Answer to exercise at descriptive stats:

Input

carbon <- read.csv("CarbonDioxideEmission.csv")

carbon <- na.omit(carbon)

carbon$CarbonDioxide <- as.numeric(carbon$CarbonDioxide)
carbon$year <- as.numeric(carbon$year)

head(carbon)
tail(carbon)

carbon  %>%
  group_by(year) %>%
  summarise(avg_emission = mean(CarbonDioxide))

Can we merge data?

We now have 2 interesting datasets:

• mydata - we can obtain the average temperature of each year

• carbon - we can obtain the average carbon emission of each year

Each dataset contains part of the data needed to answer our research questions:

• H₀₁ the earth average temperature has not dramatically increased since the advent of electronics

• H₀₂ the emission of carbon dioxide does not influence the earth average temperature

However, we may need to merge the two datasets to answer our research questions!

Notice that both datasets contain a column for the year. Let’s use year to merge the data!

merge( ): Merge two datasets using one or more columns

Input

carbon <- group_by(carbon, year) %>% # 1
    summarise(AverageCarbonEmission = mean(CarbonDioxide)) # 2

newdata <- group_by(mydata, year, era) %>% # 3
    summarise(AverageTemperature = mean(AverageTemperature)) # 4

carbon <- merge(newdata, carbon[, c("year", "AverageCarbonEmission")], by="year") # 5

A detailed explanation explained line-by-line:

1. the carbon data will be updated with the result of the right-hand operation. The right-hand operation groups the dataset by year

2. after grouping, we apply the summarise to create a column named carbon the value of the column is the mean of the CarbonDioxide emission

3. we do a similar grouping operation creating a newdata data.

4. this time, we are interested in the mean of the AverageTemperature

5. we now have two variables, one with the year and AverageCarbonEmission and another with year and AverageTemperature. Let’s merge these two tables in a final carbon table.

   • by indicates the key column to merge the two datasets

     • Our merging criteria is the year

   • We will copy two columns from the initial carbon table using carbon[, c("year", "AverageCarbonEmission")]

     • this is the same as select(carbon, year, AverageCarbonEmission)