Create bivariate graph function

[scheidec]

Proposed syntax:

bivariate_plot(tbl, x, y, xlab, xylab)

where:

• tbl is the dataframe containing x and y
• x is the independent variable (bivariate only)
• y is the dependent variable
• xlab is the x-axis label which defaults to the label metadata but uses the variable name if empty
• ylab is the y-axis label which defaults to the label metadata but uses the variable name if empty

Outputs:

• If both x and y are continuous --> Scatterplot and loess fit
• If x is categorical and y is continuous --> Boxplot
• If both x and y are categorical --> Grouped bar graph where y-axis is labeled as "Within Group Percent (%)" and bars are labelled with counts
• If x is continuous and y is categorical --> Grouped bar graph where y-axis is labeled as "Within Group Percent (%)" and bars are labelled with counts. If number of x values > 15, then convert to deciles first.

Other features should include:

• If x is a vector, then the plot should facet_grid or facet_wrap by x