Notes:

We can see from the additional EDA above that out of the 5001 observations in this data set with 8 variables, half of which are numerical and the other are categorical variables. We can also show that 12 RevperEmp data points are of NA value, and each column contains a varying number of unique values. It is interesting to note that there are 52 unique data points in the State column, when we know that there are 50 states, which leads me to believe that terratories other than states are included in this dataset. We also see that there are 2 duplicated rankings - which may be interesting to explore the reason why that is in more detail because the data seemes really different for the companies that have the same ranking in this dataset. Upon deeper insection of the variable Growth_Rate we can see that it is clearly correlated to the Rank variable. Upon further inspection of the correlation between numeric variables we can also see that Growth_Rate and Rank are highly correlated (negative correlation), and Revenue and Rank also seem to be significantly correlated (positive correlation). Finally by using the describe() function from the psych library, we can uncover some more helpful information about the variables in this dataset (such as kurtosis, skew)

Question 1

Create a graph that shows the distribution of companies in the dataset by State (ie how many are in each state). There are a lot of States, so consider which axis you should use. This visualization is ultimately going to be consumed on a ‘portrait’ oriented screen (ie taller than wide), which should further guide your layout choices.

library(ggplot2)

q1 <- inc%>%
  group_by(State)%>%
  count(State)%>% #counts the number of company in each state
  arrange(desc(n))%>% #arranges them in decending order
  as_tibble(q1) #changes the format of the data for ggplot()


#graph with count of companies in each state
ggplot(q1, aes(x=reorder(State,n), y=n, fill = State))+
  geom_bar(stat="identity", width=.6, show.legend = FALSE)+
  geom_hline(yintercept=seq(1,800,100), col="white", lwd=1)+
  theme(panel.grid.major.x = element_blank())+
  coord_flip()+
  geom_text(aes(x = State, label = State, y = n +20), colour="blue", size = 2)+
  labs(y = "Count", x ="State", color="legend", title = "Fastest Growing Companies by State") + 
  theme(axis.text.y=element_blank()
        ,axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

Quesiton 2

Lets dig in on the state with the 3rd most companies in the data set. Imagine you work for the state and are interested in how many people are employed by companies in different industries. Create a plot that shows the average and/or median employment by industry for companies in this state (only use cases with full data, use R’s complete.cases() function.) In addition to this, your graph should show how variable the ranges are, and you should deal with outliers.

#initial group
NY <- inc %>% 
  filter(State== "NY") %>%
  filter(complete.cases(.)) %>%
  group_by(Industry)

#Initial boxplot with outliers
ggplot(NY, aes(Industry, Employees, fill = Industry)) + 
  geom_boxplot(outlier.colour="blue", 
               outlier.shape=4, 
               outlier.size=2, 
               show.legend=FALSE) + 
   theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
   coord_flip()+
  labs(title="Employees by Industry in NY - GRAPH A", y="Number of Employees")

#Boxplot without outliers (ignoring all points greeater than 1500)
ggplot(NY, aes(Industry, Employees, fill = Industry)) + 
  geom_boxplot(outlier.colour="blue", 
               outlier.shape=4, 
               outlier.size=2, 
               show.legend=FALSE) + 
  ylim(0,1500) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  coord_flip()+
  labs(title="Employees by Industry in NY - GRAPH B", y="Number of Employees")

#Specifically compare mean and median
NY2 <- NY%>%
  summarise(IndMean = mean(Employees),
            IndMedian = median(Employees))%>%
  gather(statistic, Employees, IndMean, IndMedian)

#Plot
ggplot(NY2, aes(x=Industry, y=Employees))+
  geom_bar(stat = "identity", aes(fill=statistic))+
  geom_hline(yintercept=seq(1,1700,100), col="white", lwd=1)+
  theme_classic()+
  coord_flip()+
  scale_fill_discrete(name = "Metric", labels = c("Mean", "Median")) +
  labs(title="Mean & Median Employees in NY - Graph C", y="Number of Employees")

#Biggest Skew 
NYskew <- NY2 %>%
  spread(statistic, Employees) 

NYskew <- NYskew %>%
  mutate(diff = IndMean - IndMedian)

ggplot(NYskew, aes(x=fct_reorder(Industry, diff), y=diff, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(diff, 2), y = diff +70), colour="blue", size = 3)+
  labs(title="(Mean - Median) Employees by Industry in NY - Graph D", y="Difference") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

#Mean Graph
NYmean <- NY2 %>%
  filter(statistic == "IndMean")

ggplot(NYmean, aes(x=fct_reorder(Industry, Employees), y=Employees, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(Employees, 2), y = Employees +70), colour="blue", size = 3)+
  labs(title="Mean Employees by Industry in NY - Graph E", y="Mean Number of Employees") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

#Median Graph
NYmedian <- NY2 %>%
  filter(statistic == "IndMedian")

ggplot(NYmedian, aes(x=fct_reorder(Industry, Employees), y=Employees, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(Employees, 2), y = Employees +10), colour="blue", size = 3)+
  labs(title="Median Employees by Industry in NY - Graph F", y="Median Number of Employees") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

Question 3

Now imagine you work for an investor and want to see which industries generate the most revenue per employee. Create a chart that makes this information clear. Once again, the distribution per industry should be shown.

#group by Industry and calculate the reveue per employee
revEmp <- inc %>%
  mutate(RevperEmp = Revenue/Employees) %>%
  filter(complete.cases(.)) %>%
  group_by(Industry)

#Boxplot to show distribution
ggplot(revEmp, aes(Industry, RevperEmp, fill = Industry)) + 
  geom_boxplot(outlier.colour="blue", 
               outlier.shape=4, 
               outlier.size=2, 
               show.legend=FALSE) + 
   theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
   coord_flip()+
  labs(title="Fastest Growing Companies in  USA", y="Revenue per Employee")

#Boxplot without outliers (ignoring all points greeater than 1500)
ggplot(revEmp, aes(Industry, RevperEmp, fill = Industry)) + 
  geom_boxplot(outlier.colour="blue", 
               outlier.shape=4, 
               outlier.size=2, 
               show.legend=FALSE) + 
  ylim(0, 10000000) +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  coord_flip()+
  labs(title="Fastest Growing Companies in  USA", y="Revenue per Employee")

#Specifically compare mean and median
revEmp2 <- revEmp%>%
  summarise(IndMean = mean(RevperEmp),
            IndMedian = median(RevperEmp))%>%
  gather(statistic, RevperEmp, IndMean, IndMedian)

#Plot
ggplot(revEmp2, aes(x=Industry, y=RevperEmp))+
  geom_bar(stat = "identity", aes(fill=statistic))+
  geom_hline(yintercept=seq(1,1700,100), col="white", lwd=1)+
  theme_classic()+
  coord_flip()+
  scale_fill_discrete(name = "Metric", labels = c("Mean", "Median")) +
  labs(title="Fastest Growing Companies in  USA", y="Revenue per Employee")

revEmp2

## Warning: `...` is not empty.
## 
## We detected these problematic arguments:
## * `needs_dots`
## 
## These dots only exist to allow future extensions and should be empty.
## Did you misspecify an argument?

## # A tibble: 50 x 3
##    Industry                     statistic RevperEmp
##    <fct>                        <chr>         <dbl>
##  1 Advertising & Marketing      IndMean     306036.
##  2 Business Products & Services IndMean     359097.
##  3 Computer Hardware            IndMean     817702.
##  4 Construction                 IndMean     465682.
##  5 Consumer Products & Services IndMean     466068.
##  6 Education                    IndMean     296454.
##  7 Energy                       IndMean    1554656.
##  8 Engineering                  IndMean     201120.
##  9 Environmental Services       IndMean     283607.
## 10 Financial Services           IndMean     394231.
## # ... with 40 more rows

#Biggest Skew 
revEmpSkew <- revEmp2 %>%
  spread(statistic, RevperEmp) 

revEmpSkew  <- revEmpSkew  %>%
  mutate(diff = IndMean - IndMedian)

ggplot(revEmpSkew , aes(x=fct_reorder(Industry, diff), y=diff, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(diff, 2), y = diff +80000), colour="blue", size = 3)+
  labs(title="Fastest Growing Companies in  USA (mean - median)", y="Difference") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

#Median Graph
revEmpMedian<- revEmp2  %>%
  filter(statistic == "IndMedian")

ggplot(revEmpMedian, aes(x=fct_reorder(Industry, RevperEmp), y=RevperEmp, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(RevperEmp, 2), y = RevperEmp +40000), colour="blue", size = 3)+
  labs(title="Revenue per Employee by Industry ", y="Median") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))

#Mean Graph
revEmpMean <- revEmp2  %>%
  filter(statistic == "IndMean")

ggplot(revEmpMean , aes(x=fct_reorder(Industry, RevperEmp), y=RevperEmp, fill = Industry))+
  geom_bar(stat="identity", show.legend = FALSE)+
  coord_flip() +
  geom_text(aes(x = Industry, label = round(RevperEmp, 2), y = RevperEmp +100000), colour="blue", size = 3)+
  labs(title="Revenue per Employee by Industry ", y="Mean") +
  theme(axis.title.y=element_blank(),
        axis.ticks.y=element_blank(),
        plot.title = element_text(hjust = 0.5))