Abstract:

The purpose of this paper is to present a program written in Matlab-Octave for the simulation of the time evolution of student curricula, i.e, how students pass their subjects along time until graduation.
The program computes, from the simulations, the academic performance rates for the subjects of the study plan for each semester as well as the overall rates, which are a) the efficiency rate defined as the ratio of the number of students passing the exam to the number of students who registered for it and b) the success rate, defined as the ratio of the number of students passing the exam to the number of students who not only registered for it but also actually took it. Additionally, we compute the rates for the bachelor academic degree which are established for Spain by the National Quality Evaluation and Accreditación Agency (ANECA) and which are the graduation rate (measured as the percentage of students who finish as scheduled in the plan or taking an extra year) and the efficiency rate (measured as the percentage of credits which a student who graduated has really taken).

The simulation is done in terms of the probabilities of passing all the subjects in their study plan. The application of the simulator to Polytech students in Madrid, where requirements for passing are especially stiff in first and second year subjects, is particularly relevant to analyze student cohorts and the probabilities of students finishing in the minimum of four years, or taking and extra year or two extra years, and so forth. It is also a very useful tool when designing new study plans.

The calculation of the probability distribution of the random variable "number of semesters a student has taken to complete the curricula and graduate" is difficult or even unfeasible to obtain analytically, and this is even truer when we incorporate uncertainty in parameter estimation. This is why we apply Monte Carlo simulation which not only provides illustration of the stochastic process but also a method for computation.

To generate the sample paths of student curricula we assume the following hypothesis:
1) The probability that a student passes a subject is a value selected at random from within the confidence interval obtained from the data of the University. 2) The students behave independently and within for a given student, future behaviour is independent of the past. 4) The probability of passing a subject increases as the student takes the exam more times. 5) The student always takes the exam as long as allowed by the credit thresholds. 6) For the second half of the curriculum vitae (CV) the students are alloted to the different majors (specialities) in accordance with its distribution along previous years.

The stochastic simulator is a useful tool for identification of the subjects most critical and subsequently for a decision making process in terms of CV planning and passing standards in the University.

The Project has been funded by the Call for Innovation in Education Projects of Universidad Politécnica de Madrid (UPM) through a Project of its school Escuela Técnica Superior de Ingenieros Industriales

Keywords:

Accreditation, Quality Processes, Curriculum Design.