Abstract:

The aim of this work is to develop and improve the abilities of (Computer) Engineering students in formulating and solving problems by following the scientific method. This work has been jointly proposed from two courses in the syllabus: “Statistics” and "User Interfaces”, both in the third year of the Computer Engineering at University Jaume I.
It is our belief and experience that many (computer) enginering students are not properly trained in the scientific method, and they tend to rely too much on prejudices and intuition. However, this situation is clearly not desirable. In particular, we believe that in the third year of Engineering, a student should be able to pose and design a problem in a rigorous way, to solve it and to interpret and validate the solution. Therefore, in the context of “Statistics”, which is taught in the first semester of their third year, before the course “User Interfaces” starts at the second semester, we propose our students to look for a real problem related to human-computer interaction, which could be of interest to them. Our aim is that students develop, since the beginning of the year, attitudes and skills that are required in “User interface”, in other future courses, and in their training as engineers in general.
After defining the problem, they must design the experiments, perform them and collect the data, choose the most appropriated Statistical models and tools, apply them to their data and interpret the results. Finally they must elaborate a full scientific report and summarize their work and results in a poster that will be displayed for several days in a common public area at our School.

In this paper we would like to present the project as it is proposed to our students and to analyze the topics chosen and their adequacy into the context of the interaction. We will detail the criteria to assess students' work in the two subjects, and the difficulties that the students are finding. We will also describe how the project relates to both courses, and the formative implications for our students.

Keywords:

Scientific method, statistics, user interfaces.