Title
Modeling individual and collective opinion in online social networks: drivers of choice behavior and effects of marketing interventions
Author
Koster, S.E.
Langley, D.J.
Publication year
2013
Abstract
We investigate factors influencing choice behavior in online social networks. We use twitter data from a Dutch television talent show. In study one, we implement a nested conditional logit model with latent classes. We find heterogeneous effects. For two latent classes, cognitive factors most strongly influence choice behavior. For two other latent classes, the influence of the television show is strongest. For all latent classes, social effects positively influence individual choice. In study two, an agent-based model uses the parameters estimated in study one, and simulates marketing interventions. Our simulations show that collective choice behavior may be influenced by increasing the amount of time a product is displayed to the individuals, although changing the sequence that candidates appear or increasing the activity of influentials has no effect.
Subject
Nested conditional logit
Latent classes
Agent-Based Model
Choice behavior
Online social networks
Opinion formation
ETP SON-M
Information Society
Organisation GI Innovation in Behaviour / Gedrag en Innovatie
SBA - Strategic Business Analysis
BSS - Behavioural and Societal Sciences ETP - Enabling Technology Programs
To reference this document use:
http://resolver.tudelft.nl/uuid:20540a87-bda2-4c65-8bc3-92ac78c29cdf
TNO identifier
477679
Source
42nd European Marketing Academy Conference Istanbul, 4-7 June 2013
Document type
conference paper