O n f e r e n c e

Carola Strobl 
Ghent University, Ghent, Belgium 
[e-mail] 
Academic Writing in a Foreign Language: Learning to Elaborate 
Instead of Copy-Paste through Online Collaboration 
Bio data 
Carola Strobl
is a Research Assistant at the Faculty of Arts and Philosophy of Ghent 
University. She has a teaching career as a German lecturer at universities in Germany, 
Italy, Portugal, and Belgium, and has been involved in several research projects on 
technology-enhanced language learning. Currently, she is writing her dissertation entitled 
“Exploring the possibilities of online scaffolding for L2 academic wri
ting in individual and 
collaborative settings”.
Abstract 
Synthesis writing from multiple sources in a foreign language challenges the writing 
strategies of novice writers. Students need to cognitively engage with the content of the 
source texts instead of copy-pasting fragments to produce an effective synthesis (Solé et 
al., 2013). They also have to broaden the focus of their linguistic attention from word 
and sentence level towards text level in order to produce a coherent synthesis (Paulus, 
1999). Online peer collaboration can help students to master these challenges by pooling 
their cognitive and linguistic resources. Furthermore, the collaborative task setting 
fosters content discussion and reader-orientation (Blin & Appel, 2011).
For this collaboration to be effective, students need to be scaffolded in the collaborative 
process (Slavin, 1992). This study sets out to investigate what type of instructional 
support works best for this specific task and target public, advanced foreign language 
students in higher education. The two instructional methods that were used to support 
the online collaboration are 
scripting
and 
observational learning
. Both instruments have 
been shown to foster effective results in Computer Supported Collaborative Learning 
(Rummel et al., 2009).
To measure their effect on online synthesis writing, a three-weeks intervention study was 
set up, implementing a cross-sectional design: In the "script-first" condition (n=34), the 
collaborative triads followed a collaboration script in week two and watched a model 
video of an online collaboration in week three, and vice-versa in the "model-first" 
condition (n=42). The collaboration process was recorded using screencast software, and 
analysed with regard to efficiency (work flow) and effectiveness (focus and/or success of 
interactions). Attitude questionnaires and focus group interviews served to triangulate 
the observational data. 
The analysis results shed new light on the effect of instructional support. Attention to 
content elaboration and text level concerns increased throughout the intervention, but to 
a different degree for the two groups: The triads in the “model
-
first” condition benefited 
more from the instructional support. This finding is corroborated by higher ratings 
regarding the usefulness of the instructional support given by the participants in this 
condition.

---

-324- 
2014 CALL Conference 
LINGUAPOLIS
www.antwerpcall.be