This paper describes the challenges faced, and rules devised, while dealing with bilingual interview data as part of a life history study of a female science teacher's conceptions of the nature of science while teaching in a school in Karachi. The interview data generated was both in Urdu and English, which underwent a number of processes (transcription, translation, and transliteration) to evolve into "interim texts," to finally become a part of the data analysis process. I have called these translated materials "transmuted texts," as they reflect the original, but have been recreated. This paper is significant because as globalization connects diverse societies, more research studies have to deal with research data in more than one language. Key Words: Qualitative Research Methods, Interview Data Analysis, Bilingual Data, Transcription, Transliteration, and Translation
This article is based on the process of data generation during a single-case, life history study of a young female science teacher in Pakistan. The purpose of the study was to understand the science teacher's ideas about teaching science, and her conceptions of the nature of science through in-depth interviews and observations of her teaching in the classroom. The data collection spanned a period of seventeen months when more than thirteen life history interviews were conducted with this science teacher. Thirty science lessons in grades seven and eight were observed over a period of six months in a secondary school in Karachi, Pakistan (Halai, 2002).
In this paper I describe the challenges faced while handling bilingual interview data in Urdu (the national language of Pakistan) and English. This includes the process of converting the data into English text, and some rules that were devised as I continued with the conversion process. My research study is rooted in science education, but the theoretical lens that I have used to view the use of bilingual interview data, among other things, involves cultural decoding (Torop, 2002) I have borrowed key ideas from translation studies for this process (Crystal, 1991; Lambert, 1997; Nida, 1982). It is important to note that the implications of this paper and the issues highlighted, as related to bilingual interview data, are just as valid for other kinds of qualitative data as it is for life history interviews or data generated through biographical genres of research. Due to globalization and pluralization of societies, research in education has to increasingly face issues of bilingualism in education and the use of bilingual data in research studies. Yet, very little has been written about the process of generation, conversion, and utilisation of this data into text. This article will illustrate some of the processes that a researcher went through to arrive at the research texts. Although, in this paper, I discuss methodological issues that arise from the conversion of data in Urdu and English, into research texts, the issues that arise are common to any bilingual interview data. Hence, the lessons learnt can be applied to bilingual data in any two languages.
Urdu is the national language of Pakistan. It occupies a unique space in the linguistic landscape of Pakistan. Less than 9% of the population speaks Urdu as their mother tongue, but it is used as the lingua franca to communicate with different ethnic communities such as the Punjabis, Sindhis, Balochis, and Pushtuns that live in the four provinces of Pakistan. Urdu as a language has its origin in the Indian subcontinent during the time of the Mughals (1526-1858) when the language of the court was Persian, but the language of the masses became Urdu. Urdu is written in Arabic script and has freely borrowed from Arabic, Persian, and Turkish languages. It has a great ability to absorb other languages and has also contributed a number of words to the English language.
In qualitative studies, interviews and conversation are data; and this data in the form of audio-taped recordings are further transformed to transcribed text. …