The objectives guiding this research were to determine the dominant learning style of the first year Building engineering students (University of Seville), the influence of the style on the grades in each school subject, and if learning style is influenced by socio-demographic variables. The sample was composed of 161 students, who completed the Honey-Alonso questionnaire (CHAEA) and the CDAT. Descriptive and correlational methods were used. For the analysis of the data, a univariate (one-way ANOVA) descriptive study was performed, and the non-parametric Mann-Whitney and Kruskal-Wallis tests, both with a confidence level of 95% (σ=.05) were used. The results show the predominance of the reflective style, and the fact that grades as a function of styles vary depending on the school subject. No significant differences were found in the variables gender, university admission scores, type of school (private or public) of the study centers attended previous to university. Significant differences were found for some styles in the variables age, type of university admission, and if a student had to work as well as to study. The conclusions highlight the need to develop teaching strategies that encompass all the styles to improve performance, as well as the need to modify the teaching practices for the students to develop the rest of the styles in a balanced manner. This latest aspect is important so that the students can freely operate professionally, as the Engineering profession requires having all four characteristic styles simultaneously.