Date of Award

May 2015

Degree Type


Degree Name

Doctor of Philosophy


Management Science

First Advisor

Purushottam Papatla

Committee Members

Amit Bhatnagar, Janice Miller, V. Kanti Prasad, Ehsan Soofi


Branding, Constructive Choice, eWOM, Retweet, Social Media


Social networks have emerged as an important channel for brands to communicate with customers both directly as well as secondarily through customers who share the communications with others. A value of this channel to brands, therefore, depends on how effective they are in increasing the retransmission of their messages by customers. This is the issue that we investigate through three essays using the social media site Twitter as our research setting. Our investigation is based on the theory that retweeting is a choice made by consumers who rely on constructive preferences (Bettman, Luce and Payne 1998) while seeking three intangible benefits: altruism, self-enhancement, and social interaction. They also have two overarching metagoals of accuracy maximization and effort minimization (Bettman et al 1998) as they seek the benefits. Within this theoretical context, we examine the design attributes of tweets that increase retweeting. Specifically, we investigate more than 14000 tweets by 62 brands, across four product categories, over periods ranging from 18 to 400 days. Our empirical results are consistent with the theory and suggest that brand and tweet characteristics that increase the recipients’ ability to maximize the benefits of retweeting, and minimize the cognitive effort required to decide whether to retweet or not, increase retweets. The key managerial implication of our findings therefore is that brands should design tweets carefully to increase altruism, self-enhancement, and social interaction benefits while reducing the amount of effort that recipients need to undertake to assess whether the tweet offers these benefits. We replicate and extend these findings in essay two in the context of celebrities as brands by investigating the volume and duration of retweets of more than 2900 tweets by 65 celebrities across seven categories of the entertainment industry. Our results from this essay suggest that traits of the sources, i.e., the celebrities, also play a role in how recipients assess whether a tweet can deliver the three benefits while realizing the two metagoals. The third essay focuses on brands’ desire to generate retweets at a rapid rate before the tweet loses its relevance. In addition to volume and duration, therefore, we also investigate the rate at which a tweet is retweeted in this essay. Our investigation examines the retweet rates, in fifteen minutes intervals over a 24-hour period, of more than 2400 tweets posted by 62 celebrities using a Modulate Poisson Process model (Soyer and Tarimcilar 2008). Our results suggest that tweets that do not need recipients to interact with them and are related to significant cultural events are retweeted at a faster rate than others.

Included in

Marketing Commons