DATA 608 Data Viz

##           installed_and_loaded.packages.
## prettydoc                           TRUE
## psych                               TRUE
## knitr                               TRUE
## tidyverse                           TRUE
## ggthemes                            TRUE
## scales                              TRUE

Principles of Data Visualization and Introduction to ggplot2

I have provided you with data about the 5,000 fastest growing companies in the US, as compiled by Inc. magazine. lets read this in:

inc <- read.csv("https://raw.githubusercontent.com/charleyferrari/CUNY_DATA_608/master/module1/Data/inc5000_data.csv", header= TRUE)

And lets preview this data:

head(inc)

##   Rank                         Name Growth_Rate   Revenue
## 1    1                         Fuhu      421.48 1.179e+08
## 2    2        FederalConference.com      248.31 4.960e+07
## 3    3                The HCI Group      245.45 2.550e+07
## 4    4                      Bridger      233.08 1.900e+09
## 5    5                       DataXu      213.37 8.700e+07
## 6    6 MileStone Community Builders      179.38 4.570e+07
##                       Industry Employees         City State
## 1 Consumer Products & Services       104   El Segundo    CA
## 2          Government Services        51     Dumfries    VA
## 3                       Health       132 Jacksonville    FL
## 4                       Energy        50      Addison    TX
## 5      Advertising & Marketing       220       Boston    MA
## 6                  Real Estate        63       Austin    TX

summary(inc)

##       Rank                          Name       Growth_Rate     
##  Min.   :   1   (Add)ventures         :   1   Min.   :  0.340  
##  1st Qu.:1252   @Properties           :   1   1st Qu.:  0.770  
##  Median :2502   1-Stop Translation USA:   1   Median :  1.420  
##  Mean   :2502   110 Consulting        :   1   Mean   :  4.612  
##  3rd Qu.:3751   11thStreetCoffee.com  :   1   3rd Qu.:  3.290  
##  Max.   :5000   123 Exteriors         :   1   Max.   :421.480  
##                 (Other)               :4995                    
##     Revenue                                  Industry      Employees      
##  Min.   :2.000e+06   IT Services                 : 733   Min.   :    1.0  
##  1st Qu.:5.100e+06   Business Products & Services: 482   1st Qu.:   25.0  
##  Median :1.090e+07   Advertising & Marketing     : 471   Median :   53.0  
##  Mean   :4.822e+07   Health                      : 355   Mean   :  232.7  
##  3rd Qu.:2.860e+07   Software                    : 342   3rd Qu.:  132.0  
##  Max.   :1.010e+10   Financial Services          : 260   Max.   :66803.0  
##                      (Other)                     :2358   NA's   :12       
##             City          State     
##  New York     : 160   CA     : 701  
##  Chicago      :  90   TX     : 387  
##  Austin       :  88   NY     : 311  
##  Houston      :  76   VA     : 283  
##  San Francisco:  75   FL     : 282  
##  Atlanta      :  74   IL     : 273  
##  (Other)      :4438   (Other):2764

Non-visual exploration

Think a bit on what these summaries mean. Use the space below to add some more relevant non-visual exploratory information you think helps you understand this data:

Let’s create a table that shows the class types, NA counts & unique values of each variable

metadata <- function(df){
  ###Takes a data frame & Checks NAs, class types, inspects the unique values
  df_len <- nrow(df)
  NA_ct = as.vector(rapply(df, function(x) sum(is.na(x))))

  #create dataframe  
  df_metadata <- data.frame(
    class_type = rapply(df, class),
    n_rows = rapply(df, length),
    complete_cases = sum(complete.cases(df)),
    NA_ct = NA_ct,
    NA_pct = NA_ct / df_len * 100,
    unique_value_ct = rapply(df, function(x)length(unique(x))),
    most_common_values_sample = rapply(inc, function(x) str_replace(paste(names(sort(summary(as.factor(x)), decreasing=T))[1:5], collapse = '; '), "\\(Other\\); ", ""))
  )
  return(df_metadata)
}

kable(metadata(inc), digits = 2)

	class_type	n_rows	complete_cases	NA_ct	NA_pct	unique_value_ct	most_common_values_sample
Rank	integer	5001	4989	0	0.00	4999	3424; 5000; 1; 2
Name	factor	5001	4989	0	0.00	5001	(Add)ventures; @Properties; 1-Stop Translation USA; 110 Consulting
Growth_Rate	numeric	5001	4989	0	0.00	1147	0.4; 0.48; 0.53; 0.65
Revenue	numeric	5001	4989	0	0.00	1069	2100000; 2200000; 2400000; 4600000
Industry	factor	5001	4989	0	0.00	25	IT Services; Business Products & Services; Advertising & Marketing; Health; Software
Employees	integer	5001	4989	12	0.24	692	15; 25; 30; 12
City	factor	5001	4989	0	0.00	1519	New York; Chicago; Austin; Houston
State	factor	5001	4989	0	0.00	52	CA; TX; NY; VA; FL

Let’s use the psych package’s describe function to create a table that shows a more comprehensive set of summary statistics for each numerical variable.

metrics <- function(df){
  metrics_only <- df[, which(rapply(df, class) %in% c("numeric", "integer"))]
  df_metrics <- t(describe(metrics_only, quant = c(.25,.75)))[2:15,]
  return(df_metrics)
}

kable(metrics(inc), digits = 2, format.args = list(big.mark = ',', scientific = F, drop0trailing = T))

	Rank	Growth_Rate	Revenue	Employees
n	5,001	5,001	5,001	4,989
mean	2,501.64	4.61	48,222,535.49	232.72
sd	1,443.51	14.12	240,542,281.14	1,353.13
median	2,502	1.42	10,900,000	53
trimmed	2,501.73	2.14	17,334,966.26	81.78
mad	1,853.25	1.22	10,674,720	53.37
min	1	0.34	2,000,000	1
max	5,000	421.48	10,100,000,000	66,803
range	4,999	421.14	10,098,000,000	66,802
skew	0	12.55	22.17	29.81
kurtosis	-1.2	242.34	722.66	1,268.67
se	20.41	0.2	3,401,441.44	19.16
Q0.25	1,252	0.77	5,100,000	25
Q0.75	3,751	3.29	28,600,000	132

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.

ggplot(data = inc) + 
  geom_bar(mapping = aes(x = reorder(State, State, function(x) length(x)))) + 
  labs(x = "State",
       title = "Number of Companies by State") + 
  theme(axis.text.y = element_text(size=6, color="black", face = "bold"),
        plot.margin=unit(c(0,3,0,3), "cm")) +
  coord_flip()

Question 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.

state_third_most <- names(sort(table(inc$State), decreasing = T))[3]
df_third_most <- subset(inc, State == state_third_most)
df_third_most <- df_third_most[complete.cases(df_third_most),]

Plot 2.1

One way to visualize these distributions is to resize the plot limits so that it does not display the most extreme outliers.
However, some of the boxes are still too tiny to see.

#calculate some parameters to deal with the outliers
bpstats <- boxplot(Employees ~ Industry, data = df_third_most, plot = F)$stats
ylimits <- c(0, ceiling(max(bpstats) / 100)) * 100
ybreaks <- seq(ylimits[1], ylimits[2], by = 200)
outliers_not_shown <- paste(sum(df_third_most$Employees > max(ylimits)), "outlier(s) not displayed")

#to preserve alphabetical order when flipped
reordered_x_lab <- scale_x_discrete(limits = rev(levels(inc$Industry)))

plt_base <- ggplot(data = df_third_most, mapping = aes(x = Industry, y = Employees))

plt_base +
  geom_boxplot() +
  labs(title = "Company Sizes in NY by Industry",
       caption = paste("Red dot = mean", outliers_not_shown, sep = "\n")) +
  reordered_x_lab +
  scale_y_continuous(breaks = ybreaks) +
  stat_summary(fun.y=mean, geom="point", size=2, color = "red") +
  coord_flip(ylim = ylimits) +
  theme_fivethirtyeight()

Stackoverflow: Ignore outliers in ggplot2 boxplot

Stackoverflow: Boxplot show the value of mean

Plot 2.2

Another way to visualize these distributions is perform a log function in order to rescale the extreme outliers.
Now the extreme outliers can be seen.

plt_base +
  geom_boxplot() +
  reordered_x_lab +
  scale_y_log10(breaks = 10^(1:4)) +
  labs(title = "NY Company Sizes (log10) by Industry") +
  coord_flip() +
  theme_fivethirtyeight()

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.

Plot 3.1

I’m assuming that we are doing the revenue/employee division at the company-level granularity. Otherwise I believe that showing distribution per industry doesn’t make any sense.
I removed 12 incomplete cases.
First, let’s resize the plot limits so that it does not display the most extreme outliers.

#create data
inc_q3 <- drop_na(data.frame(
  Industry = inc$Industry,
  revenue_per_employee = with(inc, Revenue/Employees)
))

#calculate some parameters to deal with the outliers
bpstats <- boxplot(revenue_per_employee ~ Industry, data = inc_q3, plot = F)$stats
ylimits <- c(0, ceiling(max(bpstats) / 100000) * 100000)
ybreaks <- seq(0, ylimits[2], by = 500000)
outliers_not_shown <- paste(sum(inc_q3$revenue_per_employee > max(ylimits), na.rm = T), "outlier(s) not displayed")

#plot
plt_base2 <- ggplot(data = inc_q3, mapping = aes(x = Industry, y = revenue_per_employee))

plt_base2 +
  geom_boxplot(outlier.size = 1) +
  labs(title = "Distribution of Revenue $ per Employee",
       caption = paste("Red dot = mean", outliers_not_shown, sep = "\n")) +
  reordered_x_lab +
  scale_y_continuous(breaks = ybreaks, labels = comma(ybreaks)) +
  stat_summary(fun.y=mean, geom="point", size=2, color = "red") +
  coord_flip(ylim = ylimits) +
  theme_fivethirtyeight()

Plot 3.2

Now, let’s use the log10 scale so we can see the most extreme outliers.

breaks_log10 <- 10^(4:7)

plt_base2 +
  geom_boxplot(outlier.size = 1) +
  reordered_x_lab +
  scale_y_log10(breaks = breaks_log10, labels = comma(breaks_log10)) +
  labs(title = "Distr. of Revenue $ per Employee (log10)") +
  coord_flip() +
  theme_fivethirtyeight()