Data for: What Makes Virtual Influencers Persuasive? A Structural Analysis of Endorsement Effectiveness
Description
This dataset supports the research findings presented in the manuscript titled "What Makes Virtual Influencers Persuasive? A Structural Analysis of Endorsement Effectiveness." It contains the empirical data collected via expert surveys and the subsequent computational matrices used for the Multi-Criteria Decision-Making (MCDM) analysis.
Files
Steps to reproduce
Data Collection Process: a) Factor Identification: A Systematic Literature Review (SLR) was conducted on Scopus and Web of Science databases to identify 28 critical success factors for Virtual Influencer (VI) endorsements. These were categorized into four dimensions based on Lasswell’s Communication Model (Source, Content, Receiver, Effect). b) Expert Panel Selection: A panel of 10 experts was recruited (5 from academia with 15+ years average experience; 5 from industry with 5+ years average experience in digital marketing/AI). c) Survey Administration: Experts completed a structured pairwise comparison questionnaire. They rated the direct influence of factor i on factor j using a 5-point linguistic scale (0 = No influence to 4 = Very high influence). Data Processing (Fuzzy MCDM): a) Fuzzification: The raw linguistic scores were converted into Triangular Fuzzy Numbers (TFNs) to handle human subjectivity. b) Defuzzification: The CFCS (Converting Fuzzy Data into Fuzzy Scores) method was applied to convert TFNs back into crisp values for analysis. c) Fuzzy DEMATEL Analysis: These crisp values were used to compute the Initial Direct Relation Matrix, Normalized Direct Relation Matrix, and Total Relation Matrix (T). The sum of rows (d) and columns (r) were calculated to determine the "Prominence" (d+r) and "Relation" (d-r) for each factor. d) Fuzzy DANP Analysis: The Total Relation Matrix was normalized and transposed to create the Unweighted Supermatrix. This was raised to limiting powers to achieve convergence (Limit Supermatrix), generating the final global weights for all 28 factors. Software Used: Microsoft Excel, R-Studio
Institutions
- National Institute of Technology Kurukshetra