Student motivation has been conceived as a crucial factor in the learning processes. However, research in motivation and creative learning in the secondary education music classroom has been limited. Student motivation is explored in this article through a collaborative action research study, in the form of several projects centred on the creation of music through group improvisation and cooperative composition, conducted over the course of five 1-year cycles, from 2008 to 2013. The aim of this study was to gain a deeper understanding of the aspects determining student motivation during creative learning processes. Three secondary schools in the region of Madrid (Spain), eleven third-year secondary school classes (n = 267), teachers, researchers and artists participated in the study. Data were gathered through participant and non-participant observations, interviews, a classroom diary, and a questionnaire. In the findings, six intrinsically linked connected forms of motivation are grouped in three levels and are related to different factors emerged during the teaching and learning processes and connected to creativity.

Most past models of student motivation have assumed that student motivation generalises across various achievement situations and curriculum domains; however, research has not fully explored the extent to which motivation may be domain-specific (Green, Martin, & Marsh, 2007; Martin, 2008). The purpose of the present investigation was to explore this issue by comparing and contrasting generalised models of motivation with domain-specific models and how they relate to achievement outcomes in mathematics and English. Secondary students (N = 476) completed both the domain-general (ISM, McInerney, 2003) and the researcher-derived domain-specific motivational measure (DSSM) followed by a standardised achievement test (WRAT-3, Wilkinson, 1993). Overall, the study resulted in mixed findings. There was some indication that there was enough independent variance between the domain-specific goal types to suggest they were tapping distinct constructs as found in previous research (Green et al., 2007). However, the small and often inconsistent correlations with achievement outcomes brings into question the usefulness for educators and the research practicality of pursuing such a division.