1.1 Background of the study

With the growing trend of globalization and the advances of information technology, English is now playing a significant role in today's world and has become the virtual lingua franca (Crystal, 1997). As China is being increasingly integrated into the international community, the government, public and especially, educators working in the area of English as a Foreign Language (EFL) teaching and learning have focused on developing students' ability to effectively communicate in English rather than the traditional focus on receptive skills. Writing ability allows graduates to communicate effectively with their international counterparts. Thus, English tests, both home and abroad, have also increased the value of their writing sections. Concomitantly, the recent years have also witnessed the flourishing of a wealth of research conducted at home and aboard in the area of English writing ability. This increased focus on writing has not yet produced satisfactory results. While millions of the College English Test (CET) certificate holders nowadays have acquired the reading ability required by the College English Teaching Syllabus, i.e., 70 wpm Chinese university graduates often have failed to develop the writing skills required by the teaching syllabus as they continue their education or move into the workplace (Yang, 2004). For example, when reading college students' essays, readers often come across such problems as vagueness, oversimplification, inaccuracy, repetitiousness, irrelevance, incoherence, and confused reasoning (e.g., Li, 2006; Pan, 1999). In addition, college students generally lack interest in English writing, as indicated in the literature(e.g.,Cai&Fang,2006;Wang et al.,2006;Zhang,2006)and also as surveyed later in this study. Worse still, when writing in English, college students are often void of communicative awareness; that is, they write only for the purpose of writing itself (Zou, 1999).

This situation, in consequence, has raised serious concerns within the field of EFL teaching where efforts have been made to explore possible ways to enhance Chinese college students' English communicative writing ability. For example, advanced theories of English writing have been continually introduced (e.g., Cai, 2001; Chen, 2001; Jia, 1998; Li, 2000; Li, 2006; Qin, 2000), and various approaches to teaching English writing have been proposed and tested(e.g.,Deng et al.,2003;Luo&Li,2003;Wang et al.,2000).Furthermore,the CET,a national standardized test sponsored by the Higher Education Department of the Ministry of Education, has undergone a series of innovations and reforms, including the changes in the component of writing, so as to keep up with the new developments of higher education in China and meet the needs of the country and society. More recently,College English Curriculum Requirements(2007)has been drawn up, which explicitly stipulates “the objective of College English is to develop students' ability to use English in an all-round way … so that in their future work and social interactions they will be able to exchange information effectively through both spoken and written channels…, ”clearly indicating the importance of English writing.

Although strenuous efforts have been made from diversified perspectives to enhance college students' English writing ability, it is noted that in the teaching, assessment and research of writing ability, a writing task must be used to elicit students' writing performance. Therefore, it follows that writing task construction is of primary importance. Writing tasks function as “a springboard, propelling students into the creation of essays” (Ruth & Murphy, 1988: 3), thereby playing a critical role in determining how students perform and bearing directly on the validity of the research and the writing test. Notwithstanding its vital importance, relatively little systematic research has focused on the study of writing task construction. Writing task was a “neglected variable” (Ruth & Murphy, 1988: xv) in the field of writing assessment and research, and it still is.

Despite the paucity of research in writing task construction, overarching guidelines have been proposed for reference when one is designing writing tasks (Kroll & Reid, 1994; Lindemann, 2001; Ruth &Murphy, 1988). Admittedly, there is no definite answer to the question of “What is a good writing task? ” However, basic assumptions of a good writing task are probably valid. For example, it is generally acknowledged that writing tasks should motivate writers to best demonstrate their writing ability and help them quickly become cognitively engaged with the task. At the same time, there is growing consensus that tasks should reflect the current understanding of the construct. For example, Weir and Shaw (2006:14) strongly recommended that test tasks should “represent the underlying real-life construct”. The construction of writing tasks should, accordingly, be guided by the research of the writing construct.

Language，as we know，is more often than not perceived as a tool of communication. Writing，similarly，is now recognized by nature as a communicative act: “We write mainly to communicate with other humans”(Hayes，1996: 5)；hence，every act of writing is done in a particular context. Just as Hamp-Lyons and Kroll (1997: 8) put it，writing is “an act that takes place within a context，that accomplishes a particular purpose，and that is appropriately shaped for its intended audience”. Given that all acts of communication take place within a certain context，writing tasks should，therefore，be contextualized for writers，as Cumming et al. (2000: 8) explicitly stated that “communicative language ability must include contextualized language use”. In view of the importance of context in human communication，Weir (2005) expanded the notion of “content validity” and introduced the term “context validity” into the field of language testing，thus strengthening the importance of context in language tests. In a similar vein，the Conference on College Composition and Communication (CCCC) Committee on Assessment (2006: 2) positioned“[B]est assessment practice engages students in contextualized，meaningful writing”.

Even though promising guidelines have been suggested to improve the practice of writing task design and construction, they are neither easily operationalized nor empirically testified. Admittedly, as early as the 1980s, a few studies were conducted, primarily in the setting of English as the first language (L1), to investigate whether specifying the writing purpose and/or the potential audience could help students display their best performance. These studies, however, produced somewhat conflicting results, and researchers differed on the effect of providing contextual features on writers' performances. Some researchers (e.g., Cherry, 1989; Odell, 1981;Quellmalz, 1986; Redd-Boyd & Slater, 1989) argued for specifying detailed contextual information in a testing situation in order to simulate realistic writing circumstances, whereas others (e.g., Brossell, 1983; Hillocks, 1986; Hoetker, 1982; McAndrew, 1982) held the view that rich contextual information in writing tasks might not help to elicit the best writing performance from the test-takers.

It should be noted that the studies mentioned above did not define nor operationalize the term “context”. Furthermore, all the above studies investigated the effects of contextual features on the act of writing merely on the basis of the essay scores given by raters. However, they didn't examine the actual cognitive writing processes employed by writers and their affective responses to those tasks with different degrees of contextual information. This is inadequate because not only the product but also the process that writers engage in should be considered for understanding the construct (Linn et al., 1991). Worse still, the rating scale used by the previous research is not provided in most cases. Therefore, it is not surprising that the results of these studies are contradictory given that a score in writing assessment is an interaction among such factors as the writer, the writing task, the rater(s), and the rating scale (Hamp-Lyons, 1990; McNamara, 1996; Weigle, 2002).

Moreover, the subjects in the previous studies are all native speakers of English. The results may be inapplicable to EFL learners. In the case of EFL learners, will contextual features affect their writing performance? How do they perform when writing on tasks with different degrees of contextual information? Do writing tasks with richer contextual information facilitate or inhibit their performance? These questions all deserve further investigation in order to enhance our understanding of the effects of contextual features on EFL learners' writing performance, and at the same time, also shed light on the design, construction, and validation of writing test tasks.

Compared with the studies carried out in the L1 setting, the research of the effects of the writing context in the second language (L2) or foreign language (FL) setting is scantier. In China, although concerns have been raised of the validity of writing test tasks and speculative suggestions have been proposed so as to make writing tests more communicative (e.g., Xu, 2000; Yan, 2004; Zou, 1999), no comprehensive empirical research has been conducted so far to investigate the feasibility of contextualizing writing tasks in English writing tests in China.

To sum up, no consensus has formed among researchers with respect to the definition of context, the operationalizability of context validity, and possible effects of contextual information on students' writing performance. Theoretically, providing contextual features on a writing task conforms to the understanding of the writing construct. From the empirical perspective, however, previous studies provided dramatically conflicting findings. As a result, some intuitive suggestions on contextualizing the writing task have been offered but limited evidence was provided in support of this type of practice. Given the relative paucity of research and the conflicting findings in this area, more empirical research needs to be conducted to understand what effect, if any, contextualizing the writing task can have on writers' performance. Therefore, the present study probed into writing task construction, and more specifically, explored what contextual features should be provided in writing tasks and what effects, if any, these contextual features will have on Chinese undergraduates' writing performance.