When you think about the many uses of tests and the influence of tests on student’s lives, it is not surprising that the testing situation may evoke anxiety reactions in many students, sometimes even so disturbing that they need professional assistance. The earliest investigations of test anxiety (1900-1950) basically dealt with physiological reactions resulting from the activation of the autonomic nervous system experienced by examinees during stressful examinations (Spielberger & Vagg, 1995). This definition neglected the experiential qualities of emotional states and individual differences in anxiety occurrence. Later, Spielberger (1972) conceptualized test anxiety as a “situation specific form of T-anxiety” with worry and emotion as major components. Liebert and Morris (1967) suggested that the worry component was associated with performance decreases on cognitive tasks and the emotionality component was unrelated with task performance. In view of that, test anxiety has been defined as “an unpleasant feeling or emotional state that has physiological and behavioral concomitants, and that is experienced in formal testing or other evaluative situations” (Dusek, 1980, p.88). Researchers in the field of education, counseling, and psychology have become interested in antecedents and consequences of the test anxiety (e.g., Hancock, 2001; Hembree, 1988). Despite the large volume of literature on test anxiety, there are a few studies addressing students’ emotional regulation strategies during test taking. Yet the ability in controlling one’s emotions is an important characteristic one should possess. The purpose of emotional regulation is neither to repress emotions nor only to have an individual always in a calm state of emotional arousal. Instead, emotional regulation includes processes of monitoring, evaluating, and changing one’s emotional experiences (Thompson, 1994). Cicchetti, Ganiban, and Barnett (1991) defined emotional regulation as “the intra and extra organismic factors by which emotional arousal is redirected, controlled, modulated, and modified to enable an individual to function adaptively”(p. 15). In terms of research on emotional regulation during test taking, Schutz, Distefano, Benson, and Davis (2004) proposed that emotional regulation during testing can be conceptualized in three dimensions: task-focusing processes, emotion-focusing processes, and cognitive appraising processes. Task focusing processes include students’ self-talk while taking tests, which help them focus on the test rather than their emotions during the test. The examples of task focusing strategies include time monitoring during the test, finding the main idea in the question, and eliminating the distracters. Emotion focusing strategies, on the other hand, shift students’ focus from the task to the feelings and thoughts related to the task. These include self-talks in the form of blaming one’s self and wishful thinking (hoping the problem will go away). These kind of internal talks tend to increase anxiety during testing. It is important to note that not only the nature of self-talk but also the length and intensity influence emotions (Schutz & Davis, 2000). Finally, cognitive appraising processes are the judgments students make about the test and their ability to cope with the problems that occur during the test. This dimension has gained attention by many researchers in the area of emotions. These judgments are affected by the way student look at the world. Considering this conceptual framework of emotional regulation, Schutz et al. developed “The Emotional Regulation During Test-Taking Scale” comprising of three major dimensions and 39 items. In the present study, we aimed to collect data from a cross-section of Turkish students in an effort to examine high school students’ emotions (specifically anxiety) and emotion regulation during tests. More specifically, we were interested in whether test anxiety would be predicted by some demographic variables and emotional regulatory strategies.

The data were collected through Emotion Regulation during Test Taking Scale (ERT) developed by Schutz et al. (2004) and anxiety subscale of Academic Emotions Questionnaire (AEQ) developed by Pekrun, Goetz, Titz, and Perry (2002). This study consisted of three stages including the instrument adaptation to Turkish context. In the first stage, the instrument was translated to Turkish language by six experts. In the second stage, as a pilot study, instrument was administered to 312 high school students. Factor analysis study was conducted to provide evidence for construct validity. The findings indicated that items were loaded on the expected constructs. Reliability coefficients were ranging from .50 to .87. The results were consisted with the results of Schutz et al. study (2004). In the last stage of the study, the instrument was administered to 460 ninth grade and 448 tenth grade students. From 908 students, 47.8% were girls and 51.4% were boys. The findings of factor analysis confirmed the factorial structure of ERT. Results also indicated that test anxiety scale had high internal consistency (.83). For cognitive appraising processes, alpha coefficients ranged from .67 and .75; for task focusing processes, they were from .52 to .75; and for emotion focusing processes, they were from .69 to .77.

Hierarchical multiple regression analysis was conducted to predict test anxiety by gender, school type, and three emotional regulation dimensions: task-focusing processes, emotion-focusing processes, and cognitive appraising processes. The result of the analysis indicated that gender and school types accounted for a significant amount of variability in test anxiety, R² = .04, F(2,1039)=22.25, p<.05, indicating that females have higher test anxiety than males and students attending to high school without entrance examinations tended to have higher test anxiety. Addition of cognitive appraising strategies resulted in a significant increase in explained variation (R² = .18, F(5,1039)=44.92, p<.05). Of these variables, testing problem efficacy had a significant negative relationship and uniquely accounting for 12.25 % of the variability in test anxiety, indicating that students who had higher testing problem efficacy had lower test anxiety. Inclusion of task focusing processes also accounted for a significant amount of variability after controlling for the effects of school type, gender, and cognitive appraising process dimensions, R² = .23, F(8,1033)=37.40, p<.05. Among these strategies, task focused strategy use and importance reappraisal significantly accounted for 2% and 0.7% of the variance respectively. Findings indicated that the more students used task focused strategies and importance reappraisal, the higher the level of test anxiety was. Finally, addition of emotion focusing processes also improved the prediction of test anxiety, R² = .33, F(10,1031)=51.55, p<.05. Among these processes, both self blaming and wishful thinking were found to be significant predictors of test anxiety uniquely accounting for 6.8% and 1.6% of the variability. The relationship between them indicated that the more students engaged in wishful thinking strategies and the more they blamed themselves for bad performance in tests, the higher the test anxiety was. Results of the study revealed that gender, school type and emotional regulation strategies (specifically testing problem efficacy, task focused strategy use, importance reappraisal, self blaming, and wishful thinking) play an important role in investigating test anxiety. These findings are preliminary. Ultimately, we are also planning to investigate whether student achievement is predicted by these variables.