47528-126324-1-PB
47528-126324-1-PB
47528-126324-1-PB
Abstract.
Purpose: The teaching matrix often relies on conventional methods involving explanations on a board. It makes
students easily bored and relatively focused for only a short period. Integrating multimedia features such as text, colors,
and sound might enhance students' comfort and engagement in the learning matrix. In this study, we discuss how to
design low-fidelity and high-fidelity prototypes to enhance the comfort of the learning matrix and give a solution to
manage course and communication between teachers and students.
Methods: We employed the User-Centered Design method, focusing on users as the central design consideration. The
User-Centered Design process involves the analysis of user needs and context, followed by requirements identification,
design solutions, and solution evaluation.
Results: Based on the testing results, we scored 86.25%, categorizing it as "acceptable" with a grade of A and an
adjective rating of "excellent." Meanwhile, the usability testing resulted in a score of 85.6%.
Novelty: The main contribution to this research is proposing a new design for the UI/UX of the matrix learning
application. The generated design can be recommended for implementing a matrix learning application.
This work is licensed under a Creative Commons Attribution 4.0 International License.
INTRODUCTION
Matrix and vector are the subjects covered in Computer Engineering or Computer Science. In the field of
the study program in Computer Science, these topics are introduced in courses like Linear Algebra,
Numerical Methods, Graphics and Multimedia, and various fields of specialization like Computer Graphics,
Image Processing, Data Science, and more. The problem with these courses is that students tend to get
bored and need more enthusiasm when the lessons are delivered in conventional teaching methods. One of
the advancements in educational technology is the use of games as an alternative solution [1].
At this time is an era of technological development that can improve the quality of human life technologies
[2], one of which is in the field of software development.. The use of technology must be carried out
optimally to help processes and activities [3]. The development of software applications involves several
stages, including analysis, design, implementation, and testing [4]. The analysis and design phases are
crucial to the success of application development. Design comprises several stages, namely architecture
design, database design, and interface design. Nowadays, the field of interface design is closely related to
user experience (UX) studies. UI/UX design involves various disciplines that bridge academia and industry
[5]. The measurement of UX maturity levels in several studies has been researched in [6]. These studies
evaluate how UX practitioners work with clients with low UX maturity and then attempt to enhance their
clients' UX maturity in an agile environment. The User Experience (UX) method [7] involved 13 UX
professionals from various industries and two countries working with agile. They employed a qualitative
approach and thematic analysis to address research questions. The results indicated that professionals
frequently used prototyping, user testing, user journeys, and workshops in their UX work.
UI/UX development has also been carried out in research for application designs [8]. Furthermore, in the
study [9], they conducted research on the Child Computer Interaction UX Playbook, which is engaging and
*
Corresponding author.
Email addresses: egia@dsn.dinus.ac.id (Subhiyakto)
DOI: 10.15294/sji.v10i4.47528
Many researchers have conducted several studies on design methods and their evaluations. For instance, in
the research conducted in [25], User-Centered Design and usability testing methods were employed
involving 23 participants to examine user perspectives on blockchain technology. In another study, the
human-centered design method [26], with a broader scope than UCD, is employed. This research discusses
advanced services for both industry and academia. Lean UX is employed in the study [27] to integrate every
process at each stage to enhance startup valuations and attract additional investor funds. On the other hand,
a study [28] discussed serious games as a potential learning tool. In this research, we aim to design a
learning game application related to matrices and vector spaces. Providing a learning game will enhance
the connection between students and the course, fostering a more substantial engagement and interest.
METHODS
The research method used in this study is the user-centered design method, as shown in Figure 1 below.
Based on Figure 1, there are four stages in the user-centered design method. Choosing this method can help
users achieve results that are easily understood and maximally usable. UCD is a method that prioritizes
users as the primary focus in the design of the built application.
The application design in this research begins with conducting user research to recognize user
characteristics and develop a deep understanding of the users. The design of mockup user interfaces, based
on user experience analysis using the UCD method, focuses on introducing the characteristics of students,
which can be done through observation, interviews, and questionnaires. Observation is done by directly
After recognizing the user characteristics, we proceed with collecting and aligning information for the
design that will be developed. We address user issues while considering the identified user characteristics
from the collected data. The collection and alignment of the design are depicted as shown in Figure 2.
Figure 2 illustrates the alignment between technical and user requirements, which can be achieved by
exploring and finding common ground. This way, the designed project or application can meet the needs
of both designers and users by the technical aspects outlined in the methodology employed in this research.
The user characteristics and the design alignment from user to designer have been identified. We generate
an idea as a solution for the design to be developed. In this research, we employ an iterative design process
that can be used to manage risks in user interface design. This iterative design process refines the design
through repeated cycles in the design journey.
In Figure 3, the design represents the initial step to visualize imagination or ideas. Next, an implementation
is carried out as a physical representation of the designed outcome, such as mockups and prototypes.
Subsequently, an evaluation is conducted to test the UI/UX outcomes against user characteristics. If not
aligned, the process is iterated continuously until a usable interface result is achieved. In this design
solution, we create a low-fidelity representation in the form of design sketches on paper, which serve as a
basis for crafting mockups by exploring and visualizing those ideas. Following that, we generate a high-
fidelity prototype that closely resembles the final product. This prototype includes colors, transitions,
illustrations, and more.
The low-fidelity and high-fidelity design phases have been completed in the previous stages. Next, a
reevaluation is necessary to compare the prototype with the gathered requirements. This testing is
conducted using the System Usability Scale (SUS). The System Usability Scale (SUS) is a widely used
questionnaire-based method for assessing the perceived usability of a system or software. It comprises a
10-item questionnaire that participants respond to using a Likert scale. Several relevant studies that have
employed the SUS include [29], which has discussed evaluating the usefulness of Microsoft Teams as an
online learning platform during COVID-19. On the other hand, [30] discussed the concept of "questionnaire
experience" (QX), introducing an innovative approach to assess a novel instrument.
The System Usability Scale (SUS) consists of 10 questions, where participants are provided with a 1-5 scale
to respond based on their level of agreement with each statement regarding the tested product or feature. In
calculating the SUS score, a 5-point Likert scale is employed, following specific rules: for odd-numbered
questions, the score answered in the questionnaire is subtracted by 1 (formula 1). For even-numbered
questions, their value is subtracted from 5 (formula 2). Then, the values of both even and odd-numbered
questions are added. Subsequently, the sum is multiplied by 2.5 (formula 3).
The questionnaire's score range is 0-100, with the standard average SUS score being 68. Scores above 68
indicate good usability, while scores below 68 suggest a need for improvement in the usability of the
designed product or feature. The following is a general guideline for interpreting the SUS Score scale,
which will be presented in Table 1.
We conducted data collection and design alignment to discover commonalities between user characteristics
and the designed solution. It is done to ensure compatibility with the technical aspects outlined in the
employed methodology and to align with the needs of both designers and users. Table 2 presents the
outcomes of the data collection and design alignment by user characteristics.
The process of design solutions involves implementing user needs and issues. In the design solution phase,
we create both low-fidelity and high-fidelity designs.
Figure 5 depicts the low-fidelity main page, while Figure 6 illustrates the high-fidelity main page. The main
page includes buttons for play, exit, survey, bug report, music, shop, unlock all worlds, and reset data.
Figure 7 and Figure 8 represent the low and high-fidelity designs for the learning material page. On this
page, there is a "back" button to return to the main menu, as well as buttons for different topics related to
matrix learning, such as matrix definitions, types of matrices, and matrix operations.
Figure 9 and Figure 10 depict the design for the content page that contains information about matrix
concepts. On this page, there are a "back" button and content scrolling options.
Figure 11 and Figure 12 provide a comparison between low and high-fidelity designs for the level selection
page. On this page, levels are unlocked as the player completes the previous ones. Each level comes with
its difficulty level. The page also includes a "back" button to return to the world selection.
Figure 13 and Figure 14 illustrate the low-fidelity and high-fidelity designs for the in-game page interface.
In this game, there is a character that will move to the right and answer questions when encountering a
question mark symbol. There are obstacles and aids in the game, along with arrow directions, to determine
the character's movement in the specified direction.
Figure 15 and Figure 16 represent the design of a question page, which displays questions related to matrix
material. Each question includes points that will increase if the answer is correct, and there is also a
countdown timer to answer the question.
The next step is to conduct design solution testing involving relevant respondents. We use a Likert scale
ranging from 1 to 5, in which 1 represents "very inefficient," 2 is "inefficient," 3 is "fairly efficient," 4 is
"efficient," and 5 indicates "very efficient" responses. We present ten pairs of opposing attribute variables.
These ten attribute variables cover efficiency, effectiveness, difficulty level, learning capability, tidiness,
clarity, satisfaction level, accessibility, error anticipation, and expectations. These variable pairs are
expected to represent the user's perceptions. We divided the testing into two parts: persona testing and
usability testing. Table 3 displays the results of persona testing from the user's perspective.
The user persona testing yielded an average score of 86.25. Therefore, the design solution can be
categorized as acceptable with an "A" grade, and the adjective rating falls within the "excellent" range. The
next stage involved usability measurements, encompassing effectiveness, efficiency, and satisfaction.
Figure 17 illustrates the usability testing results, covering these three aspects.
The figure demonstrates the analysis outcomes of testing the prototype design for the matrix learning game
application, with an efficiency score of 82%, effectiveness at 86%, and a satisfaction rating of 89%.
Table 4 illustrates a comparison of usability testing results for several application designs. In the study
conducted by [31], the focus was on the analysis and UI/UX design for a local medicine and medication
mobile application using the task-centered design process. The usability testing yielded a result of 75.75%.
The results of another study [32] discussing the design of a prototype for a mobile community based at the
Catholic Church of Mary Mother of Carmel in Jakarta yielded a usability result of 83.11%. The testing
results in this study, however, achieved a higher outcome, precisely 85.6%. Despite the [33] research
obtaining a User Interface effectiveness of 100% for all tasks, improving usability scores makes it a
preferable outcome. Based on the average usability score of 85.6%, the design solution created can be
recommended as the UI/UX design for the matrix learning application. In a study by [34], the focus was on
aesthetic semantics, in which user experience (UX) designers often prioritize beauty over the intended
impact on end users. This inverted priority order contradicts the sequential UX design process, leading to
unexpected perceptions and responses from end users. The study involved 1,782 participants worldwide,
evaluating affective user responses to 43 aesthetic atoms using 153,252 data points presented as Affect
Ratings (ARs).
CONCLUSION
The UI/UX design for the matrix learning application has been created as the result of the research. The
user-centered design method was systematically followed, from analyzing user needs and context to
requirements identification, design solutions, and solution evaluation. We have developed both low-fidelity
and high-fidelity designs. These designs were subjected to user evaluations through persona testing and
usability testing. The persona testing yielded a score of 86.25%, classifying it as "acceptable" with a grade
of A and an adjective rating of "excellent." On the other hand, the usability testing resulted in a score of
85.6%. It might indicate that the design can be recommended for implementation. The future study involves
the implementation of the design outcomes into a matrix learning game application.
ACKNOWLEDGMENT
Thank you to the Research and Community Service Institute (LPPM) of Universitas Dian Nuswantoro for
funding the Basic Research of Higher Education with contract number 049/A.38-04/UDN-09/V/2023.