Research on Reading-Writing Assessment of Multilingual Learners

• In Event: The Interrelation of Literacy Skills Across Languages: Reading-Writing Connections in Monolinguals and Bilinguals

Abstract

In developing assessments that reflect language use in context, integration of language modalities has become a useful approach. Treating language as holistic or as an amalgamation of reading, writing, listening, or speaking, rather than as isolated abilities may allow for different or more substantive interpretations of performances that are transferrable to real world domains. While integrating skills presents challenges for test developers and users, such tasks provide measures that may overcome some gaps left by independent skills assessment as well as introduce new directions for assessing language (Knoch & Sitajalabhorn, 2013; Plakans, 2008; Read, 1990; Weigle, 2004; Yang, 2014).

This presentation will provide an overview of research in the past decade on assessments that integrate reading and writing with adult multilingual learners of English. The attention to integrated assessments has proliferated during this time, providing substantial evidence to answer questions useful for research and application. The review of research will cover several critical areas: defining the construct, task types, and source texts.

Given the importance of defining a construct for test development, it is not surprising that scholars have undertaken studies targeting the underlying abilities involved in integrated assessment. One question being pursued is whether the resulting score from an integrated task is a measure of all the skills that are integrated. For example, Sawaki et al. (2009) analyzed the Internet-based Test of English as a Foreign Language (TOEFL) products to inform score interpretation. They found that scores on the integrated speaking and writing items were only slightly related to the less dominant skills in the tasks (listening, reading), and thus their scores were better interpreted as measures of writing or speaking. Another approach to defining the construct in integrated tasks investigates the abilities that underlie integration and the synthesis of the skills, rather than individual skills added together. Ascención (2005) and Plakans (2009) investigated the concept of “discourse synthesis” in second language integrated writing tasks using think-aloud verbal protocols. In both studies, the results of recording writers’ online processes revealed the sub-processes of discourse synthesis did occur in composing integrated writing tasks.

While a good deal of variation exists in integrated tasks, a few studies have looked at the differences across them. Researchers have explored the genre of writing, topic/audience, and, most recently, the use of source materials in integrated writing. The topic selected for a task has emerged as an issue for integrated tasks (Huang et al., 2018). For example, Cumming et al. (2005) discovered differences in source text use between two task topics, one regarding political science and the other on cinema. Related to the issue of topic, the texts used in integrated tasks have generated increased interest recently. Researcher have looked at text structure finding complex interactions with genre and difficulty of source texts (Cho, Rijmen & Novák, 2013; Li, 2014; Yu, 2009).

Based on the review of research, the presentation will conclude with a suggested directions for future research and development of integrated reading-writing assessments.

Author

• Lia Plakans, University of Iowa