Variable names and contents

• roads: The R dataframe containing the original data
• roads.2011: The 2011 column of the roads dataframe
• not.na: An array of logical that corresponds to the non-NA values of the 2011 column
• roads.2011.cleaned: The 2011 column from the roads dataframe with the NA values removed
• total.2011: The sum of the 2011 values
• roads.num: The roads dataframe without the first column (just the numerical data)
• roads.means: A vector containing the mean value of each row
• roads.keep: A vector of logical that is True for rows for which the mean is less than 2000 (non-outliers)
• roads2: The roads dataframe with outliers removed
• roads.num2: The roads dataframe with the first column removed (just the numerical data) and the outliers removed
• roads.means2: The vector of means with outliers removed
• roads.num2.rowsums: The sum of the values in each row with the outliers removed (where a sum of 0 indicates that all values in the row are NA)
• roads.keep2: A logical vector used to index the rows for which there is at least one non-NA value
• roads3: The roads dataframe with outliers removed and the rows with all NA values removed
• roads.num3: The roads dataframe with the first column removed (just the numerical data) and the outliers removed, and with rows containing all NA values removed
• roads.means3: The vector of means with outliers removed, and with rows containing all NA values removed
• roads.replace.na: The roads dataframe with outliers removed, the rows with all NA values removed, and the NA values replaced with 0
• impute: A function that takes a vector with NA values and a vector with replacement values of the same length, and returns the original vector with the NAs replaced
• roads.impute.na: The roads dataframe with outliers removed, the rows with all NA values removed, and the NA values replaced with the row mean