Coping as Self-Regulation of Anxiety: A Model for Math Achievement in High-Stakes Tests

Abstract

ABSTRACT The aim of this study was to investigate the impact of cognitive and variables on the mathematics achievement of students taking the University Entrance Examination in Turkey, with metacognition and coping as self-regulatory variables. A model was proposed and tested by a program designed for structural equation modeling. Included in the model were variables in the cognitive and domains: mathematical background in the cognitive domain, and math anxiety, test anxiety, achievement motivation, and self-efficacy in the domain. Metacognition, taken to be self-regulation of cognition, and coping, taken to be selfregulation of anxiety, were specifically included. The sample comprised 751 high school seniors and recent high school graduates. The results indicated that the students' mathematical background was the dominant predictor of mathematics achievement on the examination. Math anxiety more than test anxiety, self-efficacy, and two coping styles also had significant influence on achievement. Non-productive coping was the operational coping style with a positive effect on achievement and a buffering effect on anxiety. According to the theoretical conceptualisation of the study, coping as the self-regulation of anxiety rather than cognitive self-regulation was found to influence performance on the examination. KEYWORDS: anxiety, coping, self-regulation, mathematics achievement, testing. INTRODUCTION Assessment procedures are used to rank and compare the academic performance of individual students and student groups, often for the purpose of admitting or rejecting candidates for the next layer of education or for employment opportunities (Chapman & Snyder, 2000). They are described as high-stakes assessments if much depends upon one's level of performance. The University Entrance Examination in Turkey is an example of a assessment; it is highly competitive and a student's score determines the nature of the student's future. Approximately 1.5 million students take the examination every year and about a third of these students are accepted into various forms of higher education programs (OSYM, 2010a). When compared with performance in other subjects, mathematics achievement measured by the University Entrance Examination is far below expectations. In 2003, the average net score for the mathematics section was 10.1 (standard deviation 12.9) out of 45 items, as opposed to an average net score of 20.7 (standard deviation 11.8) out of 45 items in the Turkish language section (OSYM, 2003b). In 2009, the average net score for two sets of 30 mathematics items were 9 (standard deviation 9.2) and 8.7 (standard deviation 7.6) while the average net score was 14.1 out of 30 items in the Turkish language section (standard deviation 6.7) (OSYM - student selection and placement center, 2010b). This research study examines the factors affecting math achievement on a test. Including and self-regulatory as well as cognitive constructs when modeling mathematics achievement is a recent trend (De Corte, Verschaffel, & Op't Eynde, 2000). Students can be described as self-regulated learners to the degree that they are metacognitively, motivationally, and behaviorally active participants in their own learning processes (Zimmerman, 1989). At this point it is important to note that human mental functions can be divided into the three areas of cognition, affect, and conation (will). But in contemporary psychology, there is a tendency to make a dual separation between cognition and functions such as emotions and motivation (Parkinson & Colman, 1995). Throughout the article, affective will refer to one side of this duality. Learners who are more proficient at self-regulation in the cognitive, affective, and behavioral domains are expected to reach higher levels of achievement. Self-regulatory skills include the setting of clear and realistic goals, selecting from a range of strategies, evaluating progress, modifying procedures, and seeking help when needed. …