female per team) so they could monitor each other’s work, perhaps gain access to different demographics of people (in some cultures it may not be appropriate for a man to ask certain questions directly to a woman, for example), discuss and collaborate on any problems that might arise, and make sure they are sticking to the agreed up on methodology.

Dividing the total number of affected people (200,000) by 150 camps gives us approximately 1330 people per camp. Assuming each household contains an average of five people, that means 165 households per camp. With the two-person assessment teams already discussed above, assessing all 150 camps, plus villages, is unfeasible. Therefore, we would aim to survey 20 camps and 5 villages total (4 camps and 1 village per team).

We would perform purposive sampling based on the locations from which the displaced people had originated. Presumably, since the event was a natural disaster, various geographic areas were affected differently depending on their proximity to the disaster. For example, some people’s homes were likely destroyed while others may have been evacuated/fled temporarily. These differences in circumstances would likely create different needs and perceptions that we’d want to account for.

As mentioned above, our sample would come primarily from camps, and from a few villages. Those in villages will not have been accounted for through a registration process and are likely staying with people they know. In general we would assume these individuals have greater access to resources because of preexisting circumstances (family or friends in unaffected areas of the country, etc). However, we would not want to ignore these groups completely because they will still have child protection concerns that are distinct from those living in camps.

We would begin by identifying key informants in each camp, and for this component we would use convenience sampling, identifying relevant influential community members who are readily available. From these interviews, the desk review, and the camp registration records, we would determine key groups to target for the rest of the assessment. We would base this break-down on information regarding geographic origin, ethnic/cultural differences, and composition of the household (female-headed, child-headed, etc).

Each assessment team would be responsible for 30 households at each of the 4 camps and 1 village assigned to it. These households would be chosen through purposive sampling--identifying the key categories raised in the KIIs and then selecting randomly within each identified category. From the 30 households, we would choose one representative to take part in the PRM. We would perform 3 PRMs per camp--one of adolescent boys (group of 10 per camp/village), one of adolescent girls (group of 10 per camp/village), and one of mothers (group of 10 per camp/village).

Due to the limited training of the social workers on staff, we would not include young children in the PRMs, for risk of further re-traumatizing them. Instead we would observe them in more informal settings and rely on the mothers and older brothers and sisters to identify issues affecting the younger children.

Key questions we would look to address include:

What are the existing child protection resources availability within each camp? What are the primary needs that are still unaddressed regarding Child Protection within the camps? How can humanitarian actors best fill these gaps?

During the three weeks leading up to the beginning of our Child Protection assessment, staff at headquarters would perform a thorough desk review to help understand the history, culture, and other specifics of the location in question.

Upon arrival, we would use a combination of Key Informant Interviews (KIIs) and Participative Ranking Methodology (PRM). KIIs will target adult community members who have relevant experience or knowledge of Child Protection issues. They would be individuals who understand the community dynamics and work closely with children, including teachers, community center employees, and religious leaders/tribal elders. It is important that these interviews be conducted with both male and female KIs of various ages. We would also conduct KIIs with emergency response health workers, who would have been tracking child protection issues from the onset of the emergency, based on recording procedures established in the preparedness phase. CPRA Guidelines recommend that a minimum of 3 KIIs be conducted at a given site. We would also perform observations during more informal gatherings that take place as part of people’s daily routine.

As stated in the Sampling section, we would use the KII as an initial step to identify which groups to target as participants in the PRMs, the issues that should be addressed/explored, and the cultural issues and communal dynamics to look out for. For the PRMs, we would prioritize adolescents (separate groups of boys and girls) and mothers as participants. This process allows participants to identify child protection issues with which they are concerned. Once issues are identified, participants participate in a ranking process, allowing them to prioritize the most pressing issues and concerns related to child protection. This method provides both quantitative and qualitative data that can be interpreted to guide development of appropriate interventions.

Our strategy is to send each team of two social workers out with a supervisor who has expertise in Child Protection and assessment methodology.

According to CPRA guidelines, the best method to collect data is to focus on and gather data from an entire region at one time. If feasible, it would be ideal to assign the each of the five “rounds” based on geography, with the five teams all beginning with camps that are relatively close to each other and then move to another set of camps as a group. Data entry can be performed as a team at the end of each working day among individual teams and any collection/analysis issues can be addressed on the spot. “Data cleaning” occurs during the daily debrief sessions between the social workers and supervisors.

Since the team as a whole would be travelling together, every few days they would have the opportunity to deal with common issues/challenges facing all of them. This strategy may take longer than sending teams out to cover assigned regions, but it would allow for more coordinated data collection and transmission of information for analysis. It would also help ensure validity because everyone would be working on the same components of the project at the same time and checking in with each other regularly.

Triangulation of the data collection through the three methodologies (KIIs, PRMs and desk reviews) would also help ensure validity.

Coordination depends on collective effort of Child Protection actors to ensure all available resources are utilized, and increase legitimacy and validity of data. The ten social workers would be responsible for collecting data; their work would be overseen and verified by a management structure, who would also be responsible for coordinating with other agencies. The role of supervisors (5 total, one per team) would be to create and coordinate data collection plans, coordinate the social workers (“assessors”), and provide technical and logistical support. Data collection can be done by hand, but it is preferable to do so in a computer database. Ideally, data collection and data entry are done simultaneously, to allow for quick data analysis. Depending on the number of KIIs and PRMs conducted, the site supervisor would collect data from the social workers in daily debriefing sessions to be compiled into a site report. This site report would later be sent for analysis. In daily debriefing sessions, KIIs and PRMs are collected, and supervisors and social workers would verify data by checking for errors and addressing challenges in collecting data. A daily report would be created following these sessions, which would be shared with the wider team of social workers in between sites.

All of the raw data collected by the 10 social workers must be compiled, entered into a system and analyzed to identify trends. The management team and supervisors who worked closely with the assessors would be in charge of data analysis. For PRM, we would employ quantitative methods to analyze the data gathered, including frequency with which priorities were mentioned and how they were ranked. We would then review the data with a qualitative lens to account for comments made by individual participants. To analyze the KIIs, we would use qualitative approaches to identify trends in issues raised and child protection priorities. Another method is to disaggregate (break up) data by specific characteristics of the data source (male, female). Once data is analyzed, a final report must be shared with all relevant actors in order to operationalize relevant interventions.

Time line
Day 1 (post-disaster) - 10 social workers identified and hired and split into 5 teams of 2, with one supervisor/technical expert assigned to each team.
Day 2 - 6 - Commence 5-day training (we feel training can take place within 5 days because social workers are community members who speak the local language, and have some institutional memory and understanding of the disaster).
Day 7 - 21 (15 days; 3 days at each round of camps/villages) - Data collection - Key Informant Interviews and Participative Ranking Methodology (first round of KIIs will help to identify relevant participants for PRMs). Each team of two social workers will be responsible for visiting 4 camps and 1 village, spending 3 days at each. During this time, data will be collected and cleaned by supervisors in daily debriefing sessions.
Day 22 - 25 - Data from all 5 groups is compiled, cleaned and analyzed.
Day 26 - 28 - Produce report, publicize Child Protection findings and make relevant recommendations for programming and interventions.