The objective of this case is to become familiar with the VA medical center appointment backlog data and the characteristics of the various medical centers. Since there are multiple variables to explore, we will focus on multivariate data visualizations. Examining the data table we see that the hospitals are identified by either Provider ID or Hospital Name. In some cases, Hospital Name is quite long, and when plotted the axis labels will take up a disproportionate amount of space on the graph. Provider ID, while shorter, is a unique identifier that is not informative. Since the data file is relatively small, a new variable could be created manually with shortened hospital names. Another option is to create a hospital identifier that contains the city and state where the medical center is located.

The data table does not have a column containing both city and state, but one can be easily created using the Formula Editor. Create a new column and invoke the Formula Editor either by double clicking the column header or right clicking the column header and selecting Column Properties >Formula. From the list of column names enter City and then select Character > Concat from the list of function groups. Enter “, “ and then concatenate State. Note that function groups can be opened using the toggle to the left of the function group name. The completed function is shown in Figure 9.1 Concatenating City and State Columns.

It is of interest to know if the VA medical centers are making progress towards reducing the backlog of appointments. To create the difference in the number of veterans waiting 31-60 days for an appointment, a new column should be created that subtracts the May 2015 backlog from the May 2016 backlog. Figure 9.2 Formula to Create the Backlog Difference shows the formula to create the new column, Backlog Difference.

It would be helpful to move the Backlog Difference column adjacent to the two columns that show the 2015 and 2016 backlogs. Select Cols > Reorder Columns > Move Selected Column as shown in Figure 9.3 Reordering a JMP Column and from the resulting Columns list move Backlog Difference after Appts May 1 2016 31-60 days.

The data file is now ready for analysis.