4 Replies Latest reply on Mar 6, 2012 6:05 AM by Shawn Wallwork

    Connecting points on a line chart using discrete and aggregate data

    Jonathan Drummey

      Data Description

       

      I've got two initial data sources, both from MS Access:

      - Blood transfusion event data, with one record per transfusion.

      - Inpatient admissions per month data, with one row per month containing a count of admissions.

       

      The Tableau data source for blood transfusions has the admissions data for that month in the Admission Count field. I've also included the admissions count data as a separate data source that can be blended on the Month Begin field. (So there are two ways to get at the # of admissions, either in the blood transfusions data or via a blend).

       

      Goal

       

      My goal is to have a line chart like the Goal chart below showing the # of tranfusions, admission count, and rate of Transfusions per Admission over time, at the month, quarter, or annual level of aggregation. I created a set of non-table calcs that work at the month level, those are used to draw the Goal chart below. I also created a set of table calcs that work at the month, quarter, and annual levels, you can see them in the BT crosstab view.

       

      goal.jpg

       

      Problem

       

      However, when going to the quarter or annual aggregates (using the table calcs), I get points and not a line chart, as shown in in the BT Points w/Out Line view below. I've tried different values for Date on the Path shelf based on this forum discussion http://community.tableau.com/thread/115881 and nothing has worked.

       

      bt points without line.jpg

       

      Questions

       

      Based on this, I have three questions:

      1. How would I connect the lines in the BT Points w/Out Line chart?

      2. Is there a simpler way to do this?

      3. Is it better to build the count of Admissions in the query, or use a blend instead? I can see advantages and disadvantages to both of them

       

      Thanks,

       

      Jonathan