We will present an innovative use of text-mining tools: helping students to identify and remedy the extreme repetition of basic verbs in intermediate language classes. We noticed that intermediate French students tend to repeat certain verbs and expressions in their writing, often relying heavily on verbs such as “to do.” Could we help students get out of this vocabulary rut by calling their attention to this repetition and giving them level-appropriate resources for synonyms? Over the past three semesters, we have been exploring the effectiveness of text mining tools as a means to build vocabulary variety. Our use of text-mining tools in foreign language writing is an innovation: what little research exists on the use of such tools in foreign language classrooms is largely devoted to their use in preparing students for reading projects. Our project focuses instead on the acts of writing and revising.

Our question:  how do we get students to notice how restricted their written vocabulary is?  How do we help them navigate the thorny process of identifying an appropriate replacement for an oft-repeated word?  We turned to digital collaboration and publication tools, and to creating a digital vocabulary database. We developed a thematically-linked pair of wikis that provide models and scaffolding to guide students through a step-by-step process of individualized research, creative production, drafting, revising, peer-editing, peer commenting, and self-reflection. While these wikis and writing cues were created for intermediate French students, the design may be adapted to fit any writing assignment .

In our project, intermediate French students submit the rough draft of their compositions to voyant-tools.org, a web-based text-mining tool that analyses word frequency. They then are guided to examine each of the repetitions of their most-used word.  Students are asked to locate the sentences in which these most-used words appear and to consult a custom online thesaurus we created with level- and theme-appropriate synonyms. Students consider this information and then use it to attempt to revise the targeted sentences before submitting their composition for evaluation.

Using the same text analysis tools as the students, we look at repetition in their composition and compare this to the students’ analysis to evaluate how successful they are at the task. We also compare the first and second drafts of each participant's work. Elements reviewed include but are not limited to changes in the number of unique words between drafts of the same composition, changes in the number of unique words between first drafts of different compositions, and changes in frequency of the most commonly used words between first drafts of different compositions. This analysis is repeated each semester in an attempt to identify consistent trends. When participants fit the eligibility criteria and agree to participate across multiple semesters, longitudinal analysis is be added.

Our initial findings show that this structured approach employing easy-to-use technologies can encourage students to successfully enrich the vocabulary of this writing project. We also find that students become more aware of the problem of repetition, particularly of certain high-frequency verbs, and we begin to notice an improvement both in their ability to use synonyms correctly when asked to do so and in their ability to avoid overusing common verbs.
 
References:
Abrams, Z & Byrd D. (2016). The effects of meaning-focused pre-tasks on beginning-level L2 writing in German: An exploratory study. Language Teaching Research, 1-20.
Baralt, M., Pennestri, S., & Selvandin, M. (2011). Using Wordles to Teach Foreign Language Writing. Language Learning & Technology. 15(2), 12-22.
Doyle, P. (2010). Viewing Language Patterns: Data Visualization for Data-driven Language Learning. In C. M.L. Ho, K. T. Anderson, & A. P. Leong (Eds), Transforming Literacies and Language: Mulitmodality and Literacy in the New Media Age (149-166). London: Continuum.
Gottron, T. (2009). Document Word Clouds: Visualising Web Documents as Tag Clouds to Aid Users in Relevance Decisions. In J. Borbinha, S. Kapidakis, C. Paptheodorou, G. Tsakonas (Eds.), Research and Advanced Technology for Digital Libraries: 13th European Conference. ECDL 2009, Corfu, Greece, September 27 - October 2, 2009, Proceedings (94-105). New York: Springer.
Johnson, M., Mercado, L., & Acevedo, A. (2012). The effect of planning sub-processes on L2 writing fluency, grammatical complexity, and lexical complexity. Journal of Second Language Writing, 21(3), 264-282.
Sorapure, M. (2010). Information Visualization, Web 2.0, and the Teaching of Writing. Computers and Composition, 27(1), 59–70.
Sun, K.-T., Huang, Y.-M., & Liu, M.-C. (2011). A WordNet-Based Near-Synonyms and Similar-Looking Word Learning System. Educational Technology & Society, 14 (1), 121–134.
Webb, S. (2009). The Effects of Pre-learning Vocabulary on Reading Comprehension and Writing. The Canadian Modern Language Review, 65(3), 441-470.