Just a reminder to let everyone know, we spend a good deal of time looking over the forums and will answer questions if we can when our real, full-time paying job allows. Remember that the majority of us are volunteers here. Tableau employees are easy to spot because they soak their pencils in coffee so they are more flavorful when they chew on them and they also have a Tableau icon to the right of their name:
Even then, many of the Tableau employees come here outside their normal job to help answer questions which is not only awesome but extremely helpful, too!
Some of us are specialists and typically answer in our area of expertise, for example, I deal with Tableau Server while Pooja and Shawn deal with Desktop so you rarely see us inside the others area of expertise. So a mass-pinging (a.k.a. shotgun pinging) isn't necessary because we look over the various forums and answer what we can, volunteers and employees.
Finding the correct @ Mention person
If you hove your mouse over a person's name you'll see their profile:
Their profile address (1) is in the lower left region of the screen and is what you use for the @ Mention text. Their Title (2) and their avatar (3) are additional helpers in finding the right person. You use those to help get the correct @ Mention link:
This way you will get the right person so you won't run into the multiple Simon Runc issue (there are a lot of those little buggers ).
Hi Toby, I am an statistican student so I am learning Tableau with no Tableau support help. I just want to share my case and the solution I found. I think my experience can be helpful for others or maybe others can tell a better idea.
I have used @ top rank user just because I thought some of you would have an advice. Please feel free to not aswer if you could not time to do it.
By the way, thanks for the @ tips. I will take it in account next time
1 of 1 people found this helpful
I've noticed that there are more statisticians using Tableau so it's good to get your feedback on different ways to attack a problem. A good area for such discussions is in the Viz Talk forum. Also, you can post solutions like what Steve Martin did with his Cumulative Distribution Function (Normal Distribution) but put yours in the subspace Tableau Workbook Library .
To help others I would recommend adding Tags to your post as this will help find the content better during searches:
Thanks agaig Toby! I will take all your tips in account.
Nice to hear lots of statisticans are using Tableau. Tableau have been improving R Integration recently, and that become Tableau more atractive for us. I will try to integrate some of my R works with Tableau and share with the community,
So you're starting with quantities in one source and prices in another source, what is your end goal exactly?
Hi Jon, the goal is to take the prices from the second source to make a calculated field quantityXprice. There are different approaches to reach de objetive. The problem is the way. For example I took prices with an union but then is difficult to use pXQ as a calculated field.
The best solution I found is to take the prices, export table and create a new data source with that information, but maybe this is not the most automatically procedure. I have tried pivoting data, but it create a unnecesary large data source.
Attached "Tableau Query 03.0216" is the final goal with my solution (get a calculated filed p1Xq1, p2Xq2, ...,etc)
1 of 1 people found this helpful
Here's a solution entirely in Tableau that does not require any changes to the source data:
Here's how I built it:
1) Connected to the Quantity source.
2) Pivoted that source on the Q fields.
3) Renamed the fields.
4) Joined on Col to the Price source. This creates a lot of extra un-needed records because the join isn't at sufficient grain...we'll get rid of them in the next two steps.
5) Created a new calculated field "Data Source Filter" with the formula 'Q' + MID([Price ID],2,LEN([Price ID])-1) = [Q]
6) Added the above field as a Data Source filter, this gets rid of the records from #4.
7) Built a QxP measure that is [Value]*[Price]
8) Built the view above.
If you could add a column to the Price Source in Excel that was a formula like ="Q" & MID(A2,2,LEN(A2)-1) then you could add that field to the join in step 4 above and be able to skip steps 5 and 6 because the join would be at the correct level of Col & Q/P.
I prefer this style of data source rather than a blended solution because having the Q1, Q2, etc. as members of a dimension gives more flexibility in filtering, sorting, and interactivity. The extra rows created by this are usually not a problem except when record sets start getting into the tens of millions of records or more.