Tableau Usage Examples

1. Create a map chart to show the number of banks in each zip code (variable name “zip”. If you see more than one zip code for a bank, use the first one).

[pic 1]

I have created a symbol map chart (geographical) by taking Zip and Name dimensions. In order to find the number of banks in each zip code, I have used the count function.

It can be seen that the number of banks are more on the east coast as compared to the west coast.

Also for example, zipcode 62401 has 3 banks in it whereas 85390 zipcode has only one bank on the west coast.

1. Create a map chart to show the median ROE (variable “roe”) for banks headquartered in each state (“state”).

[pic 2]

I have added geographic role to the state dimension (stalp) and added the median function to the ROE measure. I have selected the filled map chart and used colors where:

A state MT has ROE over positive 12 but state AZ which is in RED is in negative.

1. Create a bar chart with “state” on the horizontal axis, and trellis based on FDIC regions (“fdicdbs”) (i.e. one bar charts for each FDIC region). The height of the bars should be the number of banks headquartered in each state, and the color of the bar should reflect the variance of ROE among those banks.

[pic 3]

I have taken fdicdbs and stalp dimensions as my columns and I am counting the No of records (measure) in the rows. Along with this I am including the ROE variance across.

For example, the variance of CO state in the Dallas region is the highest with 156 banks.

1. Create a pivot table to show the variance of ROA (“roa”) across banks for each combination of two variables: (1) FDIC region (“fdicdbs”) and (2) bank charter class (“bkclass”)

[pic 4]

I have used text table to create this pivot table by setting fdicdbs dimension as the column and bkclass dimension as the row with the data for each relation coming in from the ROA measure. I also added the variance function on ROA. For example for bkclass of SA for region Chicago, the variance of ROA is 223.

1. Explore the relationship between age of a bank (based on “estymd”) and its efficiency level (“eeffr”).

[pic 5]

Age column has been added and derived from estymd in the “All_Reports_20080331_Performance and Condition Ratios” file. It is also used as a dimension. We also take the average efficiency to find a relationship and made use of the line chart.

It can be noted that the efficiency of the oldest bank upto a bank whose age is 10 has been more or less consistent. The newer banks have a much better efficiency whose ages is less than 10 years.

1. Create boxplots of the number of employees (“numemp”) per bank in each FDIC region (“fdicdbs”) (Note: you will need to link two tables in the data, search the readme file to find out where those variables are).

[pic 6]

I have used fdicdbs as a dimension (column) and changed numemp from measure to be used as a dimension (row). I used the box and whisker plot chart and changed the axis to logarithmic for a better understanding of the representation.