From problem to method : designing multiview sequential canonical covariance analysis for online word-of-mouth dynamics /
This case study introduces a novel method--Multiview Sequential Canonical Covariance Analysis (MultiSeqCCoA)--designed to address key methodological challenges in analyzing dynamic, high-dimensional, and multientity online word-of-mouth (WOM) data. Grounded in our 2024 publication in Decision Suppor...
| Main Authors: | , , , |
|---|---|
| Format: | eBook |
| Language: | English |
| Published: |
London :
SAGE Publications Ltd,
2026.
|
| Subjects: | |
| Online Access: | Connect to the full text of this electronic book |
| Summary: | This case study introduces a novel method--Multiview Sequential Canonical Covariance Analysis (MultiSeqCCoA)--designed to address key methodological challenges in analyzing dynamic, high-dimensional, and multientity online word-of-mouth (WOM) data. Grounded in our 2024 publication in Decision Support Systems, we developed MultiSeqCCoA to overcome three persistent limitations in existing approaches: handling unstructured and high-volume data, capturing time-sensitive changes, and identifying shared sentiment signals across competing entities or social groups.We describe the rationale behind the method's design, including our review of existing WOM analytics and the need for a sequential, multiview extension to canonical correlation analysis. We explain the practical steps involved in developing the method, discuss how we managed missing data, and outline ethical considerations in working with social media content. We also introduce two empirical settings from our original research that demonstrate the method's application: (1) the reputational crisis surrounding United Airlines during the Flight 3411 incident, and (2) partisan sentiment alignment on Twitter during the COVID-19 pandemic. These cases illustrate the method's value in detecting real-time spillovers, sentiment convergence, and dynamic shifts across multiple entities.Throughout, we reflect on practical and ethical challenges, highlight interpretability strategies, and offer guidance for researchers developing or applying complex analytical tools in digital and organizational contexts. |
|---|---|
| Physical Description: | 1 online resource |
| ISBN: | 9781036242411 1036242412 |