Introduction

The didactic strategies used for teaching and learning software programming are complex because information must be acquired, coded and retrieved based on logical-mathematical thinking to design instructions that will be executed by a computer in order to solve problems of diverse nature. This justifies why these contents are in high demand in today's higher education, despite the fact that the degree indices are, on the contrary, very low.

Objective

Propose and evaluate general and specific didactic strategies for teaching and learning software programming in order to create an alternative to reduce low-performance records and dropout in careers that include subjects from the above area.

Method

Design quasiexperiemtal, quantitative analysis not parametric, longitudinal (four evaluations), with a sample of 78 students of the career of engineering in Computation, with random distribution in two experimental groups and two groups of control in scenarios of intelligent classrooms.

Material

Software programs with C and C++ languages in educational platform, as well as the use of Autogradr and the Inventory of metacognitive strategies.

Results

A Kruskal-Wallis test was used because we had a sample without normal distribution and with scale by hierarchies. The acceptance of the null hypothesis was observed in both tests: 1) H = 0.04, in superior academic performance in the experimental groups, the interdependence in knowledge is regular high 0.657, with difference of 1.7 in average performance, and 2) H = 0.24, in superior metacognitive strategies in the experimental groups, the interdependence in metacognitive strategies is low 0.342. The planning area is highlighted as an elementary phase for the solution of problems.

Conclusions

It was demonstrated that the didactic strategy exposed in three specific blocks of teaching and learning elevated the academic performance and the metacognition of the students.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )