Abstract:

BS in Statistics and Business students at the Carlos III University learn computer programming. This subject is critical for the future of these students for three main reasons:
1) Statisticians handle a lot of data samples which have to be processed automatically. Many of these processes are not automated by existing tools on the market. Therefore, students must be able to adapt, create and/or modify these processes by scheduling functions required for data processing;
2) This subject serves as the basis for other degree programme subjects which use the R programming language applied to specific fields of statistics;
3) Programming skills and abilities are one of the requirements of 90% of graduate job offers.

From student outcomes and comments on this subject in previous years, it is clear that statistics and business students find the contents of this subject hard to grasp because it is the first time that they come in contact with programming and they do not have the skills required to develop software in their field of work: statistics.

Engineering students do not find it hard to learn programming because they have gained enough background in previous degree programme subjects. However, programming concepts are very tricky for statistics and business students to understand. At first, one might think that it is question of student application in terms of hours of study, but one would be very much mistaken. There is a big difference between the profile of the two student groups and the courses that they have completed. Engineering students will have taken other subjects that provide software engineering skills and capabilities, techniques and methodology empowering them to “think” about the solution to be implemented before they do any programming. Statistics and business students, however, have to acquire and infer these skills and knowledge from just one programming subject. This is the reason why the teaching of one and the same subject has to be adapted to its target audience.

The aim of this teaching innovation project is to teach BS in Statistics and Business students engineering techniques in order to provide a formal and repeatable methodology for the process analysis, coding and validation process. With this methodology, students learn to:
• Abstract and analyze problems
• Design different solutions and analyze which it would be best to implement
• Break down the problem into sub-problems to facilitate coding
• Learn to understand the R programming language paradigm for statistics and business students
• Program, run and debug programs in R
• Validate that the program works properly and monitor program implementation.

In order to demonstrate the impact of teaching software engineering techniques in the statistics and business field, we compare the outcomes for different academic years during which software engineering techniques were and were not applied. Also, students were interviewed, and their comments and opinions on learning programming through software engineering techniques are shown.

The paper is structured as follows. Section 2 describes different types of teaching. Section 3 presents the teaching innovation project, the methodology and how the software engineering techniques were taught to statistics and business students. Section 4 compares the data on outcomes with and without teaching software engineering techniques. Finally, Section 5 presents the conclusions and future work.

Keywords:

Teaching methodology, teaching techniques, best practices.