Brand Comparison

Author

Francisco Mendiburu

Introduction

For this project we are going to answer a question many athletes have, what is the best athletic shoe brand for elite performing athletes? As an amateur athlete who has played a variety of sports, I can say when attempting to play a new sport I have always struggled to find a shoe that will help me perform my best. In this project we are going to look at a variety of factors that affect the performance of a shoe, and look at data sets to see what brands tend to fulfill these requirements more often.

Data Sets Being Used

To answer this question we used multiple data sets, the first one is a website called, “The Hoops Geek.” This website rates different basketball shoes, shows the price of each shoe, and gives an explication of why these shoes have the rating they were given.

The next data set that is going to be used is a website called, “Trustpilot.” This website is a page where customers come to give their opinions on their experiences with the companies.

Data Libraries

Data scraped from “The Hoops Geek” (shoe_reviews):

name = Name of the shoe
rating = Rating given to the shoe
price = How much the shoe currently costs
dates = When this shoes were launched
comment = This is the comment that the person that rates the shoes has given
price_group = In what price group does the shoe fall. Price increases by $50 each group.
brand = The brand that created the shoe
comfort = If comfort was mentioned in the description of the shoe
traction = if traction was mentioned in the description of the shoe
lightweight = if lightweight was mentioned in the description of the shoe

Data scraped from Trustpilot (shoe_df):

name = Name of reviewer
review = Review given
date = When the review was posted
Source = The company that the review belongs to

Loading in Libraries and Data Sets

When you click the Render button a document will be generated which includes both content and the output of embedded code. You can embed code like this:

Analysis

For the analysis, I will be asking multiple questions. Through graphs we will understand the questions asked.

What affects Consumer Preference?

When looking to buy a new shoe, no matter for what sport I played, I usually tried to get a shoe within these three categories: comfort, traction, and how light the shoe was. These things are important for a shoe to have because it helps improve performance. In the following visualization, we will observe which attributes have an greater impact on the rating of the shoe.

NULL

Interpretation: In this table, we are analyzing how much comfort effects the rating given to a shoe. We can see that when a shoe has higher comfort the rating seems to increase.

NULL

Interpretation: In this graph we analyze how much traction matters for the rating of a shoe. We can observe the more traction a shoe has the higher rating it will be given. This makes sense because for basketball shoes, tennis shoes, and any other hard court shoe, it is important to have shoes that do not make you slip around.

NULL

Interpretation: This last visualization helps us see how important having a lighter shoe is. We see that having a heavier shoe for a hard court sport gives you a better rating than having a lighter shoe. This surprised me because I tend to use lighter shoes because they help me move faster around the court. This measurement will also change depending on sports. For example when purchasing soccer cleats customers will most likely buy a shoe that is lighter than heavier.

Interpretation: I decided to analyze the words being used in the comments and saw besides the words “shoe” and “players,” the next most common words explain attributes we would expect any shoe to have. These words are, cushioning for comfort, traction, and quick for improvement in performance.

Overall Interpretation: As we have observed in the previous graphs, having a more comfortable shoe or having a shoe with better traction will give you a slightly better rating than if you did not have these qualities. On the other hand, if you have a heavier shoe it will give you a much better rating than if you have a lighter shoe. One thing we should note is these attributes affect ratings much more for basketball shoes. If you play a different sport these qualities might have a different impact on how the attributes affect the rating of a shoe.

Does the Price affect Performance?

This is something most people debate with because if a shoe is more expensive we tend to think it must perform a lot better. But is this really true? In the following graph we will analyze this.

Interpretation: In the visualization, we can observe the price of a shoe does have a slight effect on ratings the shoe receives. The cheaper the price of a shoe, the lower the rating will be on average. However, if you look through the data, shoes that have the highest rating are not the ones that cost the most; rather, these shoes happen to be around the $75 - $100 range. The reason why this is not shown in the data is because the average shoe in that price range tends to be of lower quality. But if a consumer takes time to do research about the product they are buying, they will be able to find affordable shoes that are better for the sport. Rather than buying a more expensive shoe that does not perform as well.

Does the Brand Affect Performance?

We are usually led to believe brands like Nike or Jordan have better quality shoes than any other brands; here we are going to compare if the brand name has an effect on how well a shoe performs.

Interpretation: In the visualization, we can observe how big named brands like Jordan, Nike, Under Armour, etc. do have very high performing shoes the website would recommend you to purchase. Having said this, they also do have some of the lowest performing shoes. Nike has a shoe with a ranking of 9 points on the performance scale, and at the same time, they created a shoe with a ranking of 7 points on the performance scale which is one of the lowest ratings given. There are other brands that are not as well known. Even though they might not have the highest rating, the average of the shoes are much higher than any other brand, having an average rating of 8.5 on the performance scale.

What is the General Brand Sentiment?

How customers view a brand is very important because it sets expectations about how they expect a shoe will perform. In this analysis we are going to look at the sentiment data from two of the most popular brands in the athletic industry.

# A tibble: 2,711 × 3
# Groups:   source [2]
   source word         n
   <chr>  <chr>    <int>
 1 Adidas adidas     137
 2 Puma   puma       131
 3 Puma   refund      73
 4 Adidas customer    66
 5 Puma   customer    65
 6 Puma   service     64
 7 Adidas shoes       59
 8 Adidas service     46
 9 Puma   received    46
10 Adidas refund      37
# ℹ 2,701 more rows

We can observe in these observations there is a very high positive sentiment for both of these brands. This is because customers understand this is a good brand, therefore, their products should be good quality. Another thing this graphs shows that is worth noting is how high the trust sentiment is on both of these brands. This is very good for the brands because this means customers trust their products. Something to keep in mind is the reason why the negative sentiment is so high is because the data was pulled from a place where people mostly review their experiences on their online shopping websites. It is important to know this data does not represent products, but the brand as a whole .

Conclusions

If one wants to purchase a high performing shoe, the moral of the story is to do one’s research. As we have seen before, there is much more to look at than the price tag and brand. Rather, key factors mentioned above (i.e. comfort, traction, weight, year, etc.) play a major role in the process of selecting the highest performing product. A brand might have a big name and people might trust their shoes, but this does ensure good performance.

--- title: "Brand Comparison" author: "Francisco Mendiburu" format: html editor: visual html-math-method: katex #Defines the pretty math render code-tools: TRUE #Allow the code tools option showing in the output. self-contained: TRUE # Embeds all components into a single HTML file execute: warning: False message: false echo: false --- # Introduction For this project we are going to answer a question many athletes have, what is the best athletic shoe brand for elite performing athletes? As an amateur athlete who has played a variety of sports, I can say when attempting to play a new sport I have always struggled to find a shoe that will help me perform my best. In this project we are going to look at a variety of factors that affect the performance of a shoe, and look at data sets to see what brands tend to fulfill these requirements more often. ## Data Sets Being Used To answer this question we used multiple data sets, the first one is a website called, "The Hoops Geek." This website rates different basketball shoes, shows the price of each shoe, and gives an explication of why these shoes have the rating they were given. The next data set that is going to be used is a website called, "Trustpilot." This website is a page where customers come to give their opinions on their experiences with the companies. ### Data Libraries Data scraped from "The Hoops Geek" (shoe_reviews): - name = Name of the shoe - rating = Rating given to the shoe - price = How much the shoe currently costs - dates = When this shoes were launched - comment = This is the comment that the person that rates the shoes has given - price_group = In what price group does the shoe fall. Price increases by \$50 each group. - brand = The brand that created the shoe - comfort = If comfort was mentioned in the description of the shoe - traction = if traction was mentioned in the description of the shoe - lightweight = if lightweight was mentioned in the description of the shoe Data scraped from Trustpilot (shoe_df): - name = Name of reviewer - review = Review given - date = When the review was posted - Source = The company that the review belongs to ## Loading in Libraries and Data Sets When you click the **Render** button a document will be generated which includes both content and the output of embedded code. You can embed code like this: ```{r} #load in the libraries library(tidyverse) library(tidytext) library(gutenbergr) library(ggwordcloud) library(textdata) library(rvest) library(lubridate) library(httr) library(stringr) library(readr) library(xml2) library(magrittr) ``` ```{r} # data from The hoops Geek shoe_reviews <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/mendiburuperezf_xavier_edu/EaZeKKAw1ldHsNscxfVFPRwB8YrWtyzhvWXQRtXZ_Jx0-g?download=1") # Data from Trustpilot shoe_df <- read_csv("https://myxavier-my.sharepoint.com/:x:/g/personal/mendiburuperezf_xavier_edu/EVALQtYAo8tFtrFB9av13WwBOFzkxT3hsWl2TGF-e4E9tw?download=1") ``` # Analysis For the analysis, I will be asking multiple questions. Through graphs we will understand the questions asked. ### What affects Consumer Preference? When looking to buy a new shoe, no matter for what sport I played, I usually tried to get a shoe within these three categories: comfort, traction, and how light the shoe was. These things are important for a shoe to have because it helps improve performance. In the following visualization, we will observe which attributes have an greater impact on the rating of the shoe. ```{r} #| echo: false # Visualization by Comfort shoe_reviews %>% ggplot(aes(x = as.factor(comfort), y = rating, fill = as.factor(comfort))) + geom_boxplot() + labs(title = "Rating by Comfort") labs(x = "Comfort", y = "Rating") + scale_fill_manual(values = c("0" = "red", "1" = "blue")) ``` **Interpretation:** In this table, we are analyzing how much comfort effects the rating given to a shoe. We can see that when a shoe has higher comfort the rating seems to increase. ```{r} #| echo: false # visualization by traction shoe_reviews %>% ggplot(aes(x = as.factor(traction), y = rating, fill = as.factor(traction))) + geom_boxplot() + labs(title =" Rating by Traction") labs(x = "Traction", y = "Rating") + scale_fill_manual(values = c("0" = "red", "1" = "blue")) ``` **Interpretation:** In this graph we analyze how much traction matters for the rating of a shoe. We can observe the more traction a shoe has the higher rating it will be given. This makes sense because for basketball shoes, tennis shoes, and any other hard court shoe, it is important to have shoes that do not make you slip around. ```{r} #| echo: false # Visualization by lightweight shoe_reviews %>% ggplot(aes(x = as.factor(lightweight), y = rating, fill = as.factor(lightweight))) + geom_boxplot() + labs(title = "Rating by Lightweight") labs(x = "lightweight", y = "Rating") + scale_fill_manual(values = c("0" = "red", "1" = "blue")) ``` **Interpretation:** This last visualization helps us see how important having a lighter shoe is. We see that having a heavier shoe for a hard court sport gives you a better rating than having a lighter shoe. This surprised me because I tend to use lighter shoes because they help me move faster around the court. This measurement will also change depending on sports. For example when purchasing soccer cleats customers will most likely buy a shoe that is lighter than heavier. ```{r} #|output: asis top_words <- shoe_reviews %>% mutate(report_id = row_number()) %>% unnest_tokens(word, comment) %>% anti_join(stop_words) %>% group_by(word) %>% summarize(n = n()) %>% arrange(desc(n)) %>% head(n = 10) # Create a ggplot bar plot top_words %>% ggplot(aes(x = reorder(word, n), y = n)) + geom_bar(stat = "identity", fill = "skyblue", color = "black") + labs(title = "Top 10 Words in Reviews", x = "Word", y = "Frequency") + theme_minimal() + theme(axis.text.x = element_text(angle = 45, hjust = 1)) ``` **Interpretation:** I decided to analyze the words being used in the comments and saw besides the words "shoe" and "players," the next most common words explain attributes we would expect any shoe to have. These words are, cushioning for comfort, traction, and quick for improvement in performance. **Overall Interpretation:** As we have observed in the previous graphs, having a more comfortable shoe or having a shoe with better traction will give you a slightly better rating than if you did not have these qualities. On the other hand, if you have a heavier shoe it will give you a much better rating than if you have a lighter shoe. One thing we should note is these attributes affect ratings much more for basketball shoes. If you play a different sport these qualities might have a different impact on how the attributes affect the rating of a shoe. ### Does the Price affect Performance? This is something most people debate with because if a shoe is more expensive we tend to think it must perform a lot better. But is this really true? In the following graph we will analyze this. ```{r} #| echo: false #Create a new row where prices are divided into groups shoe_reviews$price_group <- ifelse(shoe_reviews$price <= 50, "<= 50", ifelse(shoe_reviews$price <= 75, "51 - 75", ifelse(shoe_reviews$price <= 100, "76 - 100", ifelse(shoe_reviews$price <= 125, "101 - 125", ifelse(shoe_reviews$price <= 150, "126 - 150", ifelse(shoe_reviews$price <= 175, "151 - 175", ifelse(shoe_reviews$price <= 200, "176 - 200", "> 200"))))))) # Create a chart to show the relationship between the Price Group that a shoe is in # and the rating that they were given. shoe_reviews %>% ggplot(aes(x = price_group, y = rating)) + geom_bar(stat = "summary", fun = mean) + labs(title = "Rating by Price Group") ``` **Interpretation:** In the visualization, we can observe the price of a shoe does have a slight effect on ratings the shoe receives. The cheaper the price of a shoe, the lower the rating will be on average. However, if you look through the data, shoes that have the highest rating are not the ones that cost the most; rather, these shoes happen to be around the \$75 - \$100 range. The reason why this is not shown in the data is because the average shoe in that price range tends to be of lower quality. But if a consumer takes time to do research about the product they are buying, they will be able to find affordable shoes that are better for the sport. Rather than buying a more expensive shoe that does not perform as well. ### Does the Brand Affect Performance? We are usually led to believe brands like Nike or Jordan have better quality shoes than any other brands; here we are going to compare if the brand name has an effect on how well a shoe performs. ```{r} #| echo: false #Create a category where we can see the brand of the product shoe_reviews$brand <- ifelse(str_detect(shoe_reviews$name,"Nike"), "Nike", ifelse(str_detect(shoe_reviews$name,"Adidas"),"Adidas", ifelse(str_detect(shoe_reviews$name,"Puma"), "Puma", ifelse(str_detect(shoe_reviews$name,"Under Armour"),"Under Armour", ifelse(str_detect(shoe_reviews$name,"Jordan"), "Jordan", ifelse(str_detect(shoe_reviews$name,"Converse"),"Converse", ifelse(str_detect(shoe_reviews$name,"New Balance"),"New Balance","other"))))))) # Create a chart to show the relationship between A brand and the rating that it has shoe_reviews %>% ggplot(aes(x = brand, y = rating))+ geom_boxplot()+ labs(title = "Rating by Brand") ``` **Interpretation:** In the visualization, we can observe how big named brands like Jordan, Nike, Under Armour, etc. do have very high performing shoes the website would recommend you to purchase. Having said this, they also do have some of the lowest performing shoes. Nike has a shoe with a ranking of 9 points on the performance scale, and at the same time, they created a shoe with a ranking of 7 points on the performance scale which is one of the lowest ratings given. There are other brands that are not as well known. Even though they might not have the highest rating, the average of the shoes are much higher than any other brand, having an average rating of 8.5 on the performance scale. ### What is the General Brand Sentiment? How customers view a brand is very important because it sets expectations about how they expect a shoe will perform. In this analysis we are going to look at the sentiment data from two of the most popular brands in the athletic industry. ```{r} #| echo: false #| messages: false #| #|include: false # unnest tokes by brand and remove stop words tidy_shoes <- shoe_df %>% unnest_tokens(word, review) %>% anti_join(stop_words) #count words for each brand tidy_shoes %>% group_by(source, word) %>% summarize(n = n()) %>% arrange(-n) #Create new table nrc <- get_sentiments("nrc") nrc_shoes <- tidy_shoes %>% group_by (source, word) %>% summarize(n=n()) %>% inner_join(nrc) # NRC by Brand #NRC sentiment nrc_shoes %>% filter(source == "Adidas") %>% ggplot(aes(x = sentiment, y = n, fill = source))+ geom_bar(stat = "summary", fun = sum)+ facet_wrap(~source)+ labs(title = "Adidas Sentiment by Brand") ``` ```{r} #| echo: false #| messages: false #| #|include: false #Create new table nrc <- get_sentiments("nrc") nrc_shoes <- tidy_shoes %>% group_by (source, word) %>% summarize(n=n()) %>% inner_join(nrc) # NRC by Brand #NRC sentiment nrc_shoes %>% filter(source == "Puma") %>% ggplot(aes(x = sentiment, y = n))+ geom_bar(stat = "summary", fun = sum)+ facet_wrap(~source)+ labs(title = "Puma Sentiment by Brand") ``` We can observe in these observations there is a very high positive sentiment for both of these brands. This is because customers understand this is a good brand, therefore, their products should be good quality. Another thing this graphs shows that is worth noting is how high the trust sentiment is on both of these brands. This is very good for the brands because this means customers trust their products. Something to keep in mind is the reason why the negative sentiment is so high is because the data was pulled from a place where people mostly review their experiences on their online shopping websites. It is important to know this data does not represent products, but the brand as a whole . # Conclusions If one wants to purchase a high performing shoe, the moral of the story is to do one's research. As we have seen before, there is much more to look at than the price tag and brand. Rather, key factors mentioned above (i.e. comfort, traction, weight, year, etc.) play a major role in the process of selecting the highest performing product. A brand might have a big name and people might trust their shoes, but this does ensure good performance.