SPANISH WRITING: WHAT DO RATERS LOOK FOR? Ana Maria Ducasse La Trobe University (AUSTRALIA) a.ducasse@latrobe.edu.au Abstract As students as 'writers' move across the globe to courses in higher education, the area faces challenges regarding academic writing across cultures. In a 1988 international comparative study edited by Gorman, Purves & Degenhart it was found that student writing across educational systems had in common what constituted a written product by its surface features. However, beyond that, each nation varied in the perception of what was valued. The CEFR very much challenges this notion bringing common standards across writing in multiple languages and cultures. On this point the current study investigates Spanish rater perceptions of what is valued in educated Spanish writing on two tasks: a reflection and an argument. The words raters focus on while using the scales (Eckes 2004), as well as the choice of band-scale wording for scale development will be presented. The grid will be compared to an English one for the same pair of tasks. Furthermore, the rating grids for academic Spanish will be compared to the CEFR C1 / C2 Spanish versions. The data will illuminate what raters consider to be valuable for judging written Spanish and how this concurs with the CEFR higher levels. Keywords: L1 writing assessment, rating criteria, CEFR, evidence based scale development. 1 INTRODUCTION The purpose of this paper is to investigate how Spanish L1 or native speaker writing is rated by L1 raters. It looks at how raters use different parts of a six point global scale with three characteristics per band global scale, to justify a decision for a mark. In addition, it is an exploratory investigation of the definition of ‘’writing’’ as a construct for native speakers in two languages, by making observations from the content of rating criteria for the same university graduate skills in a writing test in: English and Spanish. The aim is to see if looking at different scales provides an insight into the expectations of writing. The motivation behind the study is that as student mobility increases, one of the many challenges facing those who judge applications for different courses in higher education is academic writing across cultures. If a student applies for course that is taught in an English medium institution what results in the academic classroom from the interaction of a grade in English on the application and the academic skills expected of a student assessed for proficient writing in Spanish? The methodology uses scale documents for a qualitative appraisal of the material and some statistics are presented of rater preference. Written reports after rating have been used by Milanovic et al. (1996) among other methods to gain an insight into the complexity of raters’ cognitive processes and they are also used here. Since it has been found that the words raters focus on while using the scales (Eckes 2004) can belong to different skills then a way to access expectations of writing is to look at wording in rating scales and what raters focus on to assign a grade. 2 REVIEW OF RELEVANT LITERATURE 2.1 Rater cognition and scales This study contributes to the area of rater cognition, as in what raters think about when they are marking. The study responds to a call from Eckes, (2008) asking that L2 raters provide the actual importance of rating criteria when they have assigned a grade; because in his study raters provide their perceived importance of scoring criteria in general, but not in an operational setting. This type of study is of interest because as Huot (1990: 258) says in reference to the rating of L1 writing “if assessment literature is to progress, more inquiries are needed about how raters arrive at judgements about writing quality”. Other L2 studies such as for example McNamara (1996), Wolfe (1997) or Lumley (2005) also look at how raters arrive at a score but from a second language testing Proceedings of ICERI2011 Conference. 14th-16th November 2011, Madrid, Spain. ISBN: 978-84-615-3324-4 004528