ABSTRACT

Proficiency and competence in English is a goal every teacher of the language strives to help his/her students achieve. English is a language of instruction as well as an examinable subject in Kenyan schools. As a language for international communication, it has a special place in the educational systems of most countries of the world. However, among non-English speakers or speakers of English as a second language (L₂), learners tend to code-switch back and forth between English and their first languages (L₁). This tendency has been viewed as a hindrance to students‟ mastery of the language. Research already conducted on the motivation for code switching and its effect on students‟ performance in English has not been conclusive. The KNEC examinations reports have also decried poor performance in English. Some candidates use mother tongue and even Kiswahili expressions in their essays. Reports from seminars conducted for teachers of English have also shown growing concern that students do code switching whenever  they write essays. A casual observation has shown students code switch in their out of class interactions, and in other activities such as symposia, debates and group discussions. The aim of this study was to establish the cause and effects of code switching on students‟ performance in oral and written English with a view to suggesting possible solutions to the adverse effects of code switching on proficiency in English. The samples of the study were seven out of seventeen secondary schools in Nyatike District, Migori County, Kenya. A total student sample of 112 was used. The researcher used descriptive survey design. The data was analyzed using descriptive statistical techniques. Howard Giles‟ Speech Accommodation Theory and Gumperz‟s Conversational Functions Theory constituted the theoretical framework employed in the study. The findings showed that the students code-switched at the intra-sentential more than inter- sentential level. Majority of the students switched from L₂ to L₁. Code-switching was influenced by the context and the school environment. Code switching affected the students‟ oral performance more than written performance in terms of syntactic, phonetic, prosodic and lexical error levels. The strategies used to navigate CS were mainly out-of class learning activities, language policy and teaching methodology. The study recommended that these strategies should be reinforced. Further research was recommended to establish the impact of language policy on students‟ English language proficiency, and the impact of teachers‟ professional qualification and experience on students‟ communicative competence.

CHAPTER ONE

INTRODUCTION

     Background

Code switching is the use of two or more linguistic varieties in the same conversation or interaction (Myre-Scotton and Ury, 1977). Poplack (2000) as quoted by Kim (2006) states that code switching is the alternation of two languages within a single discourse, sentence or constituent. In an educational context, code switching is defined as the practice of switching between a primary and a secondary language or discourse (Coffey, 2008). It is a common occurrence among the bilingual and multilingual communities in Africa, India, immigrants in Europe and America. Code Switching (CS) is far from homogeneous and the actual linguistic behavior involved varies depending on the sociolinguistic circumstances. As far as language learning is concerned, CS is a global problem. In multilingual communities, research has revealed that the use of two or more languages within a single discourse, sentence or constituent is widespread phenomenon that extends from daily life and workplaces (Ting, 2002; Ting, 2007) to classrooms in which specific languages have been instituted as the official languages of instruction (Ting and Then, 2009).

There has been a global outcry on poor linguistic performance in discourse by code- switchers who blame it on “a lapse of attention” (Gumperz, 1982). While Johansson et.al. (2004) postulates that CS entails a proficient knowledge of two or more languages, Theo et al. (1984), on the other hand, point out that CS is a window to linguistic weakness inherent in the code switchers and leads to negative transfer of L₁ into L₂. Students switch

codes due to lack of appropriate words in the L₂ discourse. Besides, the code switchers face the difficult task to effectively communicate with the native speakers of L₂.

In Kenyan learning institutions English is a language of instruction from standard four onwards and a teaching subject even as early as nursery school. Apart from making a student to be culturally and linguistically diverse, it also has some economic returns through the availability of job opportunities (Kimemia, 2001; Kimemia, 2002). Despite this merit, the bilingual student prefers to use L₁ as the language of social interaction and for group identity (Labov, 1972 as quoted by Milroy, 1987).

Labov (1972) views vernacular as socially functional and an important marker of group identity. Bourhis (1982) maintains that loyalty to African languages for use in informal settings seems to prevail in most African states. In a Kenyan school setting, however, students are required to use English and Kiswahili throughout hence they find themselves linguistically deficient and resort to direct transfer of lexical, phonological and syntactic structures from their L₁ into L₂ (Theo, 1984). Their oral and or written expressions sometimes have errors as a result of code switching. This dilemma prompts the need to research on the effects of CS on students‟ performance in English.

Secondary schools students in Kenya have language proficiency problems. KNEC examinations reports from 2006 to 2009 have captured the growing concern on the candidates‟ poor performance in English functional skills (English paper one) and imaginative compositions (English paper three). There are a number of factors which contribute to poor performance namely poor syllabus coverage, lack of reading culture and bad language policies. Code switching could be one of the factors under bad language

policies leading to poor performance. The table below shows an example of performance in English in the years 2006-2008.

Table 1 KNEC English Examinations Report for the years 2006, 2007, 2008

The performance in the three years indicates a consistent decline in paper 1 but a fluctuating performance in paper 2 and 3. Most candidates lost marks in the essay questions in paper 3 and functional writing in paper 1. The overall mean also showed a fluctuation. In the 2009 KCSE examination the percentage mean for the English composition paper dropped by 8.08 points from 40.48 in 2008 to 32.40 in the previous year, the lowest mean in four years under review (KNEC Report, 2008). In the same year, Nyatike District recorded in English a mean grade of 4.774 down from 4.834 from the year 2008.The best school had a mean of 7.477 while the worst had 2.813 (DEO‟s Report

on Nyatike District Education Day, 2010). The KNEC report on the KCSE examination results of 2009 indicates that some candidates have the habit of code-switching hence write essays with errors due to negative transfer of L₁ expressions into their imaginative compositions (KNEC Report, 2010). In so doing they write distorted and incomprehensible essays. In some sections of functional skills, tests on homophones and silent sounds pose serious challenges to the candidates. According to Theo et al. (1984), using audio lingual approach, intra-lingual and inter-lingual errors which L₂ users make would be attributed to interference from L₁. Clearly, non-English speakers are generally challenged since English is not their mother tongue. The study sought to establish whether CS leads to negative transfer of the syntactic and prosodic features of L₁ into L₂. Further, the study sought to establish whether report on the impact of CS on the students‟ performance on the KCSE examination would traced back to their oral and written discourse in school long before they do the final KCSE.