상세 보기
초록
Purpose: This study aimed to trace the long-term structure and change of physical education (PE) teacher-related discourse using SNS big data, to examine its dynamics through bigram analysis and Latent Dirichlet Allocation (LDA) topic modeling, and to discuss theoretical and practical implications. Methods: SNS text data related to PE teachers (N = 19,482 documents; 21,516 KB) were collected from Naver, Daum, and Google web platforms covering the years 2015 to 2025. Bigram analysis and LDA topic modeling were conducted in parallel for the entire period and for four sub-periods (2015–2017, 2018–2020, 2021–2023, 2024–2025), and the convergence between the two methods was examined. Results: First, discourse on PE teacher training, certification, and recruitment preparation ranked among the top two topics in every period, indicating that concerns of the pre-service socialization stage are highly visible in the online public sphere. Second, recruitment discourse emerged as an independent topic in Period 2, and discourse on school-based PE activities and after-school programs ranked first for the first time in Period 4, marking a shift in emphasis toward practice-oriented discourse. Third, discourse on teacher misconduct appeared as an independent topic in Periods 1 and 4 and remained latent as sub-vocabulary in the other periods, showing a recurring pattern of negative-incident discourse with changing content. Fourth, discourse on sports-related occupations and career information became established as an independent topic from Period 2 onward. Conclusion: By applying bigram analysis and LDA topic modeling to eleven years of longitudinal online big data and examining their convergence, this study empirically explored the composition and temporal change of PE teacher-related discourse and provides a basis for further discussion on PE teacher education and in-service professional development.
키워드
- 제목
- SNS 빅데이터로 본 체육교사 담론의 구조와 변화*: 2015~2025년 LDA 토픽 모델링과 바이그램 분석을 중심으로
- 제목 (타언어)
- The Structure and Change of Physical Education Teacher Discourse in SNS Big Data: LDA Topic Modeling and Bigram Analysis, 2015-2025
- 저자
- 오현정; 이병준
- 발행일
- 2026-10
- 유형
- Y
- 저널명
- 한국체육교육학회지
- 권
- 31
- 호
- 4
- 페이지
- 23 ~ 42
- 언어
- KOR
- 출판사
- 한국체육교육학회
- 발행국가
- 대한민국
- 분량
- 20 페이지
- ISSN
- P 1229-9685