The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Analyzing the causal impact of streaming service usage on IPTV viewing

스트리밍 서비스 사용이 IPTV 시청에 미치는 인과적 영향 분석

  • Dahai Jung (Department of Statistics, Sungkyunkwan University) ;
  • Yongho Yoon (Department of Statistics, Seoul National University) ;
  • Kwonsang Lee (Department of Statistics, Seoul National University)
  • Received : 2024.08.01
  • Accepted : 2024.08.15
  • Published : 2024.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

In modern society, the rapid growth of streaming services has significantly changed the way people consume television. This study aims to analyze the causal impact of streaming service usage on IPTV viewing. To achieve this, we compared users who use streaming services with those who do not while controlling for many possible confounders. We employed causal inference matching methods, focusing particularly on several matching techniques to compare groups with similar characteristics. Additionally, we used regression methods using matching-driven weights to assess the statistical significance of the causal effect. The results indicate that streaming service usage has a significant impact on how IPTV is consumed. These findings provide important insights for content providers, broadcasters, and advertisers to understand viewer behavior patterns and make strategic decisions accordingly. This study offers new insights into the relationship between streaming services and traditional TV viewing and can serve as a foundation for future related research.

현대 사회에서는 스트리밍 서비스의 급속한 성장이 전통적인 TV 시청 방식에 큰 변화를 가져왔다. 본 연구는 스트리밍 서비스 사용이 IPTV 시청에 미치는 인과적 영향을 분석하는 것을 목적으로 한다. 이를 위해, 스트리밍 서비스를 사용하는 사용자와 사용하지 않는 사용자를 비교하였으며, 다양한 잠재적 교란변수를 통제하였다. 특히 유사한 특성을 가진 그룹을 비교하기 위해 여러 매칭 기법을 활용한 인과추론 매칭 방법을 적용하였다. 또한, 인과적 효과의 통계적 유의성을 평가하기 위해 매칭으로부터 얻어진 가중치를 활용하는 회귀 분석 방법을 사용하였다. 연구 결과, 스트리밍 서비스 사용이 IPTV 시청 방식에 유의미한 영향을 미치는 것으로 나타났다. 이러한 결과는 콘텐츠 제공자, 방송사, 광고주들이 시청자의 행동 패턴을 이해하고, 이에 맞춘 전략적 결정을 내리는 데 중요한 정보를 제공한다. 본 연구는 스트리밍 서비스와 전통적인 TV 시청 간의 관계에 대한 새로운 통찰을 제공하며, 향후 관련 연구의 기초 자료로 활용될 수 있을 것이다.

Keywords

Acknowledgement

본 연구는 한국연구재단의 지원 (NRF-2021R1C1C1012750)과 서울대학교 신임교수 연구정착금으로 지원되는 연구비에 의하여 수행되었음.

References

  1. Abadie A (2005). Semiparametric difference-in-differences estimators, The Review of Economic Studies, 72, 1-19.
  2. Ding P (2024). A First Course in Causal Inference, CRC Press, Boca Raton, FL.
  3. Hansen BB (2004). Full matching in an observational study of coaching for the SAT, Journal of the American Statistical Association, 99, 609-618.
  4. Holland PW (1986). Statistics and causal inference, Journal of the American Statistical Association, 81, 945-960.
  5. Imai K and Kim IS (2019). When should we use unit fixed effects regression models for causal inference with longitudinal data?, American Journal of Political Science, 63, 467-490.
  6. Imai K, Kim IS, and Wang EH (2023). Matching methods for causal inference with time-series cross-sectional data, American Journal of Political Science, 67, 587-605.
  7. Robins JM, Hernan MA, and Brumback B (2000). Marginal structural models and causal inference in epidemiology, Epidemiology, 11, 550-560.
  8. Rosenbaum PR and Rubin DB (1983). The central role of the propensity score in observational studies for causal effects, Biometrika, 70, 41-55.
  9. Rosenbaum PR (2010). Design of Observational Studies, Springer, New York.
  10. Rosenbaum PR (2020). Modern algorithms for matching in observational studies, Annual Review of Statistics and Its Application, 7, 143-176.
  11. Rosenbaum PR and Zubizarreta JR (2023). Optimization techniques in multivariate matching. In Handbook of Matching and Weighting Adjustments for Causal Inference (1st ed), (pp. 63-86), Chapman and Hall/CRC, United Kingdom.
  12. Stuart EA (2010). Matching methods for causal inference: A review and a look forward, Statistical Science, 25, 1-21.
  13. Yu R and Rosenbaum PR (2022). Graded matching for large observational studies, Journal of Computational and Graphical Statistics, 31, 1406-1415.
  14. Yu R, Silber JH, and Rosenbaum PR (2020). Matching methods for observational studies derived from large administrative databases, Statistical Science, 35, 338-355.
  15. Zubizarreta JR, Paredes RD, and Rosenbaum PR (2014). Matching for balance, pairing for heterogeneity in an observational study of the effectiveness of for-profit and not-for-profit high schools in Chile, The Annals of Applied Statistics, 8, 204-231.