Sentiment Analysis of Tweets in Nike Ego Network
On September 3, Nike released an ad featuring Colin Kaepernick, who had been tapped to star in their 30th Anniversary Campaign. Nike quickly drew the ire of the far right due to the inclusion of Kaepernick, a former professional football player who leveraged his celebrity to engage in racial activism. Kaepernick’s choice to kneel during the national anthem to protest police violence angered groups that represented his peaceful protest as a direct insult against veterans. He has remained a controversial figure, and Nike’s choice to feature him was an endorsement of his activist work as aligned with their corporate values.
In this article, Forbes features a sentiment analysis of tweets about Nike following the initial announcement and subsequent release of more marketing materials. Nike is the ego in an ego network consisting of twitter users as nodes. This analysis is useful because it captures a type of interaction (an edge between the user node and ego node) that might get lost in a traditional model.
“Subtweeting” is a term used to describe tweeting about a person or account without tagging them (e.g. @nike). This method is utilized so that the subject of your tweet does not get a notification and must search for their name or handle in order to see the (usually negative) tweet. In the case of a multinational company like Nike, there is little difference between the choice to tag or not to tag them, but it is important when modeling the network because an ego network based on direct interactions like replies, likes, retweets, or @mentions is woefully incomplete. Twitter users might not bother tagging an account that will never see their tweet or might not want to make the account trend higher.
The sentiment analysis gives us an idea of the relationship between the tweeters and Nike. Negative sentiments, which were found to often include the hashtags #BoycottNike and #MAGA, likely indicate negative relationships, while positive sentiments (among which #JustDoIt was very popular) indicate positive relationships. It’s valuable for the company to be able to understand which possible customers have a positive view of the company and how they might be linked. Nodes tweeting about Nike forges new connections and expands the reach of the brand.
Many popular celebrities, like Lebron James and Serena Williams (who are both Nike athletes), voiced their support for Kaepernick and the campaign. If their fans dislike Kaepernick, their outspoken endorsement might strain their beliefs or call them into question, illustrating the instability of a triad with one negative and two positive relationships (fan –(+)– celebrity –(+)– Nike –(-)– fan). Similarly, to a fan of Kaepernick’s who believes that capitalism and racial inequality are inexorably linked, Kaepernick’s positive relationship with Nike might strain their relationship with him. New connections can create unstable triads that force a node to change their attitude towards another node.
This polarizing campaign elicited very strong negative reactions and an overwhelming backlash to the initial backlash, making it a very interesting study in the positive and negative edges in an ego network.