Washback Effect of EFL Teacher-Made Test on Teaching-Learning Process of Communicative English Skills Course at Ambo University

Prompted by the problem the effects of washback, this study was mainly intended to evaluate the washback effect of EFL teacher-made tests/examination on teaching-learning process of Communicative English Skills at Ambo University. Specifically, it examined the influence of the examination on the domains of teaching and learning. It further explored mediating factors of wash back on teachers and students’ practices. To this end, the data were collected from three years (2008, 2009, 2010 E.C.) communicative English skills course teachers’ made examinations through document analysis method. Supplemental data were also collected from EFL teachers and English major students respectively using interview and questionnaire. Both teachers and the students were selected on the bases of randomly sapling techniques. The study employed mixed methods to analyze and interpret the results of the study. The findings of the study reveal that teacher-made examination affected the domains of teaching-learning process negatively. This means it affects students’ learning attitudes, motivations, and readiness in leaning some components of the materials, and experience in taking examination, practice in using the learning materials and methods in general. Some components of the course, such as writing and listening, were not included in the final examination of the course. As a result, students were influenced to give more time for the work of previous examination questions rather than studying the materials. Based on the findings of the study, teachers are recommended to set examination questions from the contents of the materials they used to teach the components of the course and give equal emphasis for all the components of the language during teaching-learning process and assessment. As the committee members usually set the examinations, the teaching material choice should be amended to align the contents of the examination questions with the contents in the teaching materials.

should incorporate authentic tasks that lead to achieve language-learning goals and to foster learner autonomy and self-assessment. The potential backwash effect should join validity and reliability in the balance against practicality (Hughes, 2003). In this case, Sukyadi and Mardiani, (2011) assert that the concept of the washback effect of the test includes validity, reliability, authenticity and practicality the test.
To find out whether testing will bring positive washback effect on teaching and learning or not, Taylor (2005) (pp. 276-277) The concept of examination washback effects can be studied in terms of two dimensions: micro effects and/or macro effects (Cheng, et al., 2004, Hughes, 2003Thaidan, 2015). Some language testers describe effect/outcome as micro washback and consider impact as macro washback in the language context (Taylor, 2005). In this regard, (Pan, 2009) states that the effect of test on classroom teaching and learning is micro washback whereas the impact of test on the community as a whole is referred to as test macro washback scenarios (Sadeghi, of 'consequential validity' in which the social consequences of testing are part of a broader, unified concept of test validity (Messick, 1996;Bailey, 1996). The following diagram 2 specifies in the components of washback effects: macro and Micro levels. Source: Thaidan (2015) The above figure reveals the components of washback effects. The first is macro washback effects. Macro washback effects refer to the influences of tests in which it influences a society and are influenced by the social climates. Studies on macro washback effects can be conducted on high-stakes tests, which incorporate a large number of test-takers and major decisions making (Thaidan, 2015;Messick, 1996;Bailey, 1996). In this case, studies focus on the effects of tests and test results on the professional and employment opportunities, teaching learning system and society more widely (Thaidan, 2015). The second part of washback is micro washback. Micro-washback effects refer to low-stakes tests, such as in-class tests, teacher-made tests, classroom assessments or teachers' assessments (Thaidan, 2015;Sadeghi, 2014 This study was therefore limited to micro level or teacher-made test washback effects. Teacher-made/micro washback refers to an assessment carried out by teachers in the classroom as a central part of the educational process (Anderson and Bachman, 2006;Spratt, 2005). The purpose of teacher-made test /micro washback is to describe instructional purposes, influence what students study, and help instructors to obtain perspective on their courses (Thaidan, 2015).
Teacher-made tests are normally prepared and administered for testing classroom achievement of learners, evaluating the method of teaching adopted by the teacher (Madsen, 1983). Teacher made-test is designed to solve the problem or check the requirements of the class for which it is prepared (Djuric, 2008;Spratt, 2005). In short, micro washback/teacher-made tests are the most worthy instrument in the hands of teacher (Shih, 2007). However, micro washback has limitation cover a broad range of abilities and skills in language skills (Green, 2007). Teachermade tests measure a limited part of a subject area in a very helpful way to define the shortterm course objectives envisioned by the teacher (Thaidan, 2015). However, the validity of teacher-made examination remains debatable in the educational assessment process.
Previously, various studies have globally and locally been conducted in different areas of test washback effects. Globally, Djuric (2008); Green (2007); Shih (2007); Thaidan (2015) have studied on English National Examination washback effects and found out the Situations of positive and negative washback effects. Locally, Simachew (2012) and Yasin (2014)  communicative English skills course that ha complex components of language objectives. For this reason, this article evaluates the washback effects of EFL teacher-made tests on the teaching-learning processes of communicative English skills course (Spratt, 2005).
Specifically, the current article has examined how teacher-made tests influence the teachers and students' perception, attitude and motivation, activities and selection of teaching-learning materials in the course.

Statement of the problem
The core of the problem is concerned with the assumption that the EFL teacher-made tests fail to measure the entire components of the Communicative English course contents (Hawkey, 2006;Ambachew, 2003). Teacher-made tests normally measure only a limited part of abilities and rely too heavily on memorizing facts and procedures (Hawkey, 2006;Hughes, 2003). Teacher-made test washback effect is debatable because it is less valid and reliable against practicality (Motuma, 2014;Hughes, 2003). This means, the result of the communicative English skills course teachers-made test usually create grievance because of its inconsistent for most students (Nugussie, 2002;Motuma, 2015). On the other hand, most (67%) of Ambo University students are trying to cheat at least on one or two sort of assessment tools the teachers use (Motuma, 2015). These students are not conscious about the effect of tests and test results on their future professional career (For example, access to educational and employment opportunities, Taylor, 2005;Nugussie, 2002).
Scholars believe that university teacher-made tests are the base for teaching learning process and for the students' future career. Teachers-made tests determine the classroom activities of both teachers and students (Davies, 2014;Nugussie, 2002). However, with the expansion of universities, the problem of teacher-made washback effects has become more and more apparent at the level. Though universities have almost harmonized curriculum and Kumar, 2011). This method is used to answer the questions as well as to assist a researcher to address the research problem. Besides, mixed research approach helps to provide both quantitative and qualitative answers to the questions of "who", "what", "why", and "how".
Based on this pragmatic research paradigm, the population, sample size and sampling techniques, data gathering tools and data analysis methods were presented below. To triangulate the data gathering instruments to gauge micro levels of washback: document analysis, pre-and-post tests interview with students, semi-structured interview with teachers, and questionnaires (Ching, 2008) were used. Document Analysis is focused on the analyses of the contents of examinations and teaching module (Davies, 2014). The aim of content analyses was to understand the alignment/relationship of the course module and the tests contents and objectives. In addition to document analyses, based on the recommendations of Ching (2008) to gauge micro levels of washback, pre-and-post tests interview was conducted with nine students ( three students with top, three students with medium and three students with lower scores in their previous examination results   reading (34%), grammar (31%), vocabulary (20%) and speaking (14.5%) to measure the students' abilities and performances. However, listening and writing were ignored from the examinations throughout the years. However, this does not mean that students' performances in writing and listening were not assessed. The EFL teachers confirmed during the interview time that though these two skills were ignored from the final examinations, they were assessed using continuous assessment techniques. The students replied to the questionnaire that they were being assesses in these skills during the continuous assessment.

Relationships between Textbooks and Examinations
There were some observable mismatch representation and disproportion between the contents in the course materials and examinations. In the materials of course, reading, vocabulary, grammar, listening, speaking and writing constituted in descending order covered respectively 33%, 20%, 17%, 17%, 10% and 4%. This shows that the language components received a varying degree of attention because of the objectives of the course. The following diagram three (pie chart) shows the comparison between the frequency percentages of the contents of the teaching materials and the teacher-made examination in general in the communicative English skills course during the three years. and the periods allotted to them as closer as possible to the reality and categorized them based on the judgement of the researcher. The relationships among the contents of the frequencies obtained from the analysis of the course materials, the items of examinations and the time allotted to cover the course materials have been presented in Figure 4 below. The Relationship of the contents of the examinations, module and time by frequency percentages.   The data in Table 4 presents the sum of the total expected value/frequencies (1211) of the teaching materials, which was computed from the total observable value (1073)

Discussion
The discussion of the results integrated the data obtained from documents' analysis, teachers' interview, students' preand post-tests interview and students' questionnaire.
Descriptive statistics such as frequency percentages were used to compare the contents of teachers to skip the practice of listening skill activities in the course content. Shih (2007) argues that students and teachers' perceptions of tests affect their learning practices.
The responses of the students to the questionnaire revealed that most students have experienced the influence of the examination on their learning. Normally, students believe that continuous, comprehensive, and proportional examination items to what they have learnt encourage them to work hard, use the teaching materials and time properly. This is because washback effects of examinations influence students' perception, attitude and motivation towards the skills they are learning. The contents of the questions in the final examination change the students' perceptions to use reference materials, time and methods of learning and studying the entire course contents. In general, the course examination has harmful washbach effects on learners/students that they fear, ignore and late to class. In relation to writing, correct punctuation and spelling, sentence, paragraph, completing dialogue, technical aspects of writing such as letter types and formats of writing were included in the teaching materials.
Hence, comparing with other skills, students were not equally motivated for listening and writing skills. This confirms that the consequence of the examination was negative washback effect.
Finally, to distinguish the EFL teachers and students' choices, the kind of teaching materials they frequently use and why they preferred these materials were analyzed.The major question asked in this sub-topic was: "Do communicative English Skills course teachers use additional materials in teaching the course? What kinds of reference materials do they usually use? Why do they prefer the materials the usually use?" Most of the teachers and students try to use additional materials to align the teaching learning processes to the nature of the final examinations of the course. The purpose of the selection of the materials is to help students be objectives to be assessed was apparent. The degree of the extent of relationship between the content area of the components of teaching materials and the teacher-made examinations was found to be very low. The correlation contingency coefficient between the two observations reads 0.38, 0.158, and 0.273 for those years. Moreover, cumulative contingency coefficient of the years reads 0.115 which shows very low relationship or harmful/negative washback effect of the teacher-made examination on the teaching-learning processes of the course. Negative washback effects of teacher-made examinations also influenced teachers' choice materials for teaching the course. It negatively affected student's confidence, attitude and perception towards the examination as a result students developed, stress, fear and anxiety towards examination; they usually come late to class and ignore the class the examination of the course.

Recommendations
The incomprehensive and disproportional distributions of the contents among the language contents within and between the teacher-made and teaching materials negatively affected the processes of the teaching learning of the course. Hence, embarking on the harmful effect of examinations, the teachers should first carefully draw up