Choose one of David Robinson’s tidytuesday screencasts, watch the video, and summarise. https://www.youtube.com/channel/UCeiiqmVK07qhY-wvg3IZiZQ
You must follow the instructions below to get credits for this assignment.
Tidy Tuesday screencast: exploring US beer production
It was published April 1, 2020.
He found the data on githubusercontent.com and it comes form the Alcohol and Tabacco Trade Bureau and it is portrayed in 4 differnt sets, brewing materials, beer taxed, brewer size, and beer states.
Brewer size: Each row represents the number of brewers, total barrels, taxable removals, and total shipped based on year and brewer size. There are 137 observations for brewer size.
Variables and meaning: Year - calander year of data Brewer_size - Size of brewery n_of_brewers - number of brewers total_barrels - total barrels produced taxable_removals - barrales removed for sales total_shipped - barrels shipped
Beer Taxed: Each row is barrels/kegs or bottles/cans based on their tax status by year. There are 1,580 observations for this set.
Variables and meaning: Data_type - barrels produced Tax_status - sub total tax-free, sub total taxable, tax free taxable and totals Year - calander year Month - year of production Type - consumed on premises, for export, for vessels and aircraft, in barrels and kegs, in bottles and cans, in kegs, production, stocks on hand end-of-month, sub total tax-free, sub total taxable, tax determined/premises use, or total removals month_current - total taxes of the month Month_prior_year - taxes of the previous month ytd_current - total taxes of the current year ytd_prior_year - taxes of the previous year tax_rate - the rate of tax
Beer states: Each row represents a state and the recorded barrels and type per year. This set has 1,872 observations.
Variables and meanings: State - state where barrels were produced Year - calander year Barrels - number of barrels Type - bottles/cans, barrels/kegs, or on premises
Brewing materials: Each row represents the type and amount of brewing materials bought per year. There are 1,440 observations in this set.
Variables and meaning: Data_type - Pounds of product material_type - what type of product (grain, non grain, etc) year - calander year month - month purchased type - type of specific product (barley, hops, sugar, etc) month_current - amount purchased during the current month Month_prior_year - amount purchased udring the same month the previous year ytd_current - total purchased during the current year ytd_prior_year - total purchased in the previous year
Dave imported the data by copy and pasting the code into a new r chunk. After adding all of the sets to his program, he looked at the table for each set to get a better idea of the recorded data. He decided at the start of his video he would focus on brewing materials to better understand what goes into brewing beer. Using the filter function, he removed any data including the word total and also filtered the set to only show the most recently recorded month. This allowed him to see all of the materials purchased for the lates month without any unnecessary data.
He then filtered it agian to only show totals of grain and non-grain products, using a plot that had the two totals stack on top of eachother to represent total products per month purchased. After completing the graph to his satisfaction and having a graph that showed total pound sof product both separated as grain and non-grain as well as the individual products, he decide to start exploring the brewer size set to findout the brewer size distribution. Primarily figuring out the amount of barrels produced by larger breweries compared to small breweries. He also used the data set beer states to find where the majority of beer produced is condumed on premises. In each of these cases he went a similar route as he had done with brewing materials, only beer states differed where he made a density map to show which states had the highest percentage of beer produced being consumed on premises.
For almost every code chunk, he used the filter function to isolate the variables he wanted to focus and the ggplot function to graph the new filtered sets.
Breweries in the US buy a lot of product every year for beer production and grain products are purchased in much larger quantities than non-grain products.
The way his final graph was set up was awesome. He was able to put in descriptions at the top of his page that then put int tabs to show differnt ways to view total pounds of product.