Dominant achievement goals across tracks in high school

Abstract

The dominant achievement goals (DAGs) of 7008 students in the third grade of Dutch secondary education (US grade 9) were investigated, based on Elliot & McGregors’ 2 × 2 framework (2001), in relation to track-level and motivational variables. We found the mastery-approach goal and the performance-approach goal, generally considered adaptive, to be more prominent among students in lower tracks. In contrast, avoidance goals were more common in higher tracks. Most notably, in the highest track, the mastery-avoidance goal was the most prominent. Additionally, we found that students with a dominant performance-approach goal scored highest on almost all motivational variables examined; students without a DAG scored mostly second highest. The implications of these findings are discussed.

Keywords:

• motivation
• achievement goal
• secondary education

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The authors wish to express their gratitude to Nico van Yperen for helping us with the formulation of the achievement goals, providing the syntax for scoring them and commenting on an earlier version of this paper.

2. We used the Cool^5–18 database, Secondary Education segment, which was composed from data gathered by GION and CITO. The data are available for researchers via Data Archiving and Network Systems (The Netherlands, http://www.dans.knaw.nl/).

3. In a ‘chance model’, exactly half of the respondents would be assigned to a DAG. Of the 2⁶ = 64 possible patterns for answering the items, 32 lead to the assignment of a DAG, 8 to each DAG.

4. Two of our scales have a below-standard reliability; the ISM scales ‘task’ and ‘affiliation’ show a Cronbach’s alpha of .59 and .68, respectively. This means that if these scales were used as outcome variables, the square roots of these alphas would be the upper bounds of validity; these are .77 and .82 for ‘task’ and ‘affiliation’, respectively. Generally, the associations between variables are significantly attenuated in the case of low alphas. Comments on the use of Cronbach’s alpha have been with us for decades. Schmitt (1996), for instance, argues: ‘When a measure has other desirable properties, such as meaningful content coverage of some domain and reasonable unidimensionality, this low reliability may not be a major impediment to its use’ (Schmitt, 1996, pp. 351–352). Because an exploratory factor analysis, in which eight factors are rotated using the varimax rotation, yields the eight subscales of the ISM (Zijsling, Keuning, Kuyper, Van Batenburg, & Hemker, 2009), the scales ‘task’ and ‘affiliation’ have reasonable unidimensionality. Deletion of one item from ‘affiliation’ would have resulted in a Cronbach’s alpha of .74. However, this was not done because then only two, similarly formulated, items would remain.

Additional information

Funding

This work was supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek [grant number 023.001.190].