Abstract The present study investigated the sentence-level errors of freshmen students at three proficiency levels and the aspects of writing that raters focused on while rating the essays. It views errors as valuable information for the following: For teachers, as it clues them on students’ progress; for researchers as it gives them valuable data as to how language is acquired or learned; for learners, as it enables them to reflect on their learning.
The data for the present study is based on the data collected for a previous study. One hundred fifty essays written by freshmen college students on their first week of classes in five private schools in Metro Manila (30 for each participating school) were collected, word-processed, and subjected to rating and coding or errors. Most of the findings of the present study corroborate the findings of previous studies on error analysis and essay evaluation—that sentence-level errors have a significant role in essay scores.
The raters still have the grammar accuracy model when checking essays, although it is just considered secondary to other aspects of writing such as the ability to address the prompt and organize the ideas logically. Introduction Areas of Writing Research Writing teachers and researchers have always set their teaching and research lenses on the variables that describe successful second language writing vis-a-vis unsuccessful writing.
Because of this preoccupation, a plethora of research has been undertaken as regards the role of L1 in L2 Writing (Cumming, 1990; Krapels, 1991), L2 writers’ characteristics and proficiency (Hirose & Sasaki, 1994; Victori, 1999; Deane et al, 2008), L2 writing process/strategies (Arndt, 1987; Becker, 2003), L2 writing feedback/evaluation (Ferris, 1995, 1997, 2002, 2003; Goldstein, 2001, 2005) L2 writing instruction (Zhang & Zhou, 2002; Liu, 2003; Chen, 2005; Coombe & Barlow, 2004) and L2 writer’s texts (Zhang, 1997; Hinkel, 1997; Hirose, 2003).
In the analysis of L2 writers’ texts, researchers focus on the rhetorical and linguistic features that impact essay scores in different linguistic backgrounds (Becker, 2010; Gustilo, 2011). These studies are significant in that they help us in understanding the complex factors that underlie proficient and less proficient writing. Emphasis on Essay Evaluation In assessing essays, what do raters consider as more proficient and less proficient writing? Studies have established that factors such as content, rhetorical strategy or linguistic (grammar, spelling, vocabulary) and non-linguistic features (organization, style, content, etc. can separate less proficient and more proficient writing. These factors, however, are weighed differently by raters depending on their background and experience (Kobayashi, n. d. ; Weltig, n. d. , Cumming et al. , 2002). According to Weltig, ESL (English as second language), EFL (English as foreign language), and ENL (English as native language) raters weigh the aspects of writing differently. On the one hand, the ENL raters put more thrust on the non-language aspects (grammar and mechanics) and consider language-related issues as secondary in judging essays.
On the other, ESL and EFL raters focus more on language erro Error Analysis As a corollary to this emphasis on essay evaluation, there is no wonder why almost inextricable in the analysis of L2 texts is the focus on errors that learners make in relation to their writing performance (Sarfraz, 2011; Sattayatham & Honsa, 2007; Kitao & Kitao, 2000). Previous studies have indicated that errors do significantly affect raters’ evaluation of the overall quality of essays. For instance, Sweedler-Brown (1993) attributed the low scores of L2 essays for the original than for the corrected essays to sentence-level errors.
Kobayashi and Rinnert (1993) found that the same essays gained higher overall holistic scores when the errors were corrected. Kobayashi (n. d. ) also found that language use errors and coherence breaks influenced the English teachers’ (both native speaker and Japanese teachers) judgment in terms of content and clarity but not the Japanese university students’ judgments. The error-free essays were praised as very good, but the error-laden essays were penalized with low scores. Terms and Definitions of Errors Making errors is one of the most unavoidable things in the orld. In language acquisition, learning, and teaching, error has been referred to and has been defined in many ways. Catalan (n. d. ) reviewed the terms and definitions of error in Error Studies and consolidated these definitions using the communicative event framework (addresser; addressee; code, norm, and message; and setting). From the perspective of norm or well-formedness of a sentence, error is regarded as “an infringement or deviation of the code of the formal system of communication through which the message is conveyed” (Catalan, n. . p. 66). Dulay, Burt, and Krashen (1982 p. 139 as cited in Catalan ) corroborates this definition by saying that error is the “flawed side of learner speech or writing that deviates from selected norm of mature language performance” (p. 7). Errors are alterations of the rules of the accepted norm and are termed as surface errors which may be further classified as omission errors, addition errors, misformation errors, wrong order, spelling error, systems error, and the like.
Also the terms overt and covert errors (Corder, 1973, Faerch, 1984, & Medges, 1989 as cited in Catalan, n. d. ) may be added in the list. The former refers to errors that are not observable within the surface but implied in the message, and the latter refers to those that are clearly identifiable in the surface (Catalan n. d. , p. 8). Benefits of Error Analysis Error Analysis (EA) is concerned with the analyses of the errors made by L2 learners by comparing the learners’ acquired norms with the target language norms and explaining the identified errors ( James, 1998).
Stephen Pit Corder’s (1967) seminal work “The Significance of Learner’s Errors” has given EA a significant turn in that it views errors as valuable information for three beneficiaries: for teachers, it clues them on the progress of the students; for researchers, it provides evidence as to how language is acquired or learned; for learners themselves, it gives them resources in order to learn (Corder, 1967 as cited in Maicusi, Maicusi, & Lopez, 2000, p. 170).
With this approach errors are regarded as resources for learning and teaching rather than as “flaws” which connote failure in the acquisition process and ,therefore, needs to be eradicated. Despite the criticisms against EA such as complete reliance on errors per se and not seeing the whole picture of the learners’ linguistic behaviour by looking also at the nonerrors (Gass & Selinker, 1994; Maicusi, Maicusi, & Lopez, 2000), EA has contributed comprehensively to Second Language Acquisition Theory and second language writing instruction. Aim of the Study The present study aims at investigating the sentence-level rrors of freshmen students at three proficiency levels and the aspects of writing that raters focus on while rating the essays. Specifically, the present study aims at providing answers to the following research questions: 1. Do writers with higher levels of writing proficiency commit the same errors that low proficiency writers do? 2. Are there significant differences in the frequency of errors committed by low, mid, and high proficient writers? 3. Which of these errors significantly decrease essay scores? 4. What aspects of writing that affect essay scores did the raters focus on while rating the essays?
Method The Essays The data for the present study is based on the data collected for a previous study. One hundred fifty essays written by freshmen college students on their first week of classes in five private schools in Metro Manila (30 for each participating school) were collected, wordprocessed, and subjected to rating and coding or errors. Rating Three independent raters who are trained ESL teachers rated the essays using a holistic scale patterned after the TOEFL writing section and SAT scoring guides (Gustilo, 2011). Kendall’s Tau coefficient of concordance (. 71, p