Comments on: Creating a Frequency Distribution with a Pivot Table

By: Daryl

Daryl — Fri, 09 Jul 2010 00:34:21 +0000

A year later and still nice…
Thanks for posting this

By: tani

tani — Sat, 13 Mar 2010 02:50:15 +0000

wonderful tip…just does what I want
thanks very much

By: Gusto

Gusto — Fri, 11 Dec 2009 01:55:10 +0000

Excellent tip. Too bad I can't set the interval manually with time data either.

By: Mark Truss

Mark Truss — Sun, 04 Oct 2009 02:18:01 +0000

Very helpful – exactly what I needed – thank you.

By: Colin Banfield

Colin Banfield — Sat, 04 Jul 2009 16:15:36 +0000

I've often wished for an option to show "empty" groups (in your example, if there was no data for the 200-300 group, the next row after 100-200 would be 300-400). Such an option would be useful for some types of frequency distributions e.g graphical stem and leaf plots.

By: Mathias

Mathias — Wed, 01 Jul 2009 00:27:04 +0000

Very nice trick – came at the right time, to, as I needed to generate the distribution of a few sets of data, thank you!
And I got it to work with Excel 2007 with no issue, on a field which was calculated, and formatted as %. Jared, General Ledger, what are your data?

By: General Ledger

General Ledger — Tue, 30 Jun 2009 17:44:05 +0000

Jared,

I got the exact same problem as you when using 2007 but fine with 2003. While in 2007, it seems the row data are left justified and being treated as text even though the raw data is formatted as Accounting. With the cursor on a row data and right click. The Field Setting option does not let you change the format of the numbers to Accounting, etc.

However, the order of the row data suggests Excel knows they are numbers (1000 is after 300).

Help!?!?

By: Ethan

Ethan — Tue, 30 Jun 2009 17:18:27 +0000

First off, great tip. Thanks!
Second, unless there's a convention I missed back in my Stats classes, don't you’re bins need to be discrete and not overlap?

This question lead me to a curious behavior in Excel. When I tried your trick I started with a simple dataset of whole numbers from 1 to 20 and created the pivot table with a group interval of 5 which gave me four discrete bins, i.e. 1-5, 6-10, etc. All is good.

Next, I changed the first value in my dataset to 0.1 which changed the bins to 0.1-5.1, 5.1-10.1, 10.1-15.1, etc. Now my bins overlap…why and why now?

Then, with bin #1 = 0.1-5.1 and bin #2 = 5.1-10.1 I changed the fifth value in the dataset from 5 to 5.1 tentatively expecting it to be included in bin #1, however Excel included it in bin #2. Is this expected and by specific design?

Thanks

By: Jared

Jared — Tue, 30 Jun 2009 14:15:12 +0000

No dates, no calculated items….no luck. Thanks anyway, I just suffer through the 25-step process.

By: DataPig

DataPig — Tue, 30 Jun 2009 13:59:56 +0000

Sounds like you're trying to group dates that are not formatted as dates. Or you have a calculated item in your pivot table already. I don't think you can group in a pivot table with calculated items.