Nike vs. Adidas is one of the longest running rivalries in the business world. These company both sell a variety of athletic products and many people poor their money into these companies. Yet, both have also dealt with scrutiny in the past as all large companies do. Currently, both companies have fallen on relatively hard times. Nike is getting a lot of hate because they dropped their deal with Kyrie Irving and Odell Beckham Jr. just filed a 20M dollar lawsuit against them. Adidas is struggling because they put a lot of time and money into Kanye and he’s also getting a lot of hate at the moment.

So, through some sentiment analysis, let’s see which company is currently fairing better in the public eye.

Question 1

The first question I want to answer is, on a very broad level, “Which company are the majority people saying more good things about than bad? This will serve as a solid staring point for the following questions.

I will answer this question, and all subsequent questions by scraping and analyzing the last ~2000 tweets about these companies. To get these specific tweets I searched for the last ~2000 tweets that included the word “nike” or the word “adidas.” Then, I will take these tweets and run each individual word through the Bing sentiment lexicon to assess if the words have a positive or negative sentiment.

Here are the graphs of the positive and negative words that were used more than 10 times in tweets about Nike and Adidas, respectively.

nike_counts %>%
  filter(n > 10) %>%
  mutate(n = ifelse(sentiment == "negative", -n, n)) %>%
  mutate(word = reorder(word, n)) %>%
  ggplot(aes(word, n, fill = sentiment)) +
  geom_col() +
  coord_flip() +
  labs(title= "Nike Sentiment", y = "Contribution to sentiment")

At first glance it would seem like Nike is doing much better than Adidas. I would argue that it is closer than it seems. While Nike has more positive words they only have a contribution score of 25, while Adidas has four words that have greater contribution scores which means they are used more often. Adidas still has more negative words, but it is definitely more balanced than that first glance. I would claim that Nike is generally doing better in the public eye at the moment.

Question 2

My next question is similar to the first, but more specific. I want to know the exact types of emotions that are being expressed towards these companies. Positive and negative words are, not only broad, but also relatively subjective so hopefully this question will be more specific and objective

Similar to the first question I will grab the last ~2000 tweets that mention Nike or Adidas. Then I will take each individual word, but this time I will run them through the NRC sentiment lexicon which will label them with a list of emotions that they convey. Then I will seperate them and sum the number of words in each emotion.

Here are the resulting graphs:

brands%>%
  inner_join(nrc) %>% 
  group_by(brand, sentiment) %>% 
  summarize(n = n()) %>% 
  ggplot(aes(sentiment,n))+
  geom_col() +
  facet_wrap(~brand)+
  theme(axis.text.x = element_text(angle = 30, hjust=0.5, vjust =0.5))

Now these graphs are much more complex and yield more interesting results. First, let’s look at the “negative” and “positive” columns and compare them to what we saw in the last graphs. This lexicon shows that tweets containing “adidas” have more positive words in them than tweets that have “nike” in them. Yet, on the flip side, there are more negatives on the Adidas side as well. It seems that the ratio of positive to negative words is about the same for both companies which is close to what we saw in the last graphs.

Next, I want to point out the “trust” column. I find this very interesting because all the other emotions are about equal for each company. Yet, when you look at the trust column, Adidas has about 75 more occurrences of words that convey trust. People seem to like the companies equally, but they trust Adidas more.

Question 3

Finally, for my last question, I wanted to try to get away from all the drama and get back to the essence of these companies. These companies are both best known for their shoes and despite any drama they continue to sell hundreds, if not thousands of pairs of shoes each day. So, I want to know which company has more buzz around their shoes.

My approach to this was slightly different than the last two questions. For this I went and scraped ~1000 tweets that contained the words “nike shoes” or “adidas shoes.” Then I realized that if people are tweeting about new shoes most of the time is complimentary. Even when the tweets are not positive, “Any publicity is good publicity.” So I decided to focus, instead, on the influx of tweets and their consistency.

It is important to note that I am creating these graphs at 4PM on November 15th

Nikes_tweets %>% 
  ggplot(aes(as_datetime(created_at)))+
  geom_histogram()+
   labs(title = "NIKE",x = "Tweet Time")

The most important thing on this graph is the x-axis. In order to get the ~1000 tweets that I requested, the Nike graph only had to go back to about noon on November 13th. On the other hand, the Adidas graph had to go back to November 6th and still only returned 715 tweets. This clearly shows that Nike shoes are talked about significantly more often than Adidas shoes are.