📈 Visualizations

In effort to support data exploration and codebook making, the library also contains functions to generate single variable descriptive graphs based on a variable’s class.

For example, take the County_Equivalent_Name variable from the hpsa_primarycare data set, which is a character. By passing this data set and variable to the visualize_sngl_var() we get a graph with bars presenting the count of each variable. More specifically, the function takes unsummarized data and returns summarized scores where each observation contributes to one unit of height for each bar. Lastly, in the example below, we set the flip argument to TRUE in order to pivot the graph on its x-axis.

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "County_Equivalent_Name",
                             xlab = "County",
                             flip = TRUE)

In the prior example, the df, var, and xlab arguments are requirements, the rest are optional. Among other optional features is the ability to change the y-axis labels and setting a cuttoff for larger data somewhat easily:

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "County_Equivalent_Name",
                             xlab = "County",
                             flip = TRUE,
                             ylab = "Count of apperances",
                             cutoff = 20)

Another optional feature is the ability to pass the function custom ggplot themes:

my_theme <- ggplot2::theme_minimal()

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "County_Equivalent_Name",
                             xlab = "County",
                             flip = TRUE,
                             ylab = "Count of apperances",
                             cutoff = 20,
                             theme = my_theme)

Note that the function automatically adapts to working with numeric variables:

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "HPSA_Score",
                             xlab = "HPSA SCORE")

Under the hood, the function acknowledges that the declared variable is continues and proceeds to generate a histogram, which bins the data, then counts the number of observations in each bin. You can control the width of the bins with the binwidth argument. It is very important to experiment with the bin width. The default just splits your data into 30 bins, which is unlikely to be the best choice. You should always try many bin widths, and you may find you need multiple bin widths to tell the full story of your data. For example:

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "HPSA_Score",
                             xlab = "HPSA SCORE",
                             binwidth = 10)

Or:

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "HPSA_Score",
                             xlab = "HPSA SCORE",
                             binwidth = 1)

Out of the two experiments above, it appears that a binwidth of 1 fits the data a little nicer. Let’s polish this graph up a little like we did above:

rcahelpr::visualize_sngl_var(df = rcahelpr::hpsa_primarycare,
                             var = "HPSA_Score",
                             xlab = "HPSA Score",
                             ylab = "Count",
                             binwidth = 1,
                             theme = my_theme)