Investigating Lexical Errors and Their Effect on University Students’ Written Performance in Sudan Dr. Mohamed Satti Hamad

Investigating Lexical Errors and Their Effect on University Students’ Written Performance in Sudan

Dr. Mohamed Satti Hamad     

 ABSTRACT

The aim of this study is to investigate the lexical errors and their effect on university students’ written performance. The researcher used the descriptive analytical approach. Data has been collected through a questionnaire for university English language teachers, and a composition test for the university students from different English departments. The findings showed that university students make lexical     errors because of many factors chief among them is the interference of the mother tongue.

Keywords: Lexical errors – Written performance- Mother tongue interference 

Introduction

Knowing a word implies knowing the limitations imposed on writing and comprehending the word according to variation of function and situation; the network of associations between the word and other words in the language; the different meanings associated with the word and the semantic value of the word. There is a general consensus among researchers on the crucial role of lexis in language learning and teaching. Lexical knowledge in second language is of great importance to the development of second language proficiency. It is through writing people express much of their knowledge and understanding of the different subject areas. The researcher is inclined to believe that writing often forms the basis of   discussions or homework tasks, (Lyon, 1968). Because of this it is essential for learners to minimize, if not eradicate lexical errors in order to convey their ideas. Otherwise, no communication would be established at neither the level of the classroom, nor the society, (Parkin, 1993).  Furthermore, lexical knowledge is important to academic settings, since it is critical to effective writing. Lexical knowledge is also important to communication between people. If foreign language learners are able to choose the right word and write foreign language without difficulties, they may succeed in conveying and sharing their ideas. However, many researchers e.g. (Carter (1987) & Ellis (1997) argued that vocabulary acquisition research was neglected and the emphasis was primarily on grammar and phonology. Such claim is supported by Musa, (1995), Carter and McCarthy,(1988) argued that the research in the 1970s and early 1980s have been under strong criticism for the little attention it paid to lexical acquisition.  Moreover, Krashen, (1981) goes further as he argues that vocabulary is restricted in order to focus on syntax to the extent of claiming that vocabulary learning has been a victim of discrimination. So the researcher was encouraged to carry out a study on this neglected area .The study problem arises from the assumption that students in Sudan encounter lexical problems in their written performance. This can clearly be seen in words written in specific context.  Sudanese university students’ weakness appears in their inability to choose the right word and convey new ideas and establish a base for easy communication in writing, (Karadawi, 1994). Students’ deteriorating performance has been associated with lexical errors and their inability to construct meaningful sentences. Al Busairi, (2002) argues that most of the university students can hardly write a brief paragraph in simple English and even one in four is unable to construct an elementary sentence. This is natural since students might have never had the chance to practice writing except in final exams.  University students as second language learners of English are also faced with many Linguistic problems, especially the ones associated with the writing skill. The researcher noticed that most of these students' writing often lacks the basic sentence structure and accuracy in grammar, as well as spelling and punctuation. Lexical error is one of the major issues that we, as English language teachers witness in almost every lesson. Therefore, this study will try to investigate this problem through a sample from the third level, university students majoring in English in Sudanese universities. 

This study attempts to answer the following questions:  

1. What are the types of lexical errors made by the Sudanese university students?

2. What are the causes and sources of the errors made by the university students?

For the purpose of investigating these study questions, the following hypotheses are formulated:

1. Word choice, transliteration, omission and redundancy might be some of the types of the lexical errors committed by university students.

 2. The causes of the university students’ lexical errors might be attributed to the interference of the mother tongue.

The Objective of the Study:

The primary goal of this study is to find evidence to support whether university students’ written performance is associated with different types of lexical errors. It further tries to address their effect on the university students’ written performance.

Literature review:

One of the related studies on lexical error is the study of Karadawi (1994). He used the cross -sectional designquestionnaire, and teachers' opinionnaires in analyzing the Sudanese Higher Secondary School third year students’ composition. The study claimed that both inadequate and ineffective exposure to composition writing in the Higher Secondary School (HSC) is the reason for the inability of the final year students of the Higher Secondary School to write error-free types of texts. The impact of the mother interference was also claimed.  The study proved that the students today do not do any sort of private readings that help them acquire the skill of composing a readable text.      It could be observed that although Karadawi’s study succeeded in investigating the syntactic errors in terms of subject tense errors and lexical errors, however, he did not specifically shed enough light on lexical errors.   

Karadawi’s study argued that the EFL lexico- semantic handling becomes a problem to the learners to the extent that they get bound to no choice except a transfer from the first language Arabic. Similar studies about errors have generated varied results of which Gharab’s (1996) study which investigated the performance errors made by the first year Iraqi university students in written English.  Gharab  analyzed the errors using a free composition writing test as an elicitation technique for data collection. Interviews were also made to students, teachers and supervisors. The study found out that Iraqi university first year student make spelling, syntactic, and lexical errors in their writing due to the  transfer from the mother tongue.

With regards to the Gharab’s study, it investigated a wide range of orthographical, syntactic and lexical errors which resulted in the failure to concentrate on lexical errors.  In terms of Gharab(1996)  it could be argued that  errors made by the university students are not sufficiently addressed and tackled by the teachers. This might ring the bell and notify the syllabus designer all over Arab region to care more and tackle these errors in general and in particular teachers’ inefficiency.   Gharab’s findings proved the impact of the mother language,  lack of interaction and cultural factors on the students’ weakness. 

Both Gharab (1996) and Karadawi, (1994) used the composition writing as a tool employed in both studies. Gharab (1996) did not use the multiple choice question to investigate the students free- error writing ability which proves that the composition test is the best tool that would sufficiently express the students ideas and enable them to interact with society.Gharab (1996) proved that the types of the lexico- semantic error has become a problem to the learners to the extent that they get bound to no choice except a transfer from the first language Arabic. The study comes to the conclusion that MT interference, overgeneralization of TL are major variables in the choice of lexical items. 

             The study of Al–Boni’s (2004) investigated the types and frequency of errors at the Faculty of Arts. The study has employed the descriptive and inductive approach. Two tests were used to collect the data, oral test and written tests. The results of the study confirmed that the students improved relatively in both written and oral production.  Also the results indicated that Arabic interference was most visible in the students’ tests. However, the study did not shed enough light on other causes of the university students’ lexical errors as the improper choice of lexical items in both oral and written communication may, more often than not, lead to a breakdown in communication. Al–Boni’s referred to the interference of the mother tongue as the most obvious source of the error committed by the students as claimed by the current study.

            Arabi’s (1999) study investigated the writing performance among preliminary year students in three Sudanese universities. The investigation arrived at different lexical errors in the students' writing performance at level of the right lexical word, sentence connection, and structure. The composition scripts for preliminary year students were sampled and analyzed in order to predict and pre–estimate the main trends of errors’ distribution among the member of the data population. The problem of the study was the weakness of the students while the methodology was a free composition on one of four topics.  The errors of each group e.g. the right uses of lexical words and idioms were analyzed in relationship to whether they were due to intralingual. The study concentrates on the syntactic aspects.  

                        The study identified the major areas of weakness of the performance of writing and referred to the errors observed in the students’ compositions which are attributed to first Arabic language background, the complex nature of writing process and linguistic difficulty as setbacks factors in writing performance. 

            Al Noor Idris’s (2006) study investigated the problem of the university students. The methodology of   the study was a test and a questionnaire. Idris’s study investigated the problem faced by the Sudanese university students based on the hypothesis that the university level students are weak in producing proper English sentences. The findings of the study have pointed to the fact that even though the Sudanese university students of English have been learning English for more than eight years, they still can't manage to produce proper English sentences. Both studies unlike many previous researchers agreed that multiple choice tests were not appropriate tool to investigate lexical errors. It could be deduced that Idris’s (2006) study did not go beyond the syntactic level as it shed light on grammatical competence only. The researcher is of the opinion that even though there is no doubt that syntax has a role to play in the students’ writing, emphasis should also be placed on the semantic structure of the written sentences. Moreover, it is not always the case that the incorrect syntactic sentences can alone account for the overall writing quality as the native speaker might understand any error- syntactic sentences while it is difficult for him to comprehend lexical and idiomatic errors. E.g. if we compare * Ali went to school tomorrow with a) * look out, b) * blonde door, and c) the exact meaning of the word “vice”.  Idris’s study (2006) attributes the weak performance of writing to the learner's first language (Arabic) and he attributes the errors to a number of things chief among them are mother tongue interference, nature of the target language, overgeneralization and ignorance of rule restriction. 

                         Abdulla (2006) study investigated the weakness of fourth year university level students in writing composition. The materials of the research were originally written answers for the final examination. The ready–made instruments which were final department examination were used and the methodology of error identification and classification has been employed. The results of the study have shown that students' writing was characterized by a poor grasp of the properties of English written discourse ,i.e. discourse cohesion, discourse coherence, and discourse mechanic. The results have also shown that the students’ written performance was below average.  It could be argued that Abdulla’s (2000) study has concentrated much on the syntactic areas at the expense of the lexical areas.

                        Other researchers, however, reported considerable variations among individual teachers in their studies. For example, Angele Aziz Tadros (1996)  investigated what problems students aged 17-18 years encounter in the learning of English. The students' errors attributed to mother interference and lack of interaction and aptitude. Angele Aziz Tadros (1996) analyzed the errors using one set corresponding to the Sudan School Certificate English Language paper 11 and other free corresponding to the Sudan School Certificate English Language paper 1.  The answers were dealt with from the point of view of grammatical and lexical correctness as an elicitation technique for data collection. Errors were classified under grammar and lexis.  With regards to Angele Aziz Tadros (1996) no interviews were made to students, teachers and supervisors. Angele Aziz Tadros (1996) found that Sudanese students make spelling, syntactic, and lexical errors in their writing due to direct and indirect interference sources.

With regards to the current study, it could be argued that Angele Aziz Tadros (1996) investigated a wide range of syntactic and lexical errors which resulted in the failure to concentrate on lexical errors which is different from the current study as it concentrates on the investigation of lexical errors and the extent of EFL success in choosing the right lexical word.  According to Angele Aziz Tadros (1996) and the current study hypotheses that errors are not sufficiently tackled and addressed by the teachers, it could be argued that they agree on the assumption that more researches  are needed to tackle this issue which supports the prime importance of the significant of the current study . Such agreement might call for more researches not only across Sudan but all over the Arab region as claimed by (Gharab (1996) study to care much and tackle these errors not for the error-free texts only but for conveying our unique messages to the world. The current study argument in terms of the attribution of the university students’ lexical errors to the interference of the mother tongue, runs in line with Angele Aziz Tadros (1996)  findings as the latter has proved the impact of the direct and indirect interference of mother language.  Moreover both studies argue that the lexical errors made might be difficult when compared with the syntactic ones. This claim draws our attention to the importance of lexical rather than the syntactic areas.  

It could be concluded that all of Angele Aziz Tadros (1996)  (Gharab (1996) and (Karadawi, (1994) used the composition writing tool which goes in line with the current study in terms of the tool employed. In line with the current study, Angele Aziz Tadros (1996)   proved that the lexico- semantic error has become a problem to the learners to the extent that they get bound to no choice except a transfer from the first language Arabic e.g. (sell  versus buy) * They sell or buy the book. The subject here uses the word sell and buy interchangeably instead of differentiating between the right word in the right context. He gets a negative transfer from colloquial Arabic where sell (Yabia) and buy (Yashtari ) have the same meaning.  The study comes to the conclusion that MT interference, and overgeneralization of TL are major variables in the choice of lexical items which runs in line with the assumptions of the current study. Al - SadigYahyaAbd Allah (2000) investigated the composing competence of the Sudanese EFL learners who were preparing to graduate with a BA degree in English. The research similar to the current study aims to investigate the university students’ types and causes of lexical errors, centered on the analysis of linguistic competence in the students' writing. The study methodology was exposure to a variety of writing courses and a number of language and literature courses. The research is evaluative as it assessed the extent to which the students have benefited from their skills, linguistics and literature training in improving their composing skills. The study materials were originally written answers to final examinations at three academic institutions. The study has been conducted in order to know the sources of these errors and the reasons behind their continued occurrence year after year with different groups of learners.  The examination of written materials from three universities has revealed that learners' overall writing quality is below average even though they were preparing to graduate with a B.A in English. The findings have also shown that learners' writing is characterized by low grasp of textual cohesion, coherence and mechanics and if sufficiently addressed can collectively contribute to enhance their performance in writing. Al - SadigYahyaAbd Allah (2000) as well as the current and the previously mentioned studies call for conducting more researches on errors. However, no follow up research have been conducted to assess the outcome of these researches in terms of the rectification of these errors. That is to say these errors have been dealt with theoretically, however no practical solutions have been materialized. 

To sum up, findings from studies of university students' lexical errors indicate the following:

1) A number of factors such as the interference of the mother tongue and lack of interactions are hypothesized and proved to influence the students' written performance. In fact, one question needs to be answered: what is the effect of lexical error on university students' written performance? This study will provide an answer to this question.

2) University students' written performance is very weak.

3) Students switch from L2 to L1 during their writing process.

                                     Another study carried out by  Mohammad Hamad Al-Khresheh  (2001) investigated the interference of L1 (Arabic) syntactic structures on L2 (English) syntactic structures amongst Jordanian learners of English. The focus of the study was on the errors committed by EFL learners in using one syntactic category, namely word order within simple sentence structure.  A quantitative descriptive method was used to investigate the frequency and causes of interlingual errors committed.   The multiple-choice test unlike the current study was utilized in this study.  The findings revealed that interlingual errors committed by the same subjects were due to differences between the subjects’ L1 and the L2.  The results attribute the interlingual errors to the influence of their L1 and the errors. As  claimed by the current study this could be ascribed to the interference of the mother tongue.  The  type of errors reflects the learners’ inability to think in English. The interlingual errors  that are committed by the subjects are due to the transfer of L1 habits. The findings suggest that the subjects are still very much influenced by their L1 knowledge in understanding the English sentences.   The findings also brings to light the important fact about the varieties of the Arabic language that contribute to interlingual errors. In other words, whilst the differences between their L1 and English may make the process of acquiring the English language more difficult and complicated for the subjects of the study, the study also suggests that the cause of interlingual errors should not be attributed to the influence of a general Arabic language.  Instead, it should be acknowledged that interlingual errors of the subjects of the study are due to the influence of non-standard varieties of Arabic language as well as the standard variety.  As Mohammad Hamad Al-Khresheh  (2001) study offers new information about the importance of the differences between the two varieties of the subjects’ native language and the confusion happening while trying to speak and write in English, especially with respect to word order syntactic category, the current study looks at this defect employing  composition as a tool which reveals L2 ideas.          

         Furthermore, the current  study differs from Mohammad Hamad Al-Khresheh  (2001) field of study  as the first concentrates on written composition while the second on the restricted multiple choices.    But the latter is in agreement with the current  study's assumption which proved that EFL  Jordanian teachers should be aware of areas of similarities and differences between Arabic and English in order to avoid the areas of differences and enhance their teaching by using the areas which are similar. Finally, and based on the analysis of the errors committed in the current study. Al-Khresheh’s  suggested that both deductive and inductive teaching methods  are to  be adopted and taken into account to address the causes of interlingual interference in the performance of EFL learners.  Finally it could be argued that even though Mohammad Hamad Al-Khresheh  (2001) study  provided us with answers to the errors made by EFLs, it did not fully cover the influence of Standard  versus the Non-Standard Arabic structure within the Arab World. The researcher claim is based on the ramifications associated with the non- standard Arabic.

                         Saeed, Aziz Thabit Fareh, and Shehdeh (2006) study invesigated the semantic and syntactic constraints of the use of the synonymous verbs burglarize, rob and steal.  The study assumes that even though English language teachers find it easy to account for their students' morphological or syntactic errors, it is difficult for them to account these errors when they encounter improper choice of lexical items or collocations.  The study attributed the difficulty that foreign language learners encounter in acquiring the lexicon of the foreign language to the complex nature of lexical competence. The study also argued that Arabic-English dictionaries provide steal, rob and burglarize as equivalents to Arabic (saraqa) . Such provision as claimed by the study motivates learners to believe that these synonyms share the same syntax and semantics, and contribute to learners' incomplete knowledge of how these verbs behave in language use.  The research  data  was culled from Newspapers , books and references on crime and criminal law, widely circulated magazines, Electronic concordances and the British National Corpus (BNC).  It could be argued that the minute details of lexical usage and where the correct word should be correctly chosen and placed would not be collected from Multiple Choice Questions only. This is in line with the current study which attempts to investigate, analyze and classify the lexical errors committed.   The multiplicity of synonyms that Arabic-English dictionaries, for instance, usually provide for a single Arabic word may be confusing to the learners since they might erroneously assume that these synonyms are semantically and syntactically alike, which is not always the case.   As claimed by the current study a functional knowledge of a lexical item  involves knowing more than the definitional information or the literal meanings of the word that are stated in a dictionary entry. In fact, the components of the semantic competence are much more complicated than to be contained in a dictionary entry.  This support the researcher’s view which calls for extensive reading and writing  to tackle FLLs lexical errors associated with the expression of  an idea in L2.  It could be argued that Saeed, Aziz Thabit Fareh, and Shehdeh’s study paid our attention to the importance of the lexical competency and encourage both learners and linguists to take actions. The findings of the study are not restricted to only incorrect sentences, but goes further and provide numerous examples on three lexical words which  share the feature of referring to an illegal criminal activity that involves dispossessing someone or something of something else. Both studies have something in common as they try to tackle the lexical usage in debth.  This broad sense is probably why they are given as synonyms to each other in thesauri, monolingual and bilingual dictionaries as well. However, These lexical incorrect productions may indicate  or answer the current study question about the types of lexical errors made by the university students that these learners are only familiar with the definitional or broad characteristics of these verbs. The researcher goes further in addressing the causes and sources of the errors made by the university students.  Saeed, Aziz Thabit Fareh, and Shehdeh study found that the students know the general meanings, lack knowledge of these words in depth  and usually resort to bilingual dictionaries to help them find the equivalents of the words or expressions that they have in L1.  The causes of such errors  could also be attributed to mother tongue trace, misunderstanding of idioms, absence of strategies as the researcher claimed. It could be argued that Saeed, Aziz Thabit Fareh, and Shehdeh study unike the previous studies and in line with the resercher’s view extensively and deeply explored the semantic  constraints that govern the use of  verbs rob, steal and burgle/burglarize and the contextual factors in determining the choice of each of these verbs

                        Hussam Rajab study (2006) investigated Arab errors  made by Arab EFL learners when directly transferring idiomatic expressions from Arabic L1 into English L2.  The study sheds light on the importance of utilizing these errors by enhancing the student’s language production. The study assumed that errors  committed by  EFL students are often due to the interlanguage influence of their first language.  The findings reached were based on data collected over a three year period for a study that investigated semantic lexical errors committed by tertiary level male Saudi students. 200 university level preparatory year (male) Saudi students were randomly selected from 12 different sections over the period of three years. They were asked to perform three different oral examinations in addition to one written exam; all of which were part of their end-of-semester evaluation The participants were university students enrolled in a preparatory year program. The lexical errors made by the students were projections of the immediate literal translation of the Arabic words/phrases due to the interference of their interlanguage that incorrectly semantically expressed in English. Errors committed by the students were over 70% lexical.  It was found that students used Arabic language specific idioms and directly transferred them into English. It was also found that all of the idiomatic expressions that students directly transferred into English were reflective of Arab culture and environment. E.g.  “Time is like sword if you don’t cut it, it will cut you.” (“Time and tide wait for no man”). Out of the 70% lexical errors committed by the students, 25% were committed in the written exam. The study reflects on several common examples of interlanguage transfer errors in agreement with the current study as it attributes lexical errors to interlanguage transfer. Both studies call for understanding and addressing the errors committed and encourage  setting new approaches to correct them and map out ways to improve the students’ self monitoring and understanding of those lexical errors as well as how they can also learn from these mistakes. Moreover, the two studies agree that errors can provide a wealth of information and pedagogical implication to the EFL teacher. Where the use of direct transfer by the EFL learner could be a permanent not a transient reflection on  EFL learner’s culture.     Hussam Rajab study attributes the errors committed to interlanguage transfer only, while the current study attributes the same to mother tongue trace, misunderstanding of idioms, absence of strategies followed in dealing with such errors, lack of interaction, cultural, geographical and interlingual sources.

                         Nada AbiSamra (2003) investigated the sources of errors and interlingual errors, negative L1 transfer/interference and intralingual errors. The study identified Arabic speakers’ errors in English essay writing.   The methodology was a sample of written work collected from 10 students. Those students are in grade 9.  The total number of Transfer/Interlingual errors was 77, whereas the total number of Developmental/Intralingual errors was 137. 

 According to this study, the problems in essay writing were translation from Arabic, in addition to incomplete learning of essay writing rules and conventions.  It could be argued that Nada’s study calls for  teachers to deal with their students' errors in a manner which  encourage students to learn from their mistakes  which answers the current study question: What are the attitudes of the teachers towards the students' errors?  Like Nada study the current study assumes the university students make lexical errors of different types , Nada attributed them to translation from Arabic,  mother tongue in addition to incomplete learning of essay writing rules and conventions. Hence, we can say that both studies pay our attention to EFL errors and call for hard work in collaboration with grammarians and linguists to train them write properly. Also both studies apply the one topic written composition method. Nada study supports the researcher’s opinion that errors are considered important in explaining their mechanism, types and causes . In conclusion from the studies reviewed above it could be concluded that findings were very consistent and errors made by the students are not regarded as a stigma but rather a normal lexical error which provide evidence of the state of the linguistic development of any learner. This leads us to the conclusion that linguists in cooperation with students, teachers  and grammairians should address errors. Moreover, the findings of the above studies supported the researcher viewpoints and indicated that learners seem to rely on their mother tongue and use their L1lexical knowledge and made negative transfer. E.g. * Heavy tea instead of strong tea.    

                        In conclusion, researchers have found out that the impact of the mother tongue is reflected in the students' written performance. 

                        It could be argued that even though the above studies are in line with the current study’s assumptions and questions, the current study unlike the mentioned studies sheds light on the lexical errors only while most of the above studies address the syntactic and lexical errors. This is seen by the researcher as an advantage highlighted by Lyons, 1968 ,which encourages the researcher to carry out the current study.  

Methodology of the study:

The researcher used the Descriptive Analytical Method. The aim was to collect data objectively through composition and questionnaire then coding data collected numerically to describe relationship in a phenomenon The data of this study was collected by means of questionnaire and a test.  The researcher used two tools for data collection: the first is a questionnaire which was given to (67) teachers of English language from different universities in Sudan. The second tool was a test which was given to (150) students from different universities in Sudan.  

In this study, 150 participants, whose L1 was Arabic, were asked to write on one topic. The researcher chose the free essays because of the assumption that they are likely to give more reliable results. Further, according to Gharab's (1996:100), view where he agrees with Keshevaz,1994 argument that free composition test was found to be a suitable technique to elicit the data needed for the written study in comparison with other techniques where the learner’s attention is focused on the content rather than the form of what he wants to say or write. These participants were level 3 students majoring in English at different  Sudanese universities in spring 2010. They were male and female students.  Their average age was 19-20. The students in level 3 were required to write short essay (250 – 300 words) about Sudan.    Most, if not all, of the students have little or no exposure to English outside class. These students were familiar with these tasks in class. The students wrote on the assigned topic about Sudan.  Most of the university English teachers are experienced native Arabic speakers. They have taught at the university for more than 15 years.  This study focused on the students' errors, but the focus will be on the lexical errors that occur only due to translation or switching between L1 and L2.Based on observation during writing and researchers' L1 which is the same as the students' L1 i.e. Arabic, the data was analyzed using researchers' knowledge of L1 as well as simple statistics i.e. percentages to help quantify the data.

Validity and Reliability of the Instruments:

To assure the validity of the test designed for the samples of the population at university level, four copies of the test were distributed to four experts in universities and general education to give their evaluation and comments. Two of the experts consulted were Ph.D. holders in the status of associate professors of universities. The other two were M.A holders, one is the head of English Department and the other was senior English language instructor. The experts were told that the purpose of the test was to know the general standards of these students in specific aspects of language (written production). The consulted experts explained their views and suggestions which were taken into consideration by the researcher. Most of the experts approved the original content materials of the test.

Validity of the questionnaire

                        Copies of the questionnaire designed for experts and teachers of English language at university level were similarly distributed to five English languageexperts. The (5) experts were all Ph.D. holders working at different universities. The experts were requested to check the validity of the two tools and give their views and advice. They approved the original versions  and the necessary modifications were made according to their recommendations and remarks.

Statistical Reliability and Validity:

The reliability of any test, refers to obtaining the same results if the same measurement is used more than one time under the same conditions. In addition, the reliability means when a certain test was applied on a number of individuals and the marks of every one were counted; then the same test applied another time on the same group and the same marks were obtained; then we can describe this test as reliable. In addition, reliability is defined as the degree of the accuracy of the data that the test measures.     

      On the other hand, validity also is a measure used to identify the validity degree among the respondents according to their answers on certain criterion. The validity is counted by a number of methods, among them is the validity using the square root of the (reliability coefficient). The value of the reliability and the validity lies in the range between (0-1). The validity of the questionnaire is that the tool should measure the exact aim, which it has been designed for.  The researcher calculated the validity statistically using the following equation:                                                                                                              

     The researcher calculated the reliability coefficient for the measurement, which was used in the questionnaire using (split-half) method. This method stands on the principle of dividing the answers of the sample individuals into two parts, i.e. items of the odd numbers e.g. (1, 3, 5, ...) and answers of the even numbers e.g. (2,4,6 ...). Then Pearson correlation coefficient between the two parts is calculated. Finally, the (reliability coefficient) was calculated according to Spearman-Brown Equation as the following:                      

r = Pearson correlation coefficient                                                                                    

For calculating the validity and the reliability of the questionnaire from the above equation, the researcher distributed about (20) questionnaires to respondents. In addition, depending on the answers of the pre-test sample, the above Spearman-Brown equation was used to calculate the reliability coefficient using the split-half method; the results have been showed in the following table:

Table (1): The statistical reliability and validity of the pre-test sample about the study questionnaire

Reliability	Validity
0,66	0,82
0,60	0,77
0,57	0,75
0,86	0,93

We note from the results of above table that all reliability and validity coefficients for pre-test sample individuals about each questionnaire's theme, and for overall questionnaire, are greater than (50%), and some of them are nearest to one. This indicates the high validity and reliability of the answers, so, the study questionnaire is valid and reliable, and that will give correct and acceptable statistical analysis.

Results and discussion

The analysis of the experiment will focus on answering the vital question: What are the types of lexical errors made by the university students?

To answer the study's questions and check the  hypotheses, the researcher computed the median for each question from the results of the test. To do that, each error was separately marked and given score. This means, in accordance with the statistical analysis requirements, transformation of nominal variables to quantitative variables.   The aim of the above hypothesis which assumes that words choice might be one of the lexical error committed by university students is to show the types of lexical errors made by the university students.  As mentioned at the beginning of this paper, and as shown below in table (1) lexical errors were classified as: word choice, transliteration, omission, misspelling and redundancy. Table (2) below shows the number and percentage of errors per total number of words in the sample:

Table (2) Distribution of type of error

Type of Error	Grand number of error	Percent%
Word choice	1195	9,3
Transliteration	628	4,9
Omission	782	6,1
Misspelling	1077	8,3
Redundancy	291	2,3

It is clear from table (2) that there are (1195) error in word choice with (9.3 %). There are (628) errors in transliteration with percentage (4, 9 %),and (782) error in omission with percentage (6,1%), and (1077) error in misspelling with percentage (8.3%), while (291) error in redundancy with percentage (2.3%). A simple view of the error distribution above would suggest that word choice and misspelling are practically equal in impact on written composition. However, a comparison between the type of error and its impact on comprehensibility of the written piece yielded more stratified results. Comprehensibility was determined based on the presence of a central theme, relevant facts, supporting detail, and an orderly presentation to the English language native reader using a standard Six-Trait rubric. Score possibilities were from 0 to 5, with a score of 2 signifying emerging competency. The errors were calculated as a percentage per class against the total word count (y-axis), and averaged per comprehensibility band (x-axis)

Figure (1) Word choice error vs. Comprehensibility

In the above figure, word choice plotted against comprehensibility indicates that there is a clear and consistent correlation between these variables that begins at the emerging writer level and continues through the advanced writer level. From above results, we see that the first hypothesis that states “Word choice might be one of the lexical errors committed by university students' is fulfilled.

To answer the study's second questions: What are the causes and sources made by university students? And to check its second hypothesis that university students’ lexical errors may be partly attributed to the interference of the mother tongue, we computed the median for each question from the questionnaire that shows the opinions of the study respondents about investigating lexical errors and their effect on university students' written performance. To do that, five degrees were given for each answer "strongly agree", four degrees for each answer  "agree", three degrees for each answer  " undecided ", two degrees with each answer  "disagree", and one degree for each answer with "strongly disagree". This means, in accordance with the statistical analysis requirements, transformation of nominal variables to quantitative variables. After that, the non-parametric chi-square test was used to know if there are statistical differences amongst the respondents' answers about hypotheses questions.

To test this hypothesis, we must know the trend of respondents' opinions about each question from the hypothesis's questions, and for all questions. We compute the median, which is one of the central tendency measures, that uses to describe the phenomena, and it represents the centered answer for all respondents' answers after ascending or descending order for the answers.

Table no.(2)

The median of respondents’ answers about the questions of the second hypothesis

Result	Median	Question	No
Agree	4	University students’ lexical errors may be partly attributed to the interference of the mother tongue.	1

From table (2, it has shown that:

The calculated value of the median for the respondents’ answers of the 1st question is (4). This value means that, most of the respondents’ agree that “University students’ lexical errors may be partly attributed to the interference of the mother tongue”.

The above results do not mean that all the respondents in the sample have agreed with the questions because there are some respondents who disagreed with the questions. So, to test the statistical significance of the differences among the answers of the respondents for the first hypothesis, the chi-square test was used to indicate the differences for each question of the second hypothesis. Table no.(3) explains the results of the test for the questions as follows:

Table no.(3)

Chi-square test results for respondents’ answers about the questions of the second hypothesis

No	Questions	Degree of freedom	Chi-square value
1	University students’ lexical errors may be partly attributed to the interference of the mother tongue.	4	32.16

According to the table, we can demonstrate the results as follows:                     

The calculated value of chi-square for the significance of the differences for the respondents’ answers in the 1^st question was (32.16) which is greater than the tabulated value of chi-square at the degree of freedom (4) and the significant value level (1%) which was (13.28). This indicates that, there are statistically significant differences at the level (1%) among the answers of the respondents, which support the respondents who have agreed with the assumption that “University students’ lexical errors may be partly attributed to the interference of the mother tongue”.

From above results, we see that the second hypothesis that states “The causes of the university students’ lexical errors might be attributed to the interference is fulfilled

In conclusion this study investigated the lexical errors. It used the descriptive analytical approach. The study found out that university students make lexical errors of different types of which the word choice, transliteration and omission. These errors have a negative impact on their written performance. Based on these findings it has been found out that the student's committed lexical errors because of many factors chief among them the interference of the mother tongue.

Recommendations:

In the light of the findings of the study, it is recommended that:

1)      Generally speaking, in line with Anne (1982) and Formkin,s (1980) lexical errors  should not be discouraged as we can have instances of positive transfer. Rather syllabus designers and teachers should do their best to collate and benefit from the errors committed. Moreover, the current texts should be revised to include the correction of lexical error tasks that lead to the use of authentic language as it is used by native speakers.

2)      Most of the previous researches concentrated on the syntactic errors at the expense of the Lexical errors. In this regard, the researcher recommends that more researches on lexical errors should be carried out.

3)      An error is not a stigma. Such type of discussionshould be encouraged. This practice may benefit not only the students correcting negative transfer, but it may promote interaction among both teachers and students about similarities and differences between the two languages.

References

1.Abdallah , A (2006 ) Investigating Sudanese EFL learners' written discourse competence :The case of fourth year English students in some national universities. M.A  thesis in English language ,University of Khartoum

2.Ahmed, M. O. (1994) Acquision of English Lexis. A study of Habits and Tactics) Khartoum University press.

3. Al. Busairi, Mohamed, (2002) Situation of the English Language at the Tertiary Level. Sudannow Journal, April, PP.34-35

4. Alboni, G (2004) An Analysis of Syntactic Errors in Written and Oral Productions: A case study of students studying English at the faculty of arts, U of K.  M.A thesis, University of Khartoum.

5.Anne, Culter ED (1982) Slips of the Tongue and Language Production, Mouton Publishers, Amsterdam, Printed in Netherlands, 1982

6. Arabi, H (1999) An Investigation of Writing Performance among Preliminary –Year Students.   M.A thesis in English language, University of Khartoum.

7.Carter, R and McCarthy, M., (1988) Vocabulary and Language Teaching Longman Group UK Limited, 1988

8.Ellis, R., (1985 ) Understanding Second Language Acquisition Oxford University press.

9.  Formkin, V.A (1980) Errors in Linguistic Performance,: Slips of the tongue, ear, pen and Hand.Academic Press, INC,1980

10. Gharab, A (1996), An Analysis of the Written Performance of Iraqi Students at the University Level “Error Analysis”. A P.H.D thesis , University of Khartoum, Khartoum University Press.

11. Idris, A.  (2006) Syntactic Errors Analysis :(  A case of University students of English in The Sudan),Unpub, M.A dissertation ,Gezira University.

12. Karadawi, E. (1994). Deficiency of English Composition Writing in the Sudanese Final Third Year of the Higher Secondary School, Analysis and Corrective Actions. A  PH.D thesis, Department of English, Faculty of Education, University of Khartoum

13.Keshavarz, M. (1994). Contrastive Analysis and Error Analysis. Library International Islamic University Malaysia. First Edition 1993, Second Edition 1994, Rahnama Publications. Enghelab Ave. ShohadayeZhandarmerie St.

14. Krashen, S. (1981) Second Language Acquisition and Second Language Learning First printed edition 1981 by Oxford Pergamon Press Inc.

15.Lyons, J. (1968) Introduction to the Theoretical Linguistics, Cambridge University Press, 1968

16..Mahmoud , A. (2000) Modern standard Arabic vs Non-standard Arabic: Where do Arab students transfer from? Language, Culture and Curriculum, 13, 126-136.

17. Musa, A.T (1995) Lexical networks: A study of valences in the structure of vocabulary, unpublished M.A Thesis .Khartoum: University of Khartoum.

18.Parkin, A. J. (1993) Memory Phenomena, Experiment, and Theory Blackwell     publishers ,108 Cowley Road ,Oxford OX4  IJF  UK 1993

19.Tadros. A (1996) M.A thesis on “An Analysis of the Interference Errors in the Written English of Sudanese University of Khartoum,

20.Yahy,A (2000) Investigating Sudanese EFL Learners’ Written Discourse Competence: The Case of the Fourth Year English Students at Some National Universities. Ph.D Thesis, University of Khartoum.

Periodicals and Papers:

1. AbiSamra, N. (2003). An analysis of errors in Arabic speakers’ English writing. Retrieved May 9, 2009, from http://abisamr03.tripod.com/nada/ Languagwacq-erroranalysis.html.

2. Hussam Rajab study From My Two Eyes Teacher!: K.F. Uof Petroleum & Minerals which has been dedicted to investigate Arab errors retrieved from http://www.ksaalt.org/conference%20proceedings/YC_Hussam_Rajab.pdf

3..Mohammad Hamad Al-Khresheh  (2001) study on Interlingual Interference in the English Language Word Order : Structure of Jordanian EFL Learners publishd on European Journal of Social Sciences – Volume 16, Number 1 (2010)

4..Saeed, Aziz Thabit Fareh, and Shehdeh study on Some Contextual Considerations in the Use of Synonymous Verbs: The Case of Steal, Rob, and Burglarize . Publication: Studia Anglica Posnaniensia: international review of English Studies, Vol. 42, Annual