Tag: Dave Markowitz

Presenter(s): Franziska Roth

Faculty Mentor(s): Dave Markowitz

Poster 141

 Session: Social Sciences & Humanities

In this research project, we evaluated the effect of hashtag frequency on musicians’ Twitter engagement. Our project specifically analyzed musicians with albums from the Billboard Top 200 list who actively posted on Twitter in 2017. In total, we scraped Twitter timelines from 93 musicians, including a sample of 53,299 Tweets using a software for statistical computing called R. The program allowed us to observe all 53,299 tweets at once, including metadata (username, engagements, date posted, etc.) Drawing on prior work that suggests that excessive hashtag use may be perceived as inauthentic (e.g., the concept of wear-out from marketing; Danaher, 1996), we found support for the idea that artists with more hashtags per post (as a proportion of their followers) had significantly fewer Twitter engagements. Therefore, the more hashtags an artist uses, the fewer engagements they will receive on their Tweets and this negative relationship was robust across musicians. Our findings emphasize the importance of using digital media traces to form perceptions about people and musicians online. We anticipate that this information can have reputational and financial impacts on artists as brands and change how social media strategy is pursued in the music settings.

Presenter(s): Brittani Lancaster

Faculty Mentor(s): Dave Markowitz

Poster 156

Session: Social Sciences & Humanities

What is the relationship between using #ad in social media posts and Twitter engagement? Consistent with prior work suggesting that consumers prefer to receive insight from unsponsored rather than sponsored advertisements, we predicted that in a comparison of Tweets, those including #ad would have fewer favorites than Tweets that do not contain #ad. We performed a case study of Kendall Jenner’s Twitter account (N = 3,200 Tweets) and used RStudio’s rtweet package to scrape the Twitter data from her feed. We ran a t-test, comparing the mean number of favorites per Tweet for those that had #ad and those that did not. The average number of favorites for Tweets with #ad was more than double the average number of favorites on her tweets without #ad (p = .0096). The results from this research were statistically meaningful but inconsistent with our prediction. We believe these results suggest that consumers respond well to posts that are clearly distinguished as sponsored advertisements because there is no deception occurring. We offer theoretical explanations for these data and future work should test this contention experimentally.

Presenter(s): Gillian George

Faculty Mentor(s): Dave Markowitz

Poster 140

Session: Social Sciences & Humanities

Does a brand’s engagement in Corporate Social Responsibility lead to higher financial returns and positive consumer engagement? This study evaluated four Fortune 500 companies (Campbell’s, Pepsi, Nike, and Kellogg’s) that made public ethics statements on current social issues (e.g., same-sex couples, interracial families, among others). We analyzed these companies within a two-week period, one week before and one week after their ethics statements were made, to understand if their social media activity (e.g., the number of favorites per Tweet) and stock price changed after the campaign period. We gathered Twitter data computationally through the R computing interface to collect number of favorites for each company before and after making their ethical statements public. We also evaluated the language data of each Tweet and performed an automated text analysis with Linguistic Inquiry and Word Count (LIWC) software to evaluate changes in language patterns related to authenticity (e.g., first-person singular pronouns) from before and after the ethical statements. The language data suggest that there was a marginally significant increase in authenticity from before to after the ethics stance (p = .075). However, stock price was not significantly different after evaluating company performance before or after the ethics campaign and stance (p = .987). We propose that changes in authenticity relate to a credibility mechanism as prior work suggests that people attempt to boost their credibility by using more “I” statements. We did notice a 10% increase in levels of authentic language as well as no change in stock price. Overall, there was no significant findings to prove a correlation between stock price, twitter engagement, and level of authenticity in language when comparing Tweets before and after an ethical campaign.

Presenter(s): Giselle Andrade

Faculty Mentor(s): Dave Markowitz

Poster: 139

Session: Social Sciences & Humanities

A lack of political activism among African American athletes over the past two decades has drawn widespread criticism. Critics posit that the modern-era African American athletes’ accumulation of wealth has influenced the de-politicization of sports. Our study directly tests the applicability of this narrative. We used the Twitter patterns of former National Football League quarterback Colin Kaepernick, as a case study to understand the Black Lives Matter social movement. We posed the following research question: Were Kaepernick’s political tweets about #BlackLivesMatter more influential than his non-political tweets? To evaluate this question, we created a dictionary of words that contained political speech as reflected by Kaepernick’s Twitter feed. We then used the automated text analysis program, Linguistic Inquiry and Word Count, to count the rate of political speech in Kaepernick’s tweets that were scraped computationally in the computing environment, RStudio. Regression tests analyzed the relationship between Kaepernick’s political speech and engagements, defined as the rate of favorites and retweets per tweet. We found political speech did not affect the level of engagement of favorites (p = .65). However, the rate of political speech was related to the number of retweets per tweet (p = .056), and for each percent increase in political speech, Kaepernick’s tweets received nearly 130 fewer retweets. We believe that these data suggest retweets are a more critical degree of expression than favorites.