Learning - Oriented Assessment in Speaking Class at University Level

This paper proposes a model of learning-oriented assessment to inform assessment theory and practice in speaking class. It focuses on two interrelated processes: the students’ assessment tasks and student engagement with feedback. The research method involves classroom observations of in English speaking class in university level and semistructured interviews with the lecturer and some sample of the students’ feedback. Findings highlight assessment tasks promoting thinking and practicing in the speaking activities and direct feedback to promote timely dialogues with students. The coherence of the model is discussed and some areas for further exploration are suggested. Based on the data analysis, the model shows significantly positive outcome. In the student side, it provides clearly path for them to reach the standard goal since they always get negative evidence from the teacher in each time they are out of the track. In the teacher side, it also provides positive effect. As they give negative evidence or correct the student’s mistake, they auomatically need to monitor the student’s development on every stages. It gives a crystal clear sight of student’s improvement. Finally it supports significantly on the assessment reliability of their speaking skill.


Introduction
The theory and practice of student assessment in higher education has generated a lot of debates.It is probably fair to say that there have been a number of positive developments during that time in promoting assessment for learning: more varied assessment tasks than merely a diet of final examinations; greater transparency in assessment criteria and strategies for promoting engagement with them; and a growing awareness of the importance, and the challenges, of developing effective feedback processes.(Norton, et al. 2013) There is now widespread acceptance for approaches to assessment focused on promoting and enhancing student learning (Sambell et al. 2013).Many teachers in higher education perceive that they lack individual autonomy and find themselves pulled indifferent directions by assessment purposes other than facilitating student learning (James, 2014).Effective assessment practice should focus on enhancing student learning processes.
Learning-oriented assessment is defined as assessment where a primary focus is on the potential to develop productive student learning processes.The processes ofworking towards well-designed summative assessment can also afford opportunities for formative assessment strategies, such as peer feedback and related teacher feedback.
Turning now to the other two strands of the model, learning-oriented assessment task design is supported by the interconnected elements (illustrated at the bottom of Fig. 1 by inverted arrows) of evaluative expertise and engagement with feedback.Evaluative expertise on the left of the figure represents the evolving ability of students to engage with quality criteria, develop their self-evaluative capacities and make informed judgments about their own work, and that of others.
Evaluative expertise is critical for student learning because to monitor and improve their learning, students need to know what quality performance involves and entails (Sadler 1989).A crucial role of the teacher is to assist students in developing this capability in discerning quality and making complex judgments (Sadler 2010).Developing assessment for informing judgment involves exposure for models and opportunities for practice (Boud and Falchikov 2007).

Khoirudin Yanuar Syam, J. Ari Wibowo| 101
Learning -Oriented Assessment in Speaking Class at University Level The two arrows leading from the top to the bottom of the figure suggest that the nature of the assessment task or tasks impact on prospects for the development of evaluative expertise and engagement with feedback.Task-type is one important issue, for example, 'on display assignments', such as oral presentations or posters in which work is openly evident to peers rather than just privately submitted to tutors (Hounsell 2003) provide opportunities for student appreciation of quality and associated development of evaluative expertise.The number and sequencing of tasks are another factor, for example, multi-stage assignments end to provide more opportunities for student engagement with feedback than a single end-of-semester task

Method
This research is a kind of descriptive qualitative research which consists of two variables and descriptively designed.It involves data that describes events, organization, depiction and description of the data collection.

Participants
The teacher was involved in 3 undergraduate student classes of Speaking for Formal Communication.In the process of observations, my co-researcher and I interacted with a range of students and invited a sample of them to participate in semi-structured interviews.

Data collection
The study tried to understand how teachers implemented and students experienced assessment in class under investigation.It did not aim to measure student achievement, instead it tried to explore the processes in which students were involved and their perceptions of issues arising.In line with this orientation, I carried out two main formal semi-structured interviews with the teacher: one at the outset of the study to understand their views on learning-oriented assessment issues and how teaching and assessment were approached in the courses; and the other to explore issues arising from the observations and the student data.
Students from each of the classes were interviewed in order to measure their perceptions of the learning-oriented assessment processes in the discussion.
Interviews focused on the relevant learning-oriented assessment issues arising in a particular course.
Selected randomly students were usually interviewed once for about 20 minutes, although longer interactions were common and a number of students were interviewed twice when the pertinent issues were significant over a sustained period of time.Overall, 15 interviews with 15 students were carried out in three classes.

Data analysis
The observational and interview data were assigned codes which represented my interpretation of their main essence.For the purposes of this paper, data reduction and selection of examples from the case reports are a central move.The aim is not to provide a comprehensive analysis of a case but to present one example from three different cases which illustrate the relevant feature of the model of learning-oriented assessment.Selection of quotations seeks to present a balanced view of the evidence from the wider data set.
The trust worthiness of interpretations was mainly developed through the following strategies: triangulation between observational data, teacher interviews and student interviews; investigated engagement with the participants in the classes under discussion; and data collection and interpretation operating directly The first feedback given to the students is about miss pronunciation, then term of diction which is inaccurate, the third is miss spelling and the last is fluency.
The fluency level are spread out from low level until advance level, however, it is dominated by moderate level.
The teaching method of the speaking class is by dividing the class members into 7 groups.Each group consists of 5 people who should present the formal speaking ability on their own, instead of in group.After that, the assessment is taken from its content, fluency and confidence The assessment pattern uses learning oriented assessment pattern focusing on students' speaking skill ability improvement based on speaking micro-skill (Brown, 2007).

Discussion
Judging from the interview and observation related to students' improvement of the formal speaking ability, it shows corresponding positive result.The students' perception of formal speaking has changed, as they get more understanding what and how they should be.When they make any mistake, they can learn from the given feedback by the teacher which gives a clear direction to go.
They have more confidence after knowing clearly the expected performance level and the way to perform in that level.They even improve themselves by finding some other techniques which help them to be better since they have known the right track.
the principal means of data collection were classroom observations and interviews.The main objectives of classroom observations were to develop an understanding of how classroom processes enrolled, particularly in relation to aspects relevant to the model of learning-oriented assessment.Classes were of 100 minutes duration in Speaking for Formal Communication 2 in English Department | Volume: 1 | Number: 1 | February 2016 | ISSN: | Khoirudin Yanuar Syam, J. Ari Wibowo| 102 Learning -Oriented Assessment in Speaking Class at University Level at University of Nusantara PGRI Kediri; activities were open-ended and did not follow a set time schedule.There 5 sessions in each class in 500 minutes.Detailed field notes were collected to describe classroom processes; identify issues for follow-up through interviews.

|
Volume: 1 | Number: 1 | February 2016 | ISSN: | Khoirudin Yanuar Syam, J. Ari Wibowo| 103 Learning -Oriented Assessment in Speaking Class at University Level Findings The findings are divided into two main sections which address the corresponding component of the learning-oriented assessment model; assessment task design and implementation.Task design across the cases involved different elements and emphases.The 2A, 2B, and 2C class involved continuous assessment of business plan presentations.The business plans presentations were chosen because it exemplifies prominent features of the design and implementation of learningoriented assessment tasks.