Scholarship system design part I - Information architecture
Problem description
The Costa Rica Institute of Technology (TEC) needed to create a system that allows to request and manage the student scholarships. The process begins when the Social Work Department of the university receives all the applications presented in paper, which can be around 1600 per semester (this number was taken from the 2017 statistics and increases every year). The social workers have to analyze all those applications in a very short period of time and also have to interview all the students who are expecting a scholarship, besides that, they also have to process all the students who already had a scholarship from the past semester
Once the scholarships are given, the Social Work Department has to manage all the data that comes from the process, they have to be able to inactivate them in any moment, modify some features, treat special cases that will come around and a lot of other tasks. It’s a very detailed process with a lot of flows not automatized yet, so, they manage the information in huge excel sheets that in the near future won’t be enough because of the population increase. This is a huge risk not only because of the amount of information, but because of the wide range of human error they have, as they manage over $9.000.000,00 per year.
The design solution:
This system is huge, I worked at it for three years (2014-2017) so I divided the design phases and will explain the way I took in order to get a possible architecture at the very beginning of the process.
I started by researching with the client the types of users that will be involved in the system, so I can separate and classify all the functionalities. There are four user profiles and they support different types of information as academic data, geographical data, personal information, parameters and configurations, and more.
Despite the size of the system, it is not an island, it works linked to other systems, some of them belongs to departments inside the university and some of them belongs to the government, this is important to identify as a designer, so you get to understand where the information comes from and where it is going to be sent.
Once the user types where identified, its time to create a list of requirements for each one, I got some of them from the client and others were identified through focus group I lead with the possible users. This part is very important in order to understand user needs that many times the stakeholders don’t see. Based on this I created an alpha architecture with the main concepts that represents the navigation for each user profile and then test them with the card sorting methodology. This test allows us to analyze if the suggested concepts are indeed understood by the users and also let us know the groups of information that makes sense.
Card sorting results are summarized in a dendogram that lets us see the user's behavior at the test and help us understand how to organize the groups of information, here we can tell if the designer's architecture is accurate or not. The testing moment is also an information source, because of all the comments and reactions users have, mostly related to the meaning of words. At the end of this process I got a validated alpha architecture that has four navigation levels, but of course it may change a little bit with the rest of the design process and testing.
The main importance of this part of the project is how dealing with huge systems makes you develop a great perspective of the problem, you get to understand the logic of the system, data bases and tables in order to organize data in a sustainable way. It also prepared me to confront any other kind of system, since this is the biggest one I have seen till now.
The following images represent the alpha architecture design for student and employee profiles. Since this system ain’t finished yet, the structure may change, and for sure it will expand. You can consult the next part of this project in the next post: Scholarship system design part II - usability testing.