Language Learning

The Process of Learning Second or Foreign Language:

Contrastive Analysis, Error Analysis, Performance Analysis, and Discourse Analysis

Introduction

In the process of learning a second or foreign language, L1 has complex roles which can lead to errors. Many researchers have tried to get better understanding of the errors made by the L2 learners by having different types of data analysis.  Over the past several decades there are 4 types of data analysis which were used to come to the conclusion of language acquisition process. They are contrastie analysis (CA), error analysis (EA), performance analysis, and discourse analysis. Their contribution to language learning theories is valuable. They give knowledge on how human acquire language and how second or foreign language learner process the language. In acquiring this second language the L1 of the learners give contribution to the language acquisition whether it is positive or negative. Besides, the error made by the language learners can be seen differently for each types of data analysis. It can be negative as in contrastive analysis or can be positif as in error analysis. In this analytical paper, there will be an elaboration in what ways each types of data analysis differ and in what ways they can be complementary to the others.

Contrastive Analysis

Contrastive analysis is the study about a pair of language which can be used to identify the similary and the differences structurally between the two languages. The focus of this analysis is on the surface form of the first and second language system. It tries to compare the forms of two languages and then describe them. By comparing and describing the surface form of first and second language system, it is expected that the area of difficulty encountered by the learners can be identified. After it is identified, the solution of this problem difficulty can be tackled.

The statement from Fries (1945) about the effective material that should be based on the scientific description of the language to be learned inspires a number of constrastive analysis. “the most efficient materials are those that are based upon a scientific description of the language to be learned, carefully compared with a parallel description of the native language of the learners” cited in Larsen (1991, p.52). Because there is a belief that error in second language is an evil sign of deficiancy, the materials are designed to make sure that the learners will not perform an error. The concept of error in this contrastive analysis is different with the concept of error in error analysis. In error analysis, the error will be seen as necessary part of learning second or foreign language which then allow the learners to commit an error.

The errors which are commited by the learners related to language systems mostly are phonological error that reflect the nature of native language. According to Lanteigne (2006, p.1) the difficulty in learning English occur due to the fact that some of English sound do not exist in the mother tongue of the learners. For Indonesian language, the English sound of [ð], [Ɵ], [Ʒ], [dƷ], and [tʃ] do not exist in Indonesian. For the example it can be known from the example when Indonesian learners says Thursday. Instead of saying [Ɵɜ:rzdei] they will say [tɜ:rzdei].

Others error are from the syntactic and morphological levels. For the example the Indonesian learners who says “I not like it.” The English utterance contains the error, whose the souce of the error is from Indonesian language. There is no auxiliary verb ‘do’ in Indonesian language. The Indonesia language will be “Saya tidak suka dia.” So, once it is transfered to English  the learner will say “I not like it” instead of “I do not like it.” It is clear then that the difficulties in learning second or foreign language can be predicted by comparing the structure differences between two languages through constrastive analysis.

The examples of error as discussed previously bring to the idea that the first language of the language learners interfere the second language acquisition. The acquisition of second language habit is usually different from the habit of the first language which makes what we called as negative transfer. Negative transfer happens when the two languages are different which then rises language learning difficulty. It is in line with statement from Lado who says “individuals tend to transfer the forms and meanings and the distribution of forms and meanings of their native language and culture to the foreign language and culture” (1957, in Larsen 1991, p.53). So, there is a belief in constranstive analysis second language learners will tend to transfer their first language feature to their second language utterances.

As there is prediction for the second language learners difficulty, the difficult pattern can be emphasized in the classroom drillings of Audio-lingual method. This method is based on Behaviorist theory of language learning that views second language acquisition as the formation of new habit acquired through association, repetition and reinforcement. The activities can be in the form of reading aloud dialogues, repetition of model sentences, and drilling. Based on Richard, Platt, and Weber (1985) there are three types of drilling; mechanical drill, meaningful drill, and communicative drill. All of them has control for the responses and they are only different in the portion of the control. The example of drilling that can be done in the classroom is the teacher shows the picture of dictionary while she/he is saying “This is a dictionary” and then shows another picture for the students to say what it is with the right structure. Another example of drilling that can be applied in the classroom is by letting the students to response based on the correct understanding. When the teacher says “It’s hot here” the students give response by saying “I will turn on the aircon.” So, in this case, there is a repetition and reinforcement of the correct response which place the language learners as imitator of the teacher’s model.

The formation of new habit through repetition and reinforcement can help the learners to make sentence pattern automatically but it also can make the language learners has limitation in the creativity since the responses are controlled. It also ignores the role of context and knowlege in learning language. It then falls to the assumption that second language acquisition is viewed as passive mechanical process rather than as active mental process. If it is the case, it can be compared to the training of animal where the animals can perform certain task by using appropriate conditioning technique. It means that the treatment of repetition and reinforcement to format new  habit can not offer the improvement in communicative ability that is essentially needed by the language learners.

Not only can not offer communicative ability in language learning with its habit formation view, constrastive analysis also can not explain the unpredictable types of error such as overgeneralization or simplification. The first language transfer alone is not able to explain about the errors made by second language learners. There is also one condition where the errors made by the language learners are also happened to the first language learner, especially the young learner. For the example the young language learners who their first language is English still make mistake in the use of irregular verb. Instead of saying “I made this cake” they will say “I maked this cake.” As there are many aspect of language errors that can not be explained by contrastive analysis itself, error analysis has different explanation of it.

Error Analysis

Error analysis is different from constrastive analysis because the focus is not in predicting the error but focus in discovering and describing the error made by second language learners. According to Hakuta (1977) “the goals of error analysis are twofold: to describe, through the evidence contained in errors, the nature of the interlanguage in its developmental stages and to infer from these descriptions the process of second-language acquisition.” For this goal, the errors that the language learners made will be classified to different types. After the errors are classified, the information then will be described to build up the pictures of language which can cause language learning problems.

There was a research about the errors made the learners by Dulay and Burt (1973) in Hakuta (1977) where the finding could against the argument from constrastive analysis. They classify error as interference, ibtralingual, and unique. It was found in this research that the contribution of interference to the learners’ error was only 5 percent, and the 87 percent of error was intralingual. It could mean that the interference of learner first language only had little influence on second language learning. Whereas the constrastive analysis relies on the belief that first language tends to interfere the second language. But still, the view of constrastive analysis can be incorporated to Error analysis approach for future studies in order to get complete picture of the learners performance.

From the study of the learners error, there will be information about the nature of language learners knowledge in the target language. By having this information, the area that still needs improvement can be tracked. The area of difficulty can be interlingual or intralingual. If interlingual is caused by the first language interferene, intralingual is caused by the overgeneralization or simplification of particular grammar rules. It can not be explained in constrastive analysis but it can be explained in error analysis. According to Richard (1971) overgeneralization can be caused by the learners’ failure to observe the boundaries of a rule. In this case the learners tend to extend the use of grammatical rules beyond its accepted uses. For the example the language learners say “I don’t have any idea why did he came to my house” instead of saying “I don’t have any idea why he came to my house.”  It is clear from the example that the learners has generalized the rule of subject-verb inversion in indirect question when it is not necessary. On the other hand, in simplification the language learners simplify the second language system so that the learning difficulties of the second langauge will be decreased. For the example the learners will say “This country has unique culture which is not same in other countirs” instead of saying “This country has unique culture which is not the same as those in other countries.” So in those two sentences there is simplification in relative clause.

Those errors made by second language learners suppose to not viewed as bad habit. It can be viewed as strategy of the second language learners to acquire the second language and to communicate a desired concept as a term of communication-based error from Selinker (1972) in Freeman (1991) where the errors resulted when the speakers invoked communicative strategy. So, in error analysis, the errors can be seen as positive sign that the learners engage in communication actively rather than be seen as something to be prevented.

According to Corder (1967) in Freeman (1991) the errors made by the second language learners are different from the mistake. Errors are considered to be systematic governed by rule. They appear because the knowledge of the language learners about the rule of the target language is incomplete. It then likely to occur repeatedly as they do not recognize it. That is why in solving this errors problem the language learners need teachers or resarchers to locae the errors made. Compared with errors, the mistakes are different. Mistakes are not related to any system. Mistakes are cause by non-linguistic factors such as memory limitation, fatigue, emotional strain, etc. Unlike the error that cannot be recognized by the learner themselves, the mistakes can be self corrected if they are brought to the language learners attention.

In the study of errors in language production error analysis has weakness as the learners attitude which avoid error in their language production can not be detected. There is possibility that the language learners will avoid to produce errors in their utterances by avoiding certain linguistic structures due to the differences between their native language and the target language or due to complexity of the stucture. For the learners with negative transfer who are predicted to produce more errors in their speech production may has possibility that they only produce less error in their production and vice versa. It can be caused by their avoidance in using complex and variety of sentence construction. This error analysis then has an effect to to the constrastive analysis because this avoidance, there is no evidence that first language interfere second language production. In the same position, error analysis also can not detect and describe this phenomenon since error analysis attempt not only to describe error but it describe the learners overall performance.

Performance Analysis

There are two types of performance analysis; morpheme studies and developmental sequence. Language learners’ performance can be described accurately by having quantitative analysis. Morpheme studies serve this analysis as grammatical morpheme include the article (a, the), the copula and auxiliary be, and the noun and the verb inflections.

Brown (1973) is the one who carried out the first study on grammatical morpheme with three children whose first language was english became the subject. He analyzed fourteen morphemes in data collected. They are  present progressive –ing, plural –s, irregular past tense, possessive –s, copula be, articles a/an and the, third person singular present tense –s, and auxiliary be. From the study there was conclusion that eventhough the children learned the morphemes at different rates and ages, the order in which they learned them was relatively the same.

Then, there was another study about morpheme acquisition. This time, it was conducted by Dulay and Burt (1974) with 300 Spanish speaker in the first study and 250 Spanish and Chinese speaker in the second study as subject of study. From the finding, there was indication that  the order of morpheme acquisition for second language learner was different from the order for the native speaker. Besides, the order was relatively the same between the group of Spanish speakers and the group of Spanish and Chinese speakers.

For the morpheme acquisition in a certain order there is tendency that it is determined by the frequency of occurence. So if for example the morpheme progressive –ing has high frequency in use, it will acquired early and vice versa. In addition the morpheme acquisition is also determine by the learners’ native language (Brown, 1973). So if Spanish learners are able to contrast in the use of artcle “a” and “the” due to their native language, the condition will be different with Chinese learners that have to learn about them before being able to discriminate the english article. In conclusion, the transfer from the native language and the frequency used have important factors in determining the acquisition of morpheme.

Discourse Analysis

If constrastive analysis focus on language system, error analysis focus on language construction, performance analysis focus on morpheme acquisition, discourse analysis will focus on language in social context. The theory of language will be different because in discourse analysis the sentence will be related to the socio-cultural context. Discourse analysis itself is based on the understanding that there is much more going on when people communicate than only about information transfer.

According to Stubbs (1983) discourse analysis refers to the study about the language use beyond the boundaries of a sentence/utterance. Besides, it concerns with the interrelationship between language and society, and concern with interactive properties of everyday communication. Another definition comes from Brown & Yule (1983) with the statement that discourse analysis is analysis of “language in use.” So it can be concluded that discourse anaysis is the study of the use of spoken or written language in a social context.

Discourse analysis is a qualitative method of analysis. In discourse analysis there might be questions about turn-taking in conversation, misunderstanding between speakers beacuse discourse analysis concern with the meaning produced by language use and communication. The aim of this discourse analysis is perceiving and categorizing various meaning-making processes, networks and practice from the data.

Conversational analysis is one sub-area of discourse analysis. Conversational analysis has an interest in talk-in-interaction. Specifically, conversational analysis focus on the spoken language.

Conclusion

There are several types of data analysis in second language acquisition; constrastive analysis, error analysis, performance analysis, and discourse analysis. Each of analysis has its strength and weaknesses. The weaknesses does not make the data analysis type can not be used for the future study. The weaknesses can be covered by other types analysis to form ccomplete description of future study. Constrastive analysis as the first types of data analysis which has behaviorist view proved to has weaknesses as it can not explain some errors made by the second language learners. Besides, many of the difficulties which is predicted by constrastive analysis do not show up in the performance of the learners. But until these days, this constrastive analysis still get attention as one of the component in language study. At the same point, another types of data analysis, error analysis can not just rely on its belief that by finding the error made by learners, the information about the area of language that need improvement can be discovered. It is because focusing only to the errors made can not give clear picture of language learners performance if there is no further studies about the sources of errors. In order to know how languge is process, performance analysis can be used. The use of this kind analysis data is also not enough to explain certain utterance related to the strategy of communication. Then, there is what is called as discourse analysis to study not only the language but also the context to get the meaning behind it. For the summary, the purpose of our study will determine types of data analysis used. The errors phenomenon in language can not be explain by only using one type of data analysis. One type of data analysis can also be complementary to the others.

REFERENCES

Brown, R. (1973). A First Language: The Early Stages. Cambridge, Mass.: Harvard University Press.

Dulay, H., & Burt, M. (1974). Errors and Strategies in Child Second Language Acquisition. TESOL Quarterly, 24, 37-53.

Fries, Charles. (1945). Teaching and Learning English as a Foreign Language. In Larsen-Freeman, Diane and Michael H. Long. (1991). An Introduction to Second Language Acquisition Research. New York: Longman.

Hakuta, Kenji and Herlinda Cancino. (1997). Trends in Second-Language Acquisition Research. Harvard Educational Review Vol.47 No.3 August, 294-315.

Lanteigne, B. (2006). Common, persistent errors in English by Brazilian Portugese speakers. TEFL Web Journal, 4(1). Retrieved May 20, 2016, from http://www.teflweb-j.org/v4n1/Brazilians.pd.

Larsen-Freeman, Diane and Michael H. Long. (1991) An Introduction to Second Language Acquisition Research. New York: Longman.

Richards, J. C. (1971). A Non-Constrastive Approach to Error Analysis. English Language Teaching Journal, 25, 204-219.

Stubbs, M. (1983). Discourse Analysis: The Sociolinguistic Analysis of Natural Language. Chicago: University of Chicago Press.

Assessment (Advantages and Disadvantages)

Multiple Choice

The advantages and disadvantages

a. Advantages of Multiple Choice

Multiple choice has many advantages. One of them is its practicallity. It can be easily and quickly scored. “By far the most popular method of testing a reading knowledge of vocabulary and grammar is the multiple choice format, mainly for reasons of practicality. It is easy to administer and can be scored quickly. The most straightforward multiple-choice items may have little context, but might serve as a vocabulary or grammar check.” (Brown, 2003, hal. 194). In line with Brown (Steven J. Burton, 1991) says that Multiple-choice items can be used for many different areas of subject matter because they can adapt the various level of learning outcomes from the simple one to the complex one.

Furthermore, Osteroof stated others advantages of multiple choice items. “Multiple-choice items have four basic advantages over some of the other formats. First, they often provide a more adequate sampling of content. Second, these items tend to structure the problem to be addressed more effectively. Third, they can be quickly scored. Forth, responses to multiple choice items are objectively scored.” (Oosteroof, 2003, hal. 85)

For the detail of these advantages, it will be explained below.

1). Multiple-choice items allow a test to obtain a more adequate sampling of content.

There are two reason why multiple choice items are able to provide a more adequate sampling of content. Firstly, multiple choice can involve many more items than in essay. It is because of the time needed by the students for recording the response is lesser than in essay test. In addition, more content in multiple choice item can be sampled than content in essay. “Therefore, more content can be sampled by using the multiple-choice rather than the essay format” (Oosteroof, 2003, hal. 86).

Secondly, multiple choice item can measure procedural knowledge. Actually essay item is also able to measure procedural knowledge. So is short answer. But, if it is compared, multiple choice has more advantage than short answer test in the limitation of the concept knowledge. There are many things that students can do in multiple choice item such as identifyingan example of a particular concept by giving an options which include examples and non-examples of the concept, demonstarting the ability in classifying illustration related to the concept which is done by identifying the appropriate response, and applying the rule by including correct and common incorrect solutions resulting from the application of the rule. For those things, short answer test is not able to do it because of the restriction of the response which only consists of two words or a short sentence.

2). Multiple choice items tend to structure the problem to be addressed more effectively.

In multiple choice item, the responses help to define the problem being addressed. In this types of item, the items make the context clearer so that the problem is easier to be formated. “The context provided by the options makes it easier to structure a problem with a multiple-choice item than with other writen formats” ( (Oosteroof, 2003).

3). Multiple choice items can be quickly scored.

Scoring the multiple choice items is very fast since the respond of the students is in the single mark form. It is even easier when the machine score the test with the stipulation that the students responses are in separate answer sheet. “When responses are marked on a separate answer sheet, tests can be scored by machine” (Oosteroof, 2003).

4). Responses to multiple-choice items are objectively scored.

Multiple choice items have objective score. It is because multiple choice items are consistent in scoring. “Inconsistency in scoring is negligible with multiple-choice items” (Oosteroof, 2003). So, the test can be scored by different people such as students, teacher, aides, or machine as long as that the asnwer key is provided.

b. Disadvantages of Multiple Choice

“Multiple-choice items have three limitations we will address. First, these items are somewhat susceptible to guessing. Second, multiple-choice items usually must indirectly measure targetd behaviours. Third, multiple-choice items are time consuming to construct” (Oosteroof, 2003).

1). Multiple-choice items are somewhat susceptible to guessing.

In multiple choice items, there is possibility to get correct answer by guessing. The possibility of getting the correct answer by guessing will improve when the students are able to eliminate some of the distractors in a test. The result of this elimination is stated by Oosteroof (2003) that “the more distractors the students can eliminate in a test, the higher the test score.” So, the students who can avoid to choose the distractor will be able to get the higher score. This is what makes the negative effect of guessing in multiple choice items happened. Guessing in a test will affect the generalizability of performance in a negative way.

2). Multiple-choice items often must indirectly measure targeted behaviours.

It is impossible to know what the students are thinking whether they really get the answer or only guess the possible answer. “As with all test formats, multiple-choice items do not allow teachers to observe directly what the students was thinking or why students selected particular wrong answers” (Oosteroof, 2003, hal. 88). So, the result of the test sometimes does not really reflect the students ability.

3). Multiple-choice items are time-consuming.

Multiple choice items can be time-consuming  in order to have effective alternatives within each item. For other formats it will need lesser time than multiple choice items. “More time is required to build a test with multiple-choice items than is required with any other written format” (Oosteroof, 2003).