Classroom Bilingualism/Multilingualism: A Study on Students' Perspective at Intermediate Level

Abstract

Classroom discourse plays an important role for conceptual understanding and learning performance of the students at any level. Pakistan is a multilingual country with Urdu as a lingua franca. When medium of instruction (MOI) is changed from Urdu to English, in most of the classroom discourses, code-switching takes place. This classroom bilingualism or multilingualism has certain impact on students' learning performance. This study was conducted to elaborate this phenomenon at intermediate level. A mixed method research strategy was applied in the study. Sample size was selected through a multi-layered sampling technique and data was collected through a questionnaire. After thorough data analysis, it was found that bilingual/multilingual classroom discourse resulting after change of MOI has negative impact on students' learning performance.

Headnote

Abstract

Classroom discourse plays an important role for conceptual understanding and learning performance of the students at any level. Pakistan is a multilingual country with Urdu as a lingua franca. When medium of instruction (MOI) is changed from Urdu to English, in most of the classroom discourses, code-switching takes place. This classroom bilingualism or multilingualism has certain impact on students' learning performance. This study was conducted to elaborate this phenomenon at intermediate level. A mixed method research strategy was applied in the study. Sample size was selected through a multi-layered sampling technique and data was collected through a questionnaire. After thorough data analysis, it was found that bilingual/multilingual classroom discourse resulting after change of MOI has negative impact on students' learning performance.

Keywords: classroom bilingualism/multilingualism, intermediate level, medium of instruction.

Introduction

Pakistan is the territory of bilingual/multilingual speakers. People from the Muslim community face the challenge of almost four languages (mother tongue: Punjabi, Pashto, Sindhi, Balochi, etc.; national language: Urdu; foreign languages: English, and Arabic) in their religious and social interaction. Urdu and English are used as Medium of Instruction (MOI) (Li & Shum, 2008) in the education system public or private sector. Mostly in the English medium schools and colleges, code switching takes place almost at every level. This study is limited to intermediate level only. There might be different reasons for it. One of the reasons is bilingualism/multilingualism: Every student or teacher knows at least two or three languages (one is mother tongue, second national language Urdu and third foreign language or language of instruction (LOI) i.e. English in this case (Nisar & Ahmed, 2011).

Psycholinguists all over the world agree that the best option of language for academic purposes is mother tongue: students get clearer academic concepts (related to natural sciences, social sciences, humanities etc.) in their mother tongue (Tsui et al., 1999). Due to this reason bilingualism/multilingualism is resorted to even at graduation and post-graduation level. In Pakistani context at intermediate level science subjects are taught in English as MOI which results into code switching or code mixing in the classroom discourse (Gulzar, 2010). This practice of bilingualism/multilingualism in the classroom discourse has a definite impact on the learning performance of the students. The researcher has consulted the intermediate level students' opinion in this regard for due consideration by the educationists in public or private sector in policy on MOI in general and at intermediate level in particular.

Rationale

The main objective behind the choice of this study was to probe the consequences of bilingualism/multilingualism at intermediate level for the learning performance of the students. Classroom discourse plays a very significant role for developing conceptual understanding among the students at any level. Present study is a step forward for illuminating this phenomenon which may result into assistance for teachers, educationists and language policy makers, for understanding the case of the students by their o wn assessment in this regard.

Hypothesis

The following null hypothesis was meant to be tested in the present study:-

H0:There will be no significant difference of opinion among male and female students in respect of the impact of bilingualism/multilingualism in classroom discourse on their learning performance at intermediate level after change of MOI.

Literature Review

Pakistan is a multilingual country, where people speak multiple languages minor (Hindco, Saraiki, Gujrati, Kashmiri, Pothohari etc.) and major (Urdu, English, Punjabi, Pashto, Sindhi, Balochi etc.). In the present state of affairs, education system in Pakistan may be divided into three categories: monolingual education (pure English medium or Urdu medium all over Pakistan), bilingual education (both English and Urdu medium all over Pakistan) and multilingual education (Sindhi, Urdu and English as in the province of Sindh or Pashto, Urdu and English as in Khyber Pakhton Khawah or KPK) system. These are discussed in details in the following:

Monolingual Education System

An education system or program that is based on only one language as a medium of instruction is termed as monolingual education system. For example English medium, Urdu medium and Sindhi medium institutions in Pakistan. Due to multi-cultural, ethnic and linguistic groups in the population of a state or country sociolinguists and educationists favour bilingual and multilingual education. Svensson (2005) has presented the case of Helsinki University with respect to MOI from monolingual to bilingual and multilingual. It is evident from his presentation that the more people get closer to each other the more they become multilingual. There is no doubt that some languages especially English take favourable space being lingua franca but multilingualism is being favoured because of political, cultural, academic and social interaction at international level. Whereas Duenas (2004) asserts that high level of proficiency can be achieved in classrooms where the target language is medium of communication rather than an entity of study.

Walter (2000) has reported the result of a seminar on multilingual education in 1999 held at University of North Dakota Session which summarizes the investigative work of the seminar participants from ten countries, which suggests that countries where mother tongue is not developed and implemented, education programs will result into long-term disadvantage outweighing any short-term benefits that are achieved from using only one language of education or monolingual education.

Bilingual Education System

Bilingual education is defined as the use of two languages. The combination may be:

* foreign language and second or national language for example English and Urdu,

* foreign language and mother tongue for example English and Sindhi or Pashto,

* second language or national language and mother tongue for example Urdu and Sindhi or Pashto,

* mother tongue of one population and mother tongue of another population for example Urdu and Sindhi in the province of Sindh for Urdu speaking Mohajirs and Sindhis especially in the city of Karachi

Both the languages can be used as MOI for the same pupil population in a well-organized program with suitable proportion in accordance with the priorities of educational objectives. There is remarkable evidence in developing and developed countries showing the effectiveness of bilingual education. In Zambia, for example between English-language reading and writing scores of children under the bilingual education program comprising mother tongue(L1) and second language(L2) showed a surprising improvement over the scores of children in non-bilingual programs (L2 only).The reading and writing scores in the Zambian languages was also improved (Williams, 1998).

According to August & Hakuta, (1997) and Dutcher, (2004), studies recommend that bilingualism has great cognitive benefits. Instead of replacing one language with another, good bilingual education promotes harmonizing relationship between languages, and consequently proficiency in one language improves proficiency in another. Dutcher, (2004) further asserts that students in bilingual classrooms may perform better in subjects other than the language. For instance, a number of United States studies conclude that children in bilingual programs in upper elementary grades performed better in both reading and mathematics than children who are being taught in programs that demand immediate switch to English. More broadly, first-language study gives children and parents a sense of pride and accomplishment.

There are two ways of implementing bilingual education: two-way bilingual education and one-way bilingual education (UNESCO, 2005).

Two-way bilingual education

All students develop dual language proficiency by receiving instruction in both the languages; generally half of the students are native speakers of one language and half of the students are native speakers of the other language. All curricula may be taught in both languages (but in limited topics).

One-way Bilingual Education

Only one group learns bilingually. Bilingualism can be termed as having knowledge of two different languages. It does not mean that the person should be flawless in a given language; the point is to have reasonable knowledge of a language other than the mother tongue, Rasul (2006).

Models of Bilingual Education

In a debate on bilingual education held by University of Massachusetts in 2002 following models are discussed:

Transitional Bilingual Education

This model encompasses all Bilingual Education programs which aim to shift students to the majority language, help students assimilate to mainstream cultural norms, and incorporate students into the natio nal society. This is the most commonly used model in the United States.

Developmental or Maintenance Bilingual Education

This model encourages students to maintain their native language, strengthen their cultural identity, and affirm their civil rights in the national society.

Enrichment or Two-way Bilingual Education

This model supports the development of minority languages on the individual and collective levels, cultural pluralism at school and in the community, and an integrated national society based on the autonomy of cultural groups. This model is becoming increasingly common in Canada and the United States.

Multilingual Education System

Pakistan is a multilingual country. Its national language, Urdu, is the mother tongue of only 7.57 percent of the population though it is very widely used in the urban areas of the country. Pakistan's official language is still English as it was when the British ruled the country as part of British India. In addition to this, the country has five major indigenous languages (Rahman, 2003).

Apart from programming the use of several languages, MLE also involves the following:

* The development of good curriculum.

* The training of good teachers in the required languages, content and methodology.

* The production of good teaching materials (error-free and culturally relevant).

* The empowerment of the community (school-based management) (Walter, 2000).

MLE will not work when one simply changes the language by translating existing materials into the local languages (UNESCO, 2005). According to Nolasco (2009) MLE is recommended and preferred in comparison to other education systems due to following reasons:

Active participation

Use of the mother tongue or the first language (L1) in class enables the students to express themselves easily, as there is no fear of making mistakes. MLE encourages active input by children in the learning process because they comprehend what is being discussed and what is being asked from them. They can instantaneously use the L1 to build and explain their world, communicate their thoughts and add new concepts to what they previously discern (Schulter, 2006).MLE empowers the teachers as well, particularly when they are more fluent and proficient in the indigenous language than in the languages of MOI. Because the students can express themselves, the teachers can more precisely assess what has been learned, and discover areas where students need additional support (UNESCO, 2005).MLE creates the environment for the assimilation of the people's community knowledge, the knowledge that informs their lives and gives them meaning into the school system. According to Walter (2000) MLEmakes it possible for the community to produce its own ethnically applicable reading materials and teaching aids, together with the local writers, illustrators, cultural groups and other stakeholders in the community. MLE also empowers the parents who can take an active part in the education of their children because the school's and the community's language are also their language. MLE brings the community closer to the school and its programs.

Multi-dimensional Approach

MLE aims to produce learners who are:

* Multi-literate -they can read and write competently in the local language, the national language, and one or more languages of wider communication, such as English.

* Multi-lingual -they can use these languages in various situations.

* Multi-cultural -they can live and work harmoniously with people of different cultural backgrounds (Nolasco, 2009).

Change over from Informal to Academic Language

What we and our children know is the informal language or the everyday variety used for daily communication. Achievement in school depends on the academic and intellectualized language needed to discuss more intangible concepts. According to studies, it takes one to three years to learn the conversational language, but four to seven years to master the educational language under well-resourced environment. It also takes time to develop higher order thinking skills and this depends largely on cognitively demanding curriculum particularly from Grade 4 onwards (UNESCO, 2003).

Purgation of Inferiority Complex

According to Kangas (2000) if the indigenous languages are denied access in education, people lose their interest in local and regional languages. The resourceful population gradually joins the affluent and elite class by achieving proficiency in national and official languages. But those who do not have access to imperialist language and only know their indigenous language feel themselves handicapped in the walk of life as it is the case in Pakistan. People who do not know Urdu and English are failed to survive an honourable position in the society. MLE is solution for this kind of situation.

The use of the local and indigenous languages in education is not disadvantageous to building one nation. It suppresses the sentiments that may lead to vicious conflicts, disunity, and rebellion due to the refusal of regional and indigenous languages in the education system (UNESCO, 2005). In the early 1970s, the Bangladeshis fought and won a war against Pakistan over the issue of language. The Lithuanians protested the compulsory use of Russian in schools and later seceded from the former Soviet Union in the early 1990s (Nolasco, 2009). The sovereign government of Catalonia enacted the 1983 Catalonian Linguistic Normalization Law, which made Spanish and Catalan co-official languages in the region. This was an act made by the Catalonians against what they perceived to be the Spanish government's attempt at practicing ling uistic imperialism in their region (Hall, 2001 cited in Rahman, 2006).

Affordable and Economical

It is a popular belief worldwide that MLE is an unaffordable system of education. Contrary to this popular conviction, L1-based education may actually cost less than a system that is based on L2. If we reflect on the capital washed out on drop-outs, repeaters, and failures, as well as other added costs, studies show that L2 -based education systems are more costly than L1 systems. A Guatemalan study, for example, showed that it is more expensive to turn out a grade level passer (in Grades 1 to 6) in a Spanish medium school ($6,013) than in a Mayan school ($4,496) (Nolasco, 2009). In Mali, a World Bank study found that French-only programs cost about 8% less per year than mother-tongue schooling, but the total cost of educating a student through the six-year primary succession is about 27% more, largely because of the distinction in repetition and dropout rates (Patrinos, 1996). One problem often raised in discussions about MLE is that it is too expensive to produce instructional materials in so many languages. This may be true if materials are produced in full colour, silky pages. The successful MLE experiences in Papua New Guinea, Indonesia, Kenya, Ethiopia, Cameroon, and the Philippines specify that teacher -made cardboard covered books, with simple black-and-white drawings on plain paper, are acceptable and just as effective in early primary education. In Papua New Guinea, the national government moved the materials development process from expensive urban contexts to the communities themselves. Because of this, the communities were able to produce instructional materials in about half of the 800 local languages in the country. The main concern should therefore be on building the capability of communities to construct their own materials in their own languages. This way we can make the MLE quite affordable (UNESCO, 2007). According to Walter (2000) critics of the idea of bilingual or multilingual education have continuously promulgated that the paramount cost required in the process of developing a mother tongue or multilingual education program is, by itself, an adequate rationale for not taking into account this option. Others have suggested that higher arrangement costs are less than compensation by resultant achievements in academic competence (Patrinos, 1996 & Thomas, 1997).

Methodology

This study is a quantitative research. A quantitative research instrument (questionnaire attached as appendix "A") was developed to collect the data. The data was analyzed statistically through SPSS (quantitatively) and qualitatively. The study was conducted in the city of Karachi that is the most thickly populated multi-ethnic city of Pakistan. The population of Karachi reflects the representation of all major areas of Pakistan. It is known as economic hub, therefore, people from all walks of life rich or poor from all over Pakistan visit and reside the city for their economic gains (business) or employments. The study was conducted in the public sector colleges. These colleges own a good reputation and cost affordable fee, therefore, students from multiple backgrounds (rich or poor ) get their education in these colleges. The study was limited to intermediate level students. Due the big size of the target population, it was not feasible to collect the data form all students. A multi-layered sampling technique was used for selection of the respondents.

There were almost 132 male and female colleges in total. Almost 50% colleges were selected though random sampling for data collection including 36 male and 31 female (67) colleges. There were 264 respondents including 144 male and 120 female in the sample size who were selected through stratified random and purposive sampling. A pilot study was conducted before formal data collection. The questionnaire (attached as appendix 'A') developed for the data collection was close ended and it comprised eight questions in total which were finalized after detailed literature review on the topic and thorough consultations with experts in the field who were holding PhD degrees in social sciences particularly in the disciplines of education and linguistics. The Likert Scale which is widely used in survey research was also used in the present study for each question. It included following five attributes:

(a) strongly agree

(b) agree

(c) undecided

(d) disagree

(e) strongly disagree

Along with the pilot study as mentioned earlier, Cronbach's Alpha was also applied to research instrument though SPSS and data was collected through personal visits by the researcher himself to ensure the reliability and validity of the research respectively. The queries of the respondents were answered by the researcher then and there. It is also worthy to mentioned over here that the questionnaire return response was 100%.

Data Analysis

Data was analyzed through SPSS as per following details:

1. Major Hypotheses Testing

2. Item-by-Item Analysis

Testing of the Major Hypothesis

One null hypothesis was made for the study, which was tested by applying t-test as mentioned in the following:

Null Hypothesis

H0: There will be no significant difference of opinion among male and female students in respect of the impact of bilingualism/multilingualism in classroom discourse on their learning performance at intermediate level after change of MOI.

Analysis of the Problem

1. H^sub 0^ : μ 1 = μ 2

2. H^sub 1^ : μ 1 ≠ μ 2

3. α = 0.05

4. Test Statistics : t-test

t-test

Conclusion

Referring to the table of critical t values, it is found that at alpha = 0.05 with degree of freedom (df) = 262 the tabulated value of t= 1.9768, which is smaller than the computed value t= 3.044. Therefore, the H0 is rejected and it is established that there is significant difference of opinion among male and female students in respect of the impact of bilingualism/multilingualism in classroom discourse on their learning performance at intermediate level after change of MOI. The difference of opinion among male and female respondents is further obvious from their mean difference as tabulated above.

Item-by-Item Analysis of the Questionnaire

In this section of the study, the data would be analyzed item-by-item. Following points should be kept in view while reading the reports:

(i) Chi-Square One-Variable test is applied through SPSS for analysis.

(ii) Level of Significance Alpha (α) = 0.05

(iii) Degree of Freedom (df) = k - 1 = 5-1 = 4

(iv) Decision Rule: Reject H0 if computed value of Χ^sup 2^ ≥ 9.49

Following is the item-by-tem data analysis of Questionnaire:

Analysis of the Problem

The problem for each item was analyzed in the same pattern as mentioned below:

1. H^sub 0^ : f^sub 1^ = f^sub 2^ = f3

2. H^sub 1^ : f^sub 1^ ≠ f^sub 2^ ≠ f^sub 3^

3. α = 0.05

4. Test Statistics : Χ^sup 2^

Item No. 1

I understand better in Urdu than English during classroom discussion.

There is no significant difference in the opinion of students in respect of item no.1.

Computation Results of Χ2

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ2= 9.49, which is smaller than the computed value Chi Χ2 = 3.072E2 = 307.2. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 147 strongly agree and 91 agree with the statement, which confirms that, there is overwhelming majority (90%) of the students who understand better in Urdu than English during classroom discussion.

Item No. 2

I understand well when classroom discussion is in both Urdu and English.

There is no significant difference in the opinion of students in respect of item no.2.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ2 = 9.49, which is smaller than the computed value Chi Χ2= 4.213E2 = 421.3. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 171 strongly agree and 83 agree with the statement, which confirms that, there is overwhelming majority (96%) of the students who understand well when classroom discussion is in both Urdu and English.

Item No. 3

I use Urdu in the classroom to ask questions.

There is no significant difference in the opinion of students in respect of item no.3.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ2= 9.49, which is smaller than the computed value Chi Χ2 = 3.975E2 = 397.5. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 171 strongly agree and 74 agree with the statement, which confirms that, there is overwhelming majority (93%) of the students who use Urdu in the classroom to ask questions.

Item No. 4

Science/commerce teachers facilitate the students by using their mother tongue.

There is no significant difference in the opinion of students in respect of item no.4.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ2 = 9.49, which is smaller than the computed value Chi Χ2= 1.943E2 = 194.3. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 131 strongly agree and 77 agree with the statement, which confirms that, there is overwhelming majority (79%) of the students who confirm that science/commerce teachers facilitate the students by using their mother tongue.

.Item No. 5

I am unable to express myself well in English.

There is no significant difference in the opinion of students in respect of item no.5.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ2= 9.49, which is smaller than the computed value Chi Χ2 = 1.903E2 = 190.3. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 89 strongly agree and 99 agree with the statement, which confirms that, there is overwhelming majority (71%) of the students who confirm that they are unable to express themselves well in English.

Item No. 6

I can perform better in Urdu-medium exam than English-medium exam.

There is no significant difference in the opinion of students in respect of item no.6.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ^sup 2^= 9.49, which is smaller than the computed value Chi Χ2= 76.341. Therefore, the H0is rejected. The frequency table further shows that out of 264 respondents 100 strongly agree and 69 agree with the statement, which confirms that, there are 64% students who claim that they can perform better in Urdu-medium exam than English-medium.

Item No. 7

Change of medium of instruction decreases my learning performance at intermediate level.

There is no significant difference in the opinion of students in respect of item no.7.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 tabulated value of Chi Χ^sup 2^= 9.49, which is smaller than the computed value Chi Χ2= 86.947. Therefore, the Ho is rejected. The frequency table further shows that out of 264 respondents 90 strongly agree and 85 agree with the statement, which confirms that, there are 66% students who accepted that change of MOI decreases their learning performance at intermediate level.

Item No. 8

I adopt rote learning (cramming) to achieve high grades in the exam due to the change of medium of instruction at intermediate level.

There is no significant difference in the opinion of students in respect of item no.8.

Computation Results of Χ^sup 2^

Conclusion

Referring to the table of Chi-Square, it is found that at Alpha = 0.05 with degree of freedom (df) = 4 the tabulated value of Chi Χ2= 9.49, which is smaller than the computed value Chi Χ2= 98.917. Therefore, the Ho is rejected. The frequency table further shows that out of 264 respondents 94 strongly agree and 89 agree with the statement, which confirms that, there are 69% students who accepted that they adopt rote learning (cramming) to achieve high grades in the exam due to the change of medium of instruction at intermediate level.

Conclusion and Findings

The major rationale of this study was to explore the phenomenon of bilingualism/multilingualism in classroom discourse resulting from change of MOI at higher secondary level in the perspective of its impact on students' learning performance and make recommendations for effective learning in the light of opinions of the students of intermediate level. A null hypothesis was formulated for the study: There will be no significant difference of opinion among male and female students in respect of the impact of bilingualism/multilingualism in classroom discourse on their learning performance at intermediate level after change of MOI. The hypothesis is rejected and it is established that there is significant difference of opinion among male and female students in respect of the impact of bilingualism/multilingualism in classroom discourse on their learning performance at intermediate level after change of MOI in public sector colleges especially for science subjects. In accordance with the findings of the study, the students confirmed through their highly positively response that they understood better in Urdu than English during classroom discussion, understood well when classroom discussion was in both Urdu and English, used Urdu in the classroom to ask questions, and teachers facilitated the students while teaching science/commerce subjects by using mother tongue of the students. However, they accepted through moderately positive response that they were unable to express themselves well in English, could not perform better in Urdumedium exam than English-medium exam, adopted rote learning (cramming) to achieve high grades in the exam due to the change of medium of instruction at intermediate level and Change of MOI decreased their learning performance at intermediate level. It is therefore, concluded that bilingual or multilingual education is beneficial for the students if managed properly at appropriate levels in accordance with the needs of the students.

References

References

August, D. & Hakuta, K. (Eds.) (1997).Improving schooling for language-minority children, a research agenda, Washington, DC: National Academy Press, USA.

Dutcher, N. (2004).Expanding educational opportunity in linguistically diverse Societies, Washington, DC: Centre for Applied Linguistics, USA.

Dueñas, M. (2004). The whats, whys, hows and whos of content-based instruction in second/foreign language education, International Journal of English Studies Vol. 4 (1), 2004, pp. 73-96, University of Murcia.

Gulzar, A. M. (2010). Code-switching: awareness about its utility in bilingual classroom, Bulletin of Education and Research, December 2010, Vol. 32, No. 2 pp. 23-44.

Kangas, S. T. (2000).Linguistic Genocide in Education-or Worldwide Diversity and Human Rights?, MahwahNJ: Lawrence Erlbaum.

Li, B. & Shum, A.O.N. (2008). A discussion on using English as medium of instruction in Hong Kong and the sociolinguistic impacts, LCOM Papers 1 (2008), 37 - 51

Nisar, S. & Ahmad, S. I. (2011). Performance of secondary school certificate (ssc) graduates of English and Urdu medium schools: a comparative study. Language in India www.languageinindia.com, 11: 4 April 2011.

Nolasco, R.M.D. (2009). 21 reasons why Filipino children learn better while using their mother tongue: a primer on mother tongue-based multilingual education (MLE) and other issues on language and learning in the Philippines, University of the Philippines, 2009. http://catalogue.nla.gov.au/Record/4739741

Patrinos, H. & Eduardo, V. (1996). Costs and benefits of bilingual education in Guatemala: a partial analysis. (Unpublished research report). Washington D.C: The World Bank, USA.

Rasul, S. (2006). Language hybridization in Pakistan as socio-cultural phenomenon, an analysis of -mixed linguistic patterns, PhD Thesis, National University of Modern Languages; Islamabad Pakistan.

Svensson, P. F. (2005). From monolingual to bi- and multilingual instruction at the university of Helsinki: bi-and multilingual universities- challenges and future prospects, a paper presented at University of Helsinki, Language Centre, 1-3 September, 2005.

Schulter, B. (2006). Language learning and multilingual education from the point of view of the child - linguistic, cognitive and personal development in education forms with various linguistic settings, Conference Proceedings 7 on Language Policies and Education in Multilingual Societies: Tbilisi, Georgia, March 2, 2006.

Thomas, W.P. & Virginia C. (1997). School effectiveness for language minority students: Washington D.C.: National Clearinghouse for Bilingual Education, USA.

Tsui, A.B.M., Shum, M.S.K., Wong, C.K., Tse, K.S. & Ki,W.W., (1999). Which agenda? medium of instruction policy in post-1997 Hong Kong, Language Culture and Curriculum: Vol. 12, No.3

UNESCO, (2005). The importance of mother tongue-based schooling for educational quality: background paper prepared for the Education for All Global Monitoring Report 2005: The Quality Imperative, UNESCO 2005/ED/EFA/MRT/PI/9.

UNESCO, (2007). Advocacy kit for promoting multilingual education: including the excluded language in education policy and practice in Asia and the Pacific. UNESCO Bankok. ISBN 92-9223-110-3.

Walter, S.L. (2000). Explaining multilingual education: information on some tough questions: Work Papers of the Summer Institute of Linguistics, University of North Dakota Session 2000 Vol. 44. Online: URL: http://www.und.edu/dept/linguistics/wp/2000Walter.PDF

Williams, E. (1998). Investigating bilingual literacy: evidence from Malawi and Zambia, Serial No. 24. London: DFID.