README.md
In Reed-Math241/denverweed: Marijuana Gross Sales in Denver, CO 2014-2017

denverweed

The goal of denverweed is to provide a simple dataset for analysis which shows the different uses of wide/long form data, while giving insight into the trials and tribulations undergone by America’s favorite smokeable/tokeable 5-to-13 leafletted plant

The development version of denverweed is available from GitHub with:

install.packages("devtools")
devtools::install_github("Reed-Math241/denverweed")

The data were collected and made available by the City of Denver Open Data Catalog.

the denverweed package contains two datasets.

library(pkgGrpm)
data(package = 'pkgGrpm')

It contains the dataset weed, which contains 4 variables and 90 observations:

head(weed)
#>   YEAR     MONTH          GROSS_SALES_TYPE GROSS_SALES
#> 1 2017 SEPTEMBER Medical Total Gross Sales    18314027
#> 2 2017 SEPTEMBER  Retail Total Gross Sales    34950895
#> 3 2017    AUGUST Medical Total Gross Sales    19043315
#> 4 2017    AUGUST  Retail Total Gross Sales    35240979
#> 5 2017      JULY  Retail Total Gross Sales    34815762
#> 6 2017      JULY Medical Total Gross Sales    17922959

It also contains weed_wider, a pivoted, untidy version of weed which may be useful for certain graphs, containing 4 variables and 45 rows

head(weed_wider)
#>   YEAR     MONTH Medical Total Gross Sales Retail Total Gross Sales
#> 1 2017 SEPTEMBER                  18314027                 34950895
#> 2 2017    AUGUST                  19043315                 35240979
#> 3 2017      JULY                  17922959                 34815762
#> 4 2017      JUNE                  17615041                 31908811
#> 5 2017       MAY                  18276904                 30671594
#> 6 2017     APRIL                  19070685                 30302993

of course, we’re curious to see how marijuana sales have changed over the years. here we plot gross sales by year, faceted by medical or retail

library(tidyverse)

ggplot(weed, aes(x = YEAR, y = GROSS_SALES)) + 
  geom_bar(stat = 'identity', fill = '#709B40') +
  facet_wrap(~GROSS_SALES_TYPE) +
  scale_y_continuous(labels = scales::comma) + 
  ylab("Gross Sales ($)") + 
  xlab("Year")+
  theme_bw()

if we plot sales by month, we can add more detail. here we plot gross sales by month, faceted by year. we can see a general increase in marijuana sales.


ggplot(weed, aes(x = MONTH, y = GROSS_SALES)) + 
  geom_bar(stat = 'identity', fill = '#709B40') +
  facet_wrap(~YEAR) +
  scale_y_continuous(labels = scales::comma) +
  ylab("Gross Sales ($)") + 
  xlab("Month") +
  theme_bw()+
  theme(axis.text.x = element_text(angle = 90))

now we’d like to see the difference in medical/retail sales over time, by month. here we plot gross sales by month, faceted by year and retail/medical

ggplot(weed, aes(x = MONTH, y = GROSS_SALES)) + 
  geom_bar(stat = 'identity', fill = '#709B40') +
  facet_grid(GROSS_SALES_TYPE ~ YEAR) +
  scale_y_continuous(labels = scales::comma) + 
  ylab("Gross Sales ($)") + 
  xlab("Month") +
  theme_bw()+
  theme(axis.text.x = element_text(angle = 90))

finally, a use for weed_wider in a scatterplot of medical vs retail sales, by year


ggplot(weed_wider, aes(x = `Medical Total Gross Sales`, y = `Retail Total Gross Sales`, color = YEAR)) +
  geom_point() +
  scale_y_continuous(labels = scales::comma) +
  scale_x_continuous(labels = scales::comma) +
  scale_color_discrete(name = "Year") +
  theme_bw()