The International Federation of Red Cross and Red Crescent Societies (IFRC) is a global humanitarian organization that works to prevent and alleviate human suffering. As a way to digitise and make processes more efficient, the IFRC has developed IFRC GO, a platform to connect information on emergency needs with the right response.

Sampling is one of the key methods for the IFRC (or any other humanitarian organisation) to understand the views of a certain population. Sampling is meant to be a scientific approach, however, as also outlined in our Abstract, country-level teams do not always have the expertise, time and/or resources to conduct proper sampling. As a consequence, sampling - a method that is thought of as a rigorous approach - yields statistically incorrect results that should not be trusted and can not be generalised.

Our Community Sampling Tool will serve as an extension to the aforementioned IFRC GO platform with the goal of encouraging correct sampling.

The general goal is to create an intuitive tool that guides users through the sampling process. Based on our research, there are several sampling approaches, such as simple random, time-location or cluster, that are used on a regular basis by IFRC staff.

However, there's no one-solution-fits-all approach. Every situation and survey has unique conditions and they must all be taken into account when sampling. Our design follows a funnel-questioning model. From the ideal sample population to the specifics of the sample, we take as much information as we can to determine the most appropriate method to sample.

Our main way of collecting requirements was meeting with several IFRC staff members including Rachel Yales, Alex Howes, Eero Sario and Vincent Annoni.

At the beginning of our project, we focused our weekly meetings on requirements gathering. In-between meetings our team also met to discuss how feasible and realistic certain requests were to then funnel this feedback back into the upcoming meetings. The IFRC team was just as responsive and enthusiastic as us, so we were able to establish a quick feedback loop and iterate quickly.

In the context of developing an IFRC sampling tool, a persona could represents a specific user group. Personas are created based on the interview and analysis of the questionare and are used as a tool to help us better understand users' needs, behaviors, and preferences. By creating a detailed persona, the we were able to developed a sampling tool that is tailored to the specific needs and interests of the users.

• Persona 1
• Persona 2

Surge Team Delegate (Smith) :

Smith has been working for the British IFRC global surge team for years. The most recent emergency that he was deployed in was the Haiti earthquake in 2023. His role was to work with Haiti Red Cross teams, local authorities and other aid organisations to mobilise resources, arrange for funds and organise personnel.(British Red Cross,2022) In order to plan all of these efficiently, the first thing he did was to survey and assess the need and load in the information to a sampling tool. Based on the tool’s output, Smith can allocate the corresponding resources. The tool provides the theoretical correct results but usually these results are not practically applicable. For example, Smith may find that the tool suggest more number of people or vehicles than the team has. It is very frustrating for Smith when he finds out that the data he has collected for months does not have a good tool to give out proper and realistic results.

Pakistan Red Crescent Society Staff (Mirral) :

     Miraal works for Pakistan RC society and she has limited knowledge in programming and statistics. She is in charge of the investigation at the local national society branch during a disaster. (M.Amin,No date)Since she rarely runs into large-scale disasters, she is not very familiar with the procedures and formalities. To improve her knowledge- Miraal uses the sampling tool provided by the global IFRC team to accomplish her task. However, the tool supports English only and Miraal has to use a translation software from time to time. The definitions of the terminologies are also too complex for her to understand. In order to finish the task on time Miraal has to work under high pressure everyday.

Having learnt more about the different user groups and also taking into account the API we developed, we have created a use case diagram. This diagram shows the different actors and their interactions with the tool and the API.

To summarise all the information we have gathered in the 'Client Requirements', 'User Interviews' and 'Personas' sections, we put together the MoSCoW list for our project. The MoSCoW list finalised the requirements that were used to guide the development of the sampling tool.

Functional list

Non-functional list