Peggy Semingson

Discourse-Analysis (Text-Based Analytics) of Discussion Boards in Elementary Literacy-Focused Online Teacher Education Courses

Slides

Peggy Semingson, PhD
Associate Professor of Literacy Studies
Department of Curriculum and Instruction, College of Education

Problem: Teachers as a whole are lacking background knowledge about the technical subject of phonics and word study to teach beginning readers and struggling readers at the elementary level. Watchdog organizations have harshly criticized the lack of phonics content in literacy courses for pre-service teachers in university-based teacher certification programs (Walsh, Glaser, & Wilcox, 2006). Recent research has identified the very weak knowledge base that practicing K-12 teachers have in order to teach beginning and /or struggling readers (Joshi, Binks, Hougen, Dahlgren, Ocker-Dean, & Smith, 2009). The role of teachers’ knowledge of phonics is central to the conversation of the specific knowledge that is required to effectively teach beginning reading and readers (Cunningham & Stanovich, 2004). Implications on instruction and student achievement are significant; the effect on this gap in teacher knowledge is especially severe in its effect for at-risk readers (e.g., those with learning disabilities, dyslexia, or other reading challenge) and beginning readers. Teacher’s knowledge base in literacy impacts the instruction of their students (McCutchen, Abbott, & Green, 2002).

Project Questions: a) Based on text-based analysis a large corpus of data, how do students feel (sentiment analysis) about the subject of phonics and beginning reading? b) How do students discuss the subjects of phonics, word study, and beginning reading (thematic analysis) c) What are key words related to the technical aspects (domain knowledge) of phonics, word study, and beginning reading that are the focus of the discussion boards. How does preservice teacher dialogue about these topics on discussion boards differ from the grad student dialogue?

Applications to Practice: Based on the outcomes of this analysis, I will redevelop my materials and structure of the discussion boards to further student learning in future iterations of the class.

Methodology: Question 1: Sentiment analysis Data: Discussion boards of graduate and undergraduate elementary literacy courses for weeks that the topic of phonics, word study, and beginning reading was a major subject (weeks 1 and 2) for iterations of the course since 2012. Analytics tool: LIWC (Linguistic Inquiry and Word Count) https://liwc.wpengine.com/ * * (Pennebaker, J., Chung, C., Ireland, M., Gonzales, A., & Booth, R., 2007; Tausczik, Y., & Pennebaker, J., 2009) Question 2: Thematic Analysis Data: Discussion board data, as above Analytics tool: NVIVO 10 (qualitative software Question 3: Keyword analysis Analytics tool: NVIVO 10 (qualitative software)

References

Cunningham, A. E., Perry, K. E., Stanovich, K. E., & Stanovich, P. (2004). Disciplinary knowledge of K-3 teachers and their knowledge calibration in the domain of early literacy. Annals of Dyslexia, 51, 139-168.

Joshi, R.M., Binks, E., Hougen, M., Dahlgren, M., Dean,E., & Smith, D. (2009). Why elementary teachers might be inadequately prepared to teach reading, Journal of Learning Disabilities, 42, 392-402.

Pennebaker, J., Chung, C., Ireland, M., Gonzales, A., & Booth, R. (2007). The development and psychometric properties of LIWC 2007. Retrieved from: http://www.liwc.net/LIWC2007LanguageManual.pdf

Tausczik, Y., & Pennebaker, J. (2009). The psychological meaning of words: LIWC and computerized text analysis methods. Journal of Language and Social Psychology, 29(1), 24-54. doi:10.1177/0261927×09351676

 

  • J. T. Dellinger

    Neat project, Peggy! It’s cool that you get to be a part of the Civitas pilot - I look forward to hearing how that goes! How do you think that the predictive tools will help your class in particular? Regarding this project, how will you differentiate between beginners and more advanced phonics users in your class, and do you plan to examine both separate and then compare?

    • Peggy Semingson

      I think the predictive tools will help me and the academic coaches to intervene early with students identified as having challenges. We can hopefully also improve retention….
      For this project, I am guessing the preservice teachers are beginners although some of our grad students are also beginners. I think I will more likely differentiate more broadly between preservice teachers in the aggregate with the grad students.

  • Melanie Mason

    So interesting. I want to use text-based analysis going forward with my research. What is your process for identifying key words in your data?

  • Kim Breuer

    I am interested in seeing how you will combine this data with Civitas data. Once we start looking at our data this fall, we will have to compare notes. It would seem to me that we should be able to identify cohorts within a live class and use both sets of data to provide more customizable learning pathways for students in online courses in nearly real time.

    • Peggy Semingson

      Let’s keep each posted as we work through the Civitas pilot/data!

  • Maria Trache

    I learned something completely new.. sentiment analysis. I looked in the past at attitudes (e.g., attitudes toward science) and classified them as positive or negative based on agreement/disagreement with some statements that were conceptually indicative of some attitude. Not sure how it works in ‘sentiment analysis’ .. is it more like a text analysis in which words perceived as positive/negative/neutral are somehow counted? Interesting

    • Peggy Semingson

      Interesting parallel/connection, Maria! I believe sentiment analysis is used heavily in the business sector, at the moment!

  • Regina Urban

    I’m so glad you chose this topic, Peggy. Your overview of how you used it to look at thematic and sentiment analysis of your students who were learning to use / apply phonics in the classroom with grade school students was very helpful. Were there any aspects of LIWC that you found challenging as you were learning the program for the sentiment analysis? Are researchers in your field of interest publishing many studies that have used LIWC for the analysis?

    • Peggy Semingson

      I am still in progress with the project, but I will keep people posted! There are a few studies with LIWC, but hardly any in my field of literacy! I think it came out of a psychology framework…..

  • Cindy

    I so want to learn more about LIWC - I have many ideas about using it!

    • Peggy Semingson

      I will keep you posted!