DDS6306_CaseStudy01

Git repository project entitled "Case Study 1." Members of this project include Heindel Adu, Gabriel Gonzales, and Amy Markum.

Introduction: This presentation provides an overview of exactly what type of manipulations were performed on two combined data sets (Beer by State and Breweries by State). Client has requested answers to the five following questions:

Number of Breweries per state

Number of NA values returned for each data set

Median ABV and IBU counts for each state, and provide a comparative bar chart

States with maximum ABV and IBU, respectively

Determination of any relationship between ABV and IBU, and provide comparative scatter plot

First, it is necessary to merge the tables using a specific code in r

Brew_Beer_Table <- right_join(Breweries, Beers, by="Brew_ID")

Second, it's a good idea to save your file before manipulating the data, in case you need to return to the original data

write.csv(Brew_Beer_Table, file = "Breweries_Beer.csv")

Third, you should rename all columns to be easily human readable

names(Brew_Beer_Table) <- c("Brewery_ID", "Brewery_Name", "City", "State", "Beer_Name", "Beer_ID", "ABV", "IBU", "Beer_Style", "Ounces")

Any data should be sorted to assist in understanding the data set, in this case sorted first by Brewery ID, then by ABV (ascending)

attach(Brew_Beer_Table)

SortedBreweryTable <- Brew_Beer_Table[order(Brewery_ID, ABV), ]

write.csv(SortedBreweryTable, file = "Breweries_Beer.csv")

It's always a good idea to review your dataset to ensure everything is as it should be

head(SortedBreweryTable, 10)

tail(SortedBreweryTable, 10)

Next we begin to answer the client's questions: Identify all of the NA's: (62 in the ABV column, 1005 in IBU)

sapply(SortedBreweryTable, function(x) sum(is.na(x)))

Then identify how many breweries are in each state

StateBreweryTotals <- SortedBreweryTable %>% group_by(State) %>% summarize(count=n())

Any data that is separated out and may be needed for later mainpulation should be stored in its own file

###write.csv(StateBreweryTotals, file = "Breweries-State.csv")

We provide a barplot indicating the number of breweries per state

ggplot(StateBreweryTotals, aes(x=State, y=count)) + geom_bar(stat="identity")

Identify the overall median ABV and IBU counts

ABV <- Breweries_Beer$ABV

median(ABV, na.rm = TRUE)

IBU <- Breweries_Beer$IBU

median(IBU, na.rm = TRUE)

Split out your data so you can work with your ABV & IBU variables by state

ABVState <-split(Breweries_Beer$ABV, Breweries_Beer$State)

IBUState <-split(Breweries_Beer$IBU, Breweries_Beer$State)

Identify your medians by state (example given)

median(ABVState[["CA"]], na.rm = TRUE)

median(IBUState[["CA]], na.rm = TRUE)

Create your new file with your states/median ABV information

write.csv(ABVStates, file = "ABVStates.csv")

Create your new file with your states/median IBU information

write.csv(IBUStates, file = "IBUStates.csv")

Provide a summary of the ABV statistics

str(ABVStates)

Provide a scatter plot of your ABV & IBU Levels

plot(IBULvl, ABVLvl, main = "ABV & IBU Comparison Chart", xlab="IBU Levels", ylab="ABV Levels", pch = 19, frame = FALSE)

Provide a comparative bar chart of the median ABV & IBU Levels

data <- data.frame(States, ABVLvl, IBULvl)

plot_ly(data, x = ~States, y = ~ABVLvl, type = "bar", name = "ABV Level") %>% add_trace(y=~IBULvl, name="IBU Level") %>% layout(yaxis = list(title = "ABV&IBU Comparison"), barmode = "Group")

NOTE: Although this was requested, a comparative barchart is inappropriate for ABV & IBU Levels due to the difference in scale