Questions and Answers

1. What was the major objective of your project and what was your plan to achieve it? 

The major objective of my project was to design and build a new mobile app that solves two problems: first, to save the time that teachers use to manually take attendance in each and every class, and second, reduce the time school nurses spend on manually tracing close contacts of infected students, to help contain the spread of coronavirus.

            To achieve this objective, I planned to build a mobile app, “Auto Attendance”, that automatically takes attendance and performs rapid contact tracing.

       a. Was that goal the result of any specific situation, experience, or problem you encountered?  

As a student at my school, I noticed two problems in our learning environment: 1) How teachers would take a considerable amount of class time to take attendance, and 2) During the COVID surge, it often took days for the school to identify close contacts when a fellow student was tested positive, which caused a major concern for parents.

First, I sought to design a mobile app that could automatically take attendance for teachers. From elementary to middle to high school, I realized how teachers would take unnecessary amounts of class time to track down absent students. At the beginning of each class, about 5-7 minutes are used for manual attendance-taking. We would easily lose 40 minutes of learning time on a day with 8 class periods. By having this mobile app automatically taking attendance, students and teachers can use more time for learning. Second, when we returned to in-person learning last year, there was a surge in COVID infections at school. Contact tracing was not timely because it was done manually by school nurses going through attendance papers. This would allow for the virus to continue spreading during contact tracing.

Along the way of creating my Auto Attendance mobile app, I realized I could use the attendance app to perform contact tracing. I added a mechanism on my app to search for users by their attendance data, making contact tracing efficient and rapid. This attendance app also allows for parents and teachers to track their students for safety reasons, such as in common areas, school gatherings, and school buses. These places are often the blind spots in contact tracing as well.

       b. Were you trying to solve a problem, answer a question, or test a hypothesis?

I was trying to solve a problem.

I looked for a solution to the problem of slow and inefficient contact tracing and attendance-taking in school environments, which were concerns for students and teachers. The goal was to develop a mobile app that can automate attendance taking and contact tracing. It should make both processes faster and more efficient.

2. What were the major tasks you had to perform in order to complete your project?

1. I familiarized myself with MIT App Inventor, an online platform that allows for developing mobile apps.

2. I had to design the logic flow of the app. It needed to have the features of taking attendance, store and retrieve data from a database, and find a list of close contacts of a given student. The app needs to interface with students, teachers, and administrators. The features, data and users are connected and the app had to make them work together. I found a way to break down the logic flow into 3 main components and designed 5 screens to connect these components. The details of the main components are described in my technical report.

2a. I first built an Attendance Taking component to record student names, classroom locations and times of classes students attend.

2b. I implemented a personal identification component that uses QR Code identification to identify students.

2c. I built a Class Information Collection component that collects the classroom section, class period and teacher of a class.

2d. I built a Contact Tracing component that returns the class information of classes attended, and close contacts of an infected student.

2e. I set up a database to store attendance records and class information.

3. I generated test cases and tested the speed and accuracy of the app.

            3a. I set up test cases with a group of students, each attending multiple classes with different class periods and classroom locations. I also selected multiple positive cases for contact tracing. I ran the app and added class rosters for each class.

            3b. Using the test cases, I ran the app to take student attendance and compared the results with the actual attendance.

            3c. I ran the app to test the data storage to a cloud database and retrieved a list of close contacts from the database. I then compared the list retrieved from the database to the actual close contacts.

            3d. I measured the accuracy and speed of the Contact Tracing component.

4. I had a meeting with my school principal and school nurses and introduced my work to them as well as a few teachers. I collected their feedback on the app and discussed practical constraints and bottlenecks in changing the contact tracing protocol at my school.

       a. For teams, describe what each member worked on.

This was not a team project.

3. What is new or novel about your project?

To my knowledge, at the time of building this app, there were no known apps for automatic attendance-taking and contact tracing in school settings. My project built a new app for this purpose. This app allows students to sign themselves into class automatically, and does contact tracing instantly using student attendance records.

       a. Is there some aspect of your project's objective, or how you achieved it that you haven't done before?

This is the first time I built an app for attendance-taking and contact tracing. This is also my first time using a Barcode scanner to build a personal identification component, and using a cloud database to store and retrieve data. In the initial design, I also used the GPS Location sensor that I have not used before.

       b. Is your project's objective, or the way you implemented it, different from anything you have seen?

Most apps are not designed for classroom use, but my app is built for school settings, where students are scheduled to move from class to class over multiple class periods. I incorporated two connected features, attendance-taking and contact tracing, into one app. This app was developed using MIT App Inventor, and is a multiplatform app.

       c. If you believe your work to be unique in some way, what research have you done to confirm that it is?

I have conducted research and literature review. From recent scientific research, I found that on average, contact tracing takes up to 6 days to perform manually. Scientific studies also suggest that by reducing the time it takes to contact-trace by just 3 days, the transmission rate of the virus more than halves. Additional modelling and simulation by Almagor and Picascia concluded that if 80% of the population were to use a contact tracing app, the percentage of the population infected would decrease from 45 to 15%. With 40-60% of the population using the contact tracing app, overall infections would decrease to 22-27%, and cases at the peak of the epidemic would be reduced by 70-85%. The app I developed makes contact tracing instant, which would help to contain virus spread.

To my knowledge, there is no app for class attendance-taking. This app is unique in automating attendance-taking and saving time for teachers and students.

4. What was the most challenging part of completing your project?

The most challenging part of this project was to develop the complex internal logic design of the app. The logic design needed to make the app instant and easy to use for students to record their attendance and for school nurses to do contact tracing. The app needed to interact with different types of users, the cloud database, and a barcode scanner using the camera sensor. It also needed to support different features of collecting, storing and retrieving attendance records, and tracing close contacts. All the parts of the app were connected and needed to work together.

      a. What problems did you encounter, and how did you overcome them?

As stated above, the app interfaces with different users and integrates different components. The first problem was to make sure that each part of the app works well with the others in the logic design, given the constraints of the coding platform.

The second problem was to find a way for person identification when students sign in to the app. There are many ways to identify people, including face recognition, voice recognition, and QR Code recognition. I compared these options for their accuracy and privacy constraints, and decided to use a QR Code scanner in my app and also have the option for students to enter names.

The third problem was how to get the location of a student for contact tracing. Originally, the app used the GPS sensor on the mobile device to record the location of student users. However, after testing this option I quickly realized the limitations of this method. The GPS sensor was inaccurate by a few meters. This app would not be feasible in school environments, since the entire student population needed to run the app all the time for it to work effectively. In addition, students might not give permission for the app to constantly record their location. I had to change the design of the app and added a classroom information setup component to record the classroom as the student’s location. In the new design, the students did not have to install the app on their phones. They only needed to scan a personal QR Code upon entering the classroom.

The fourth problem was encountered when testing the attendance-taking component. I realized the app needed to cover different situations that could happen during student sign-in. For example, the student may be absent or sign in multiple times to the same classroom.

 

      b. What did you learn from overcoming these problems?

I learned that complex logic systems can be constructed more efficiently by breaking the system down into different parts. A helpful approach would be to build and test each part of the app separately, and then put them together. I also learned there are often multiple solutions to each technical problem. It’s a good idea to pick a solution with a good balance of accuracy, ease to use, and consideration of practical constraints. Additionally, when testing solutions, be sure to test for different ways users interact with the system, including corner cases, such as multiple sign-ins and absent students in my app.

5. If you were going to do this project again, are there any things you would you do differently the next time?

If I were to do this project again, I would add an alert and notification component. This component uses the user’s contact address to notify them when they become a close contact.

6. Did working on this project give you any ideas for other projects? 

Yes. I would like to extend this project and build an app for tracking student riders on school buses. Currently, there is no monitoring or tracking technology for school transportation. Students can miss their bus or get off at the wrong stop. It frequently happens for elementary school students. Parents are not notified about the whereabouts of their child. An app for tracking bus riders would be very helpful.

7. How did COVID-19 affect the completion of your project?

The COVID-19 pandemic urged me to expand on my project. This project originally aimed to design a mobile app that can automate attendance taking and save teachers’ time and effort. However, the COVID-19 pandemic made me realize that the information collected in the attendance app could also be used to rapidly and effectively perform contact tracing. I talked to my school nurses and found out that manual contact tracing took copious amounts of time and effort, and put a huge strain on them. Because the process took too long, parents and students were worried about the virus spread at school. I expanded Auto Attendance to include features of rapid contact tracing. The need to reduce the development time also led me to select MIT App Inventor as my coding platform.