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.
The screencast that I chose is called “Tidy Tuesday: Exploring U.S. Beer Production”.
The screencast was published on April 1st, 2020.
Hint: What’s the source of the data; what does the row represent; how many observations?; what are the variables; and what do they mean?
The data in this anaysis is from the Alcohol and Tobacco Tax and Trade Bureau. It is broken down into different categories consisting of brewing materials, beer taxed, brewer size, and data for each state. The brewer size section shows rows of the size of the brewer per year, based on how many barrels of beer are produced. The variables include number of brewers, total barrels, taxable removals, and total shipped. The brewing materials section displays rows containing pounds of materials, with variables such as material type, the year and month which they were used, and the previous months data as well. The tax data shows the year and months production of beer and what the taxes were on different amounts of barrels. This is useful for seeing how much beer was sold. The last section shows rows by state using year and number of barrels as variables.
Hint: For example, importing data, understanding the data, data exploration, etc.
The first thing Dave did was import the data from github by pasting the data set into R. Then he organized it so it was easier to read. Dave quickly looked through each of the four data sets to get a sense of what he was going to do with them. The first thing he did while exploring the data was write down questions that he was going to later answer. The first question was “What ingredients are used in beer?”. To answer the question, he started to put the data into charts to visualize it. Dave looked at each ingredient that was included in the data set then graphed them to see the largest vs. smallest volumes. After playing around with different charts and scales, he used a bar graoh that showed the use of materials over timne for each different ingredient. Dave continued these steps for the other data sets regarding beer production as well.
Yes, there were many things from the course that he used in this project. Since he used a lot of graphs there was constant coding of ggplot and geom functions. This was highly used in our course. Another thing that he included was answering his own questions about the graphs that he had, which is what our class did as well. It was similar to the mosaic plots we analyzed in quiz 3-b.
One of the major findings I found interesting was the amount of beer that was consumed on the premesis of the breweries. To see this, Dave created a map of the United States and used the individual states data for on premesis consumption. Then he made the colors darker for low consumption and lighter for high consumption. States such as Oklahoma and North and South Dakota had very high on premesis comsumption, reaching about 75% of beer consumed. This was a huge finding, as I did not know jjst how much beer was consumed at some of the breweries.
The most interesting thing about the analysis was the amount of data that Dave was able to analyze for this topic. Although the video was titled for beer production, there were hundreds of data points to look at such as ingredients, time of year or month when beer is consumed the most, beer consumed ojn the premesis of the breweries, and all of the tax information for beer sold. For me, it was very interesting to see his thought proccess and how he tackled a massive data set.