Assessing representation techniques of programs supported by GreedEx

In systems supporting learning of programming, it is common to use several techniques to represent algorithms, some of textual and graphical nature. This kind of representations is used by the GreedEx system, a system for interactive experimentation with greedy algorithms. The assessment made of this system thus far is based on the use of questionnaires and, therefore, on the subjective perception of the student with respect to the usefulness and/or complexity of the representations provided. This paper presents two empirical evaluations conducted with the aim of learning about both aspects (usefulness and complexity) of each of the representations supported by the GreedEx system. In this evaluation several information sources are combined: subjective perception questionnaires and the metrics provided by an eye-tracking device.