The Dominant Errors of Speech Production Committed in Speaking Class Interaction

This study is focused on the students’ errors in speech production. It aims at describing the dominant errors that committed by the students in speaking class. The objects of the study are the students of the first semester and the third semester of English Education Study Program of Universitas Islam Indonesia (UII). This research was qualitative research study. The researcher collected the data through audio recording, listened and made the script from the audio recording, read the script and identified the data, selected the data, and classified the silent pause of error and filled pause. The researcher analyses the data used the theory of Clark and Clark and Dulay. The result indicated that the total errors are 84 utterances containing 108 silent pauses for the first semester and 32 Utterances containing 34 silent pauses for the third semester, and 51 utterances containing 57 filled pauses for the first semester student and 89 utterances containing 124 filled pauses for the third semester students. Silent pause is the dominant errors made by the first semester students and filled pause is the dominant errors made by the third semester student. The error sources are cognitive reason and situational anxiety.


INTRODUCTION
Speaking is one of obligatory subject that should be taken by student of English education study program. In the same manner as English education department of UII defends the students to take speaking class continuously from the first semester until the third semester, because it becomes the compulsory subject. In Speaking I (one) students are thought daily interaction such as interpersonal communication and situational communication. After completing the course the student is able to understand the types of communication used in daily life, to communicate their feelings and thoughts using the appropriate expression, able to apply the degrees of formality and informality in speaking appropriately, able to set clear objectives for speaking and organize talks in a logical manner.
However, Speaking II (two) called classroom English. Students were thought a job interview and presentation. In Speaking III (three) students were thought about public speaking such as speech and debate. Each skill focuses on the different concentration. Even though they have different level of learning speaking, it cannot be supposed which semester makes an error in speech production more than the other. It cannot predict the speech error made by these two levels.
In fact, that speaking becomes the important act. The speaker speaks and gives the effect to the listener. The speaker gives the information and the listener absorbs it. Therefore, listening and speaking is almost closely interrelated" (Brown, 2004). In speaking the speaker delivers message to the listener, how they can give the information clearly, how they can transfer the idea, and opinion to the listener. During transferring the idea and opinion speakers tried to construct the correct utterances to avoid misunderstanding, it can be the grammatical, the phonological etc. According to Fauziati (2013) said that "speaking seems to be instrumental act". However, students speak to formulate plans and executes them or producing them, but in daily life speech, student common made an error in speaking. Sometimes the students made execution, such as filled with pauses and hesitate or stop in the middle of the sentence for a while to think the appropriate word. The student also usually makes corrections, repeats, replacements and even slip of tongue. According to Dell in Poulise (1999) as cited by Fauziati (2013) "people slip their tongue now and again, when the speakers are tired, a bit drunk, and rather nervous. According to Clark and Clark,(1977: 263) as cited by Fauziati (2013) said that, the common type of speech errors as follows: Silent Pause; A period of no speech between words, such as turn on the // heater switch. Filled Pause: A gap filled by ah, er, uh, mm, such as in turn on, uh, the heater witch. Repeats: The repetitions of one or more word in a row, such as turn on the heater // the heater switch. False Start ( Unretraced): False starts are correction of a words, such as turn on the stove // heater switch. False (Retraced): It starts are repetitions of one or more word before the corrected words, such as turn on the stove // the heater switch. Corrections: They are like false starts, but they contain an explicit correction, such as turn on turn on the stove switch -I mean the heater switch. Interjections: They like hesitation pause, indicate that speakers have had to stop to think about what to say next. In English often emerge with sounds oh, ah, well, and say, for example turn on, oh, the heater switch. Stutters: Speaker who stutter speak rapidly the same sound or syllable, as in turn on the h-h-h heater switch. Slip of Tongue: Speaker may make errors in sounds, word parts, words and even sentence structures. They may include substitution, metathesis, omission, or addition of segments as in, turn on the sweeter hitch (Fauziati, 2013). According to Clark and Clark (1997) as cited by Fauziati (2013) There are three possible sources of planning difficulty are cognitive reasons, anxiety, and social reasons. Cognitive difficulty is people obtain a longer time to create sentences. Anxiety is when people anxious, they become tense, and their planning and execution of speech become less well-organized. And the last is Social Factor, for example speech plan appears difficult when conversation gets place under force.
Unfortunately, it appears a number of undergraduate students of English Education Department of Islamic University of Indonesia they still make the types of errors in speech production. Even though, they have learned speaking for numerous semesters. English is still one of language that is not be mastered by the student of English Education Department major. Speech error still appears in student speech production, especially in the speaking class. During the learning of daily communication and public speaking, speech error is becoming a common that made by them. The writer can see that the student made some error in their speech production, such as hesitation, repeats, pauses and even slip of tongue. It can be supposed that the lack of knowledge and anxiety can be influenced by the student's erroneous in speaking. Therefore, the writer is going to conduct the research on error analysis on speech productionthe purpose is to describe the student's erroneous in speech production. In addition, there are three kinds of previous research related with this research. There are similarities and differences from the previous research. Here are the previous researches, first has conducted by Hidayati (2011) "Error Analysis on a Short Speech: a Case of an ESL Indonesian Learner". The aim of this research to analyze the errors produced by an Indonesian learner in speaking in a given short speech task. This study is a case study of a learner of English as a foreign language middle learning these languages in Australia. The data of this research is Speech error made by ESL Indonesian Learner. Short speech and interview were recorded as data collection. And the data analysis is the recording was transcribed for the purpose of analysis. The analysis first focused on the pronunciation errors, and then morphological and syntactic errors were analyzed. Each type of errors was listed and presented in tables for easy reading and analysis. The lists of the errors are coded following the categories of errors proposed by Brown (2000): addition, omission, substitution. Second research has conducted by Wijayanti (2012) "An Analysis of Speech Errors in A Talk Show Program Of Metro TV Face to Face With Desi Anwar Broadcasted in January To June 2012". This article explores the types of speech errors, the frequency of each type of speech error, the dominant of speech error, and the sources of speech errors in the talk show program of Metro TV Face to Face with Desi Anwar broadcasted in January to June 2012. The findings of this research shows that there are 253 utterances consist of 428 speech errors which are gained from 9 types of speech errors based on Clark and Eve, Gleason and Ratner, and Poulisse theory, the frequency of the speech errors, it can be described that the most dominant error is filled pause, the speech errors are mostly caused by three sources; they are cognitive difficulty, situational anxiety, and social reasons. During their stay at a university, students are expected to write answers on exams using paragraphs and complete essays as well. They are also required to carry out various written activities, such as field and/or lab reports, senior essays orfinal year projects. When these students write, they face a variety of problems. One of such problems is committinglinguistic errors which adversely affect the structure of their sentences and the idea they want to communicate. Brown (2007) stated that making mistakes [errors in writing] is a natural process of learning and must be considered as part of cognition. Learners' errors, of course, give insight to the teacher about the learners' difficulty in their learning and therefore they are considered indispensable in learning teaching process. Thus learners' errors must be studied systematically and appropriately analyzed in order to give effective remedial. Analyzing learners' errors, in general, has two fold advantages: Firstly, it gives a good understanding of the nature and types of errors so as to devise appropriate ways to avoid them (pedagogical advantage); Secondly, it provides an insight about the process of second language acquisition, for the study of learners' errors is part of the systematic study of the learners' language (Theoretical advantage), (Corder, 1981). These two significances of error analysis, therefore, are absolutely essential to make wellfounded proposals for the development and improvement of the materials and techniques of language teaching in general and writing skills teaching in particular. To enable students, avoid such errors and construct grammatically well-formed and meaningful sentences, our responsibility is to systematically study such errors and bring to the attention of material developers and curriculum designers as Lightbown and Spada (2006) have indicated. Thus, the major purpose of this paper is to study the nature, type and magnitude of the errors that AMU students commit when they write paragraphs and to provide a means of avoiding those errors. The third previous research has been done by Hojati (2013) attempts An Investigation of Errors in the Oral Performance of Advanced-level Iranian EFL Students. The goals of this research are to find the frequently-committed errors in the oral performance of the participants, to find the most frequently-committed errors of the participants in categories of vocabulary, grammar and pronunciation, to find the highfrequency errors of the participants be interpreted and qualitatively explained. The findings illustrate that, contrary to what might be assumed, advanced-level learners commit numerous errors in all the foregoing categories, especially in pronunciation and grammar. From the previous study above there are similarities and differences with the current research that will be conducted by the writer. The similarity of the current research with previous research is about errors on speech production. Well, the differences are the types of errors that committed by the target people. The current research focuses on nine types of speech errors by Clark and Clark to analyze the data.

METHOD
The type of the research is qualitative research. The subjects of the study are the first and the third semester students of English Education Department UII in academic year 2016/2017, containing of 10 students for the first semester and 10 students for the third semester. The objects of the study are the errors made by the first and the third semester students of English Education Department in academic year 2016/2017. The data of this research are in the type of speech production consist of errors utterances taken from transcription of audio recording. The researcher takes the data from the audio recording of speech production in the class of English education department UII, especially for the first and the third semesters. The technique of collecting the data is observation, documentation and in-depth interview. In the technique of analyzing the data, the writer adapted theory from Miles and Huberman (1994:10) analyzing data refers to three concurrent flows of activity: data reduction, data display and conclusion (Bazeley, 2013).

FINDING AND DISCUSSION
Error Analysis (EA) Richards & Schmidt (2002) defined EA as a technique for identifying, classifying and systematically interpreting the unacceptable forms of a language in the production data of someone learning either a second or foreign language. Such systematic analysis of errors eventually provides useful insights about the system operating in the learners' mind and reveals the learners' knowledge about the grammatical systems of the target language. By identifying what is exactly lacking in the learners' competence, EA brings the problem areas to the attention of teachers, syllabus designers and textbook writers, and suggests remedial action. EA is usually operated on the production data of language learners (compositions, speeches, etc.), and any EA activity entails the following procedures (Ellis, 1985).
1. Defining a corpus of language 2. Identifying errors in the corpus 3. Description of the errors 4. Explaining the errors Defining a corpus of language: This step involves collecting and defining a set of utterances produced by L2 learners. Error identification: Ellis (1997) claims that comparing the sentences learners produce with what the normal or 'correct' sentences in the target language, which correspond with them enable us to identify errors. This process involves "…a comparison between what the learner has produced and what a native speaker counterpart would produce in the same context", (Ellis & Barkhuizen, 2005, p.58). Errors are those sentences which are ill-formed grammatically or wellformed grammatically but inappropriate for a particular context. After identifying the erroneous utterance, it will be possible to compare the reconstruction with the original erroneous utterance and then we can describe the differences in terms of the grammar of the target language.
Describing errors: The description procedure involves specifying how the forms produced by the learner differ from those produced by the learner's native speaker counterparts in the same context. The most useful taxonomies for error descriptions are linguistic taxonomy, surface structure taxonomy, communicative effect taxonomy, and comparative analysis taxonomy. The following discussion of error description taxonomies is based on the presentation given in Dulay et al., (Dulay et al. 1982: 150-163). 1. Linguistic taxonomy: It operates on the basis of the linguistic component (phonology/orthography, grammar, semantics, lexicon, and discourse) that is affected by an error. This taxonomy improves teaching since it uses wellestablished grammatical categories which are utilized to organize language lessons in textbooks and workbooks. 2. Surface structure taxonomy: This taxonomy works on mechanisms in which surface forms are modified or altered in erroneous utterances. There are four main ways in which learners alter target forms.
Addition errors: such errors refer to the presence of an element or form which must not appear in a well-formed utterance. Addition errors are sub-categorized into: regularization i.e applying rules used to produce the regular ones to those exceptions to the rules; double-marking, a kind of addition error in which one feature is marked at two levels; simple additions are those which are neither regularizations nor doublemarkings. Omission errors: the absence of an item that must appear in a well-formed utterance. Mis ordering errors: caused by incorrect placement of a morpheme or group of morphemes in a given.
In this research is found type of errors that committed by the first and the third semester students. Silent pause is type of dominant error that committed by the first semester students. Silent pause means a period no speech between one word to another word (Fauziati, 2013 (19SPrdSMT) There are 81 utterances of similar categories from the first semester student and 29 utterances from the third semester students. The examples above shows that the first and the third semester students of English Education Department of Islamic University of Indonesia stopped for a while or no speech between one word to the other word because they got difficulties to find the vocabularies in their minds before producing the sound.
From six utterances above showed that the students got difficulty to find the next word that they are going to say. The first semester speakers are done with no speech between the words raja ampat and is, between I think and we, between one and the most, between the most and modern, between island and we, and the last between can and see. However, the third semester students are done with no sound between the words make and our, between such as and what and the last between is and more. The speakers took longer time to produce the next word, there are the result of silent pause. The speakers got difficulties to find the next right word to be executed, because they have not completely planned their utterances, therefore that cause them made the pause in uttering the utterances. In order to have the smooth and fluent speaking, the speaker should plan the utterances well before producing the sound. Lack of vocabularies can be the reason that the students think too hard to produce new word.
On the other hand, the writer conducted in depth interview with the students. The writer asked to the students, what is the reason that makes them stop between one word to another word in speaking. The students answered, nervous is become the reason that they usually think hard or stop too long before producing the next word.
In addition, filled pause is the dominant error committed by the third semester students. Filled pause is type of error happens when the speaker filled up the expression ah, er, uh, mm before expressing the next word (Fauziati, 2013). In this research, the writer found 51 utterances consist of 57 filled pauses that made by the first semester student and 89 utterances consist of 124 filled pauses that made by the third semester students.
Here are the examples of filled pause that made by the first and the third semester students of English Education Department of UII as follow: (1) ah [fp] Raja Ampat is a great island.
( There are 48 utterances having similar types of filled pause from the first semester and 86 utterances from the third semester students.
From the examples above, we can see that the first and the third semester students of UII are committing filled pause such as ah before producing raja ampat, ah before producing good evening and em before producing the Camp Nou, uh before producing don't, uh before producing good, uh before producing now and em before producing more. They created the speech error by committing of the word ah and em because they try to find the appropriate word will be expressed next. The students try to think hard about the next vocabulary.
The finding of the current research shows that the differences of dominant errors are committed by the first and the third semester students of English Education Department of Islamic University of Indonesia. Silent pause is the dominant error made by the first semester students, while filled pause is the dominant error made by the third semester students. The current finding corresponds with Wijayanti's (2012) finding of the types of speech error. The most dominant error Wijayanti's study is filled pause the same finding of the writer research for the third semester students. The speech errors are mostly caused by three sources; they are cognitive difficulty, situational anxiety, and social reasons. These types of speech error are common. It's suitable with Clark and Clark ,(1977: 263) as cited by Fauziati (2013) said that, the common type of speech errors as follows: Silent Pause; A period of no speech between words, such as turn on the // heater switch. Filled Pause: A gap filled by ah, er, uh, mm, such as in turn on, uh, the heater witch (Fauziati, 2011).
The other reason can be a factor to the differences of the common dominant error made by the first semester and the third semester is the cognitive demanding and less cognitively demanding. The cognitive demanding belong to the third semester. The third semester students felt self-assurance when conveying the presentation in the classroom. They felt enjoy to produce the sound. While, the first semester students are less confident to convey the presentation in the classroom. They felt anxious to express the wrong word and sentence, they used to keep silent. Therefore, there are few errors committed by the first semester, because they afraid of uttering the sentence.
The current research finding demonstrates that the speech errors are frequently caused by two sources; they are cognitive difficulty and situational anxiety. The cognitive difficulty consists of lack of vocabulary and lack of grammar mastery, while the situational anxiety consists of nervous and hesitation. The present research concentrates on speech error, which uses the theory from Clark and Clark, therefore the current research finding correspond with Wijayanti's research. These types of errors are frequent, because between current research and Wijayanti's research match with the theory of Clark and Clark.
The finding is proper with Clark and Clark theory, the finding of this research only found two sources, while Clark and Clark theory of source of speech error has three types. According to Clark and Clark (1997) there are three possible sources of planning difficulty are cognitive reasons, anxiety, and social reasons. Cognitive difficulty makes people need a longer time to produce sentences. Anxiety is when people anxious, they become tense, and their planning and execution of speech become less well-organized. And the last is Social Factor, for example speech plan appears difficult when conversation gets place under force (Fauziati, 2013).

CONCLUSION
The most dominant error made by the first semester students is silent pauses with the total number of error 108 and the percentage 36.73%. while the most dominant error made by the third semester students is filled pause with the total number of error 124 with the percentage 38.50%. These two semesters have different dominant of error made by them. We can point that the first semester students got trouble to execute and plan the utterances before producing the sound. Therefore, the first semester students common filled no word between the other words. They kept silent and start to find the next vocabularies. While the third semester students used to feel nervous Therefore, they used to filled em eh uh between one word to the other words. They also tried to find the next vocabulary.
Both the quantitative and qualitative analyzes revealed that the core components of the English language (morphology and syntax) are hugely affected by errors in the learners' compositions. The analysis also showed that morphological errors are the most pervasive in learners' written productions. The other notoriously difficult area for learners is the right ordering of words to produce well-formed utterances (syntax). Errors in core grammar of a language negatively affect both the forms and meanings of utterances. Such errors are observed in all the sampled students of AMU though they are acute in CNS, CSSH, AMIT & CBE. Omission is the most persistent error type followed by addition errors. Although grammatical morphemes are more frequently omitted, a significant amount of content morphemes has also been omitted. The most disruptive of the mis formation errors is the use of erroneous lexical items. Almost all of the lexical mis formations distort the meanings that learners intended to convey in their compositions. Besides, errors in word order are manifested in misplacement of verbs, objects, adverbs, and modifiers in one hand and using passive constructions for active or vice versa and wrong cleft sentence formation on the other hand. This paper is an indicative of learners' errors are systematic and regular in the sense that their addition, omission, mis formation and mis ordering of grammatical items reveal that learners are employing some strategies, such as overgeneralization, undergeneralization, or incomplete application of rules in learning the different aspects of English. The interplay of intralingual and interlingual factors triggered learners' errors. The majority of the errors in this study are attributed to intralingual factors. L1 induced errors, which are restricted only to the direct translation of Amharic words and sentences into the target language (English), borrowing, code-mixing and switching, have also been sorted out.