| Language testing washback effects have been studied recently by many language educators. Both positive and negative washback effects have been found in various empirical studies (Alderson & Wall, 1993; Shohamy, 1996). However, that research has focused mainly on standardized commercial tests and their influences on program participants and on teaching contents and methods. Few, if any, studies have explored the washback effects directly from teacher-made tests, which are the most frequently used in classroom settings. Thus, it can be expected that such tests play a crucial role in the mediation of teaching and learning effects. The purpose of this thesis was to review the different studies of washback effects, determine the positive or negative effects they have, and suggest what language teachers can do to increase the positive effects and reduce the negative effects. Individual item formats and their effects on vocabulary learning were also examined. On the affective side, the level of test anxiety accompanying the testing method was investigated with the hope that FA (FA) would surpass DA (DA). Two research projects in high school and college level English classes were conducted to investigate washback effects, relationships between item formats and vocabulary acquisition, and test anxiety levels.;In the first study, 467 college freshman participated. They were given an English Proficiency Pretest and an Achievement Test Anxiety Questionnaire (ATAQ) at the beginning and the end of a semester respectively. The results indicated that students' FA level increased after the edumetric testing method while in some classes, the DA level also increased. This result partly supported my hypothesis. Perhaps more measures should be taken to resolve the DA problems.;The second study explored the interrelationships among item format, vocabulary acquisition, and test anxiety. One hundred thirty-eight students were given a pretest and ATA at the beginning of the study. They were also given three posttests after each session. At the end of the semester, they completed a post questionnaire and a posttest. Results showed that the three different test formats did not show any significant influence on high school students' acquisition of the foreign language vocabulary. Students' anxiety level also did not change significantly from debilitative to facilitative. This ran counter to my hypothesis, which posited that the testing method would reduce DA level.;However, the edumetric testing method did result in a significant change in the students' performance on their final proficiency test scores compared to the control group. Still, further research should be done to determine the cause or other possible influences. |
