Spatial reasoning within High-Stakes Tests: Indonesian National Exam (UN), TIMSS and PISA

Year: 2017

Author: Winarti, Destina

Type of paper: Abstract refereed

In recent years, international high-stakes testing has become more prevalent, despite some of the documented pitfalls in the tests (i.e., Trends in International Mathematics and Science Study (TIMSS) & Programme for International Student Assessment (PISA)). One of the criticisms of international testing relates to the political impact of the results. While the tests may cover a limited range of content, the rankings released have a bearing on the international standing of the participating countries. The tests are seen as a reflection of the performance of the respective countries' education system. Nevertheless, both tests have the potential to create understanding and identify where the strengths and weaknesses of education systems lie.
An increasing number of studies provide strong evidence for a link between spatial reasoning and mathematics performance. For some time, researchers have encouraged the integration of spatial reasoning within mathematics curriculum. Since curriculum developments are influenced by assessment practices, we were interested in investigating how spatial reasoning is assessed in high-stakes tests, particularly in mathematics.
The studies of high-stakes tests thus far have covered academic content, social and economic impacts, as well as focusing on the education systems. Somehow, the content items which are related to spatial reasoning have not become a consideration to date. This study aims to undertake a content analysis of mathematics tasks that require spatial reasoning across national and international tests. Data were collected from TIMSS released items 2011, PISA released items 2012 and UN items 2011-2016. Items were coded into graphic and non-graphic, next; every graphic item will be coded again into spatial reasoning constructs namely spatial visualization, spatial orientation, mental rotation or a combination of two constructs.
Results revealed that the proportion of spatial reasoning items within each test were approximately 37% for TIMSS, 31% for PISA, and 24% for UN. These data suggest that spatial reasoning has become an influential aspect of item development despite the fact that these constructs are not identified in test composition. These data show that UN has the least proportion of spatial reasoning items. The study highlights the importance of explicitly teaching spatial reasoning, as school-based curriculum content inevitably aligns to assessment practices. The findings in this research could shed light on the need for governments to start focusing attention on the refinement of mathematics curriculums, particularly in Indonesia, to reflect the prominence of spatial reasoning in today's mathematics assessment.