2016 Fiscal Year Research-status Report
Data-Driven Learning for English Language Beginners: An Empirically-Grounded and Evidence-Based Investigation
| Project/Area Number |
16K02878
|
|---|
| Research Institution | Niigata University |
|---|
Principal Investigator |
G.S Hadley 新潟大学, 人文社会・教育科学系, 教授 (70311008)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ハドリー 浩美 新潟大学, 教育・学生支援機構, 准教授 (60534732)
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Keywords | Data-Driven Learning / Corpus Linguistics / Extensive Reading / Mixed Methods Research |
|---|
| Outline of Annual Research Achievements |
Our statistical investigation, using dependent means t-Test analyses of the pretest and post-test data of learners within groups, found that the lexicogrammatical proficiencies of both the experimental and control Extensive Reading (ER) groups improve significantly, with high Cohen's d impact factors, over the course of one semester. This is regardless of whether the class is taught using Data-Driven Learning (DDL) or more traditional approaches to ER. However, independent means t-Tests have found that, when using a harder from of DDL with the experimental group, it was the control group that showed greater improvement. Qualitative data using Repertory Grid analysis indicated resistance on the part of the experimental groups. We conclude that hard DDL using comprehensible corpus data is likely to be inappropriate for undergraduate learners. Data findings suggest that even though DDL was seen as useful by the learners in the Experimental Groups, they nevertheless did not feel it was worth the investment of time needed to make it work. These findings necessitate further investigation into whether softer, flexible, and more aesthetically-pleasing forms of DDL are able to have a beneficial effect.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
Discoveries of what does not work are as important as finding out what does work. Both represent progress, as they will aid other teachers to either avoid or reconsider such a pedagogical approach. Our discovery that the use of comprehensible Hard DDL in an ER program for undergraduate learners does not lead to greater gains in reading speed, lexical knowledge or grammatical improvement than those achieved by comparable students in a non-DDL ER program is helpful. Our initial qualitative findings, we believe, will help to guide us towards discovering the conditions within which Data-Driven Learning might have the greatest impact among undergraduate learners in the Japanese ELT context. This will also aid us in producing better materials and processes for corpus usage within an ER context.
|
| Strategy for Future Research Activity |
We are currently designing DDL materials that promote affective engagement, based upon our initial findings. We anticipate that a repackaged, recontextualized form of soft DDL might be more effective. We also intend to adjust the approach to one that is data-directed rather than data-driven, that is, one which is directed by the teacher for the learners in an explicit, structured way, and which is as affectively engaging as possible. The impression of solitary hard study needs to be masked by purposeful activities that foster enjoyable social interaction in a non-threatening environment. We will then assess the effectiveness of this approach before proceeding to a grounded theory based investigation of student social processes arising from regular problems encountered within DDL-based ER.
|
