Strategies to Process Large Mixed Data Sets in Curriculum Intervention Research

• In Event: Invisible College: Session 4

Abstract

The primary objective of this session is to discuss common and unique challenges in handling mixed-data set generated in large-scale PE curriculum intervention research through interactive slide presentations and group discussions. The secondary objective is to stimulate and brainstorm strategies to be utilized at different stages of curriculum intervention research, including research conceptualization, research design, data collection and management, data cleaning and data analysis. The objectives are built on previous three Invisible College sessions in 2013 (Chen & Ennis: Designing large-scale curriculum intervention studies), 2014 (Ennis & Chen: Scaling up research to increase opportunities for grant funding), and 2018 (Zhang & Chen: Challenges and solutions in large-scale Intervention research) to address unique and common challenges in the operation of large-scale PE curriculum intervention research.
 Session Format: We will incorporate lead-up presentations and group discussions. First Slide Presentations (20 minutes) will focus on the data types in large-scale curriculum intervention research. Chen will briefly discuss data considerations at the conceptualization and design stages. Zhang will present strategies to sequence data collection and organize/manage data. Wang and Deng will share their unique experiences of structuring mixed-data sets to answer distinctive research questions. Participants will be encouraged to share their experiences of working on qualitative, quantitative, and/or mixed data with individual and institutional as the unit of analysis. and data structure (including but not limited to structured numerical datasets, individual or group interviews, field notes, and documents collected for ethnographic analyses).
 First Group Discussion (10 minutes) will be devoted to identifying data-processing related challenges in their research settings. A think-pair-share discussion will be used for participants to (a) think about the challenges in their own research settings and write them down (2 minutes), (b) join other 2-3 participants to share and compare the challenges (5 minutes); and (c) each group will identify the challenges common to all participants as well as those unique to each (3 minutes).Each group’s spokesperson will share the group’s discussion results to all participants (5 minutes).
 The Second Slide Presentation (10 minutes) will be given AFTER the first group discussion to show major data processing challenges that we encountered and solutions we used to address them. The purpose is to start the group discussion leading to solutions to the challenges identified in the first group discussion. During both presentations, the participants will be asked to share their experiences and ask questions.
 In the Second Group Discussion (15 minutes) participants will (a) brainstorm possible solutions to their groups’ common challenges and solutions to unique challenges as well; and (b) develop strategies to handle data set effectively. The group discussion will again be shared with all participants. At the end of the discussion, the strategies will be summarized as session closure.

Authors

• Tan Zhang, Winston Salem State University

• Anqi Deng, University of South Carolina - Columbia

• Yubing Wang, University of Wisconsin-Whitewater

• Ang Chen, University of North Carolina - Greensboro