The Research Design RESEARCH METHODOLOGY

used to know how well students‟ score as a whole on speaking skill. It uses the formula: 5 X = X = mean n= number of student n x = individual Score Second, the researcher tries to get the class percentage which pass the minimum standard. It will use the formula: 6 P = F x 100 P = The class percentage N F = Total percentage Score N = Number of Students Third, after getting mean of student s‟ score per actions, the researcher identifies whether or not there might have students‟ improvement score on speaking skill from pre-test and post-test score in cycle 1 and cycle 2. In analyzing that, the writer uses the formula: 7 P = y1-y x 100 P = Percentage of students‟ improvement Y y = pre-test result y1 = post-test 1 P = y2-y x 100 P = Percentage of students‟ improvement Y y = pre-test result Y2 = post-test 2 2. Analyzing Students‟ Response Besides analyzing the result of the test and observing the activity during classroom Action Research, the researcher also uses questionnaire to analyzing th e students‟ response, the researcher uses formula: 5 W. James Popham, Educational Statistic, New York, Harper Row Publishers p.21. 6 Anas Sudijono, Pengantar Statistik Pendidikan, Jakarta: P.T. Raja Grafindo Persada,2008,p.43. 7 David E. Meltzer, The Relationship between mathematics Preparation and Conceptual Learning Gains in Physics: A Possible Hidden Variable in Diagnostic Pretest Scores, lowa: Department of Physics and Astronomy,2008,p.3. P = F x 100 P = The class percentage N F = Total percentage Score N = Number of Students 4 = Strongly agree 2 = Disagree 3 = Agree 1 = Strongly Disagree

I. The Assessment of Speaking

Speaking is a complex skill requiring the simultaneous use of different ability which often develops at different roles. Speaking skill are generally recognized in analysis of speech process that are pronunciation, grammar, vocabulary, fluency, and comprehension. Heaton presented the sample of an oral English rating scale that used 1-6 points. 8 Below is the frame of Heaton‟s oral English rating scale: Table 3.1 The Rating Score s’ criteria of Oral Test 6 - Pronunciation good - Only 2 or 3 grammatical errors - Not much searching for words - Very few long pauses - Fairly easy to understand - Very few interruptions necessary - Has mastered all oral skills on course. 5 - Pronunciation slightly influenced L1 - A few grammatical errors but most sentences correct -Sometimes searches for words 8 JB Heaton, Writing English Language Test, Longman: London and New York, 1990 p. 100 -Not too many long pauses -General meaning fairly clear but a few interruptions necessary -Has mastered almost all oral skill course 4 - Pronunciation influenced a little by L1 - A few grammatical errors but only 1 or 2 causing serious confusion searches for words - A few unnatural pauses - Conveys general meaning fairly clearly - A few interruptions necessary but intention always clear -Has mastered most of oral skills on course 3 - Pronunciation influenced by L1 - Pronunciation and grammatical errors - Several errors cause serious confusion - Longer pauses to search for word meaning - Fairly limited expression - Much can be understood although some effort needed for parts - Some interruptions necessary - Has mastered only some of oral skills on course. 2 - Several serious pronunciation errors - Basic grammar errors - Unnaturally long pauses - Very limited expression - Needs some effort to understand much of it