[posted only]
On Oct 19, 2005, at 3:48 PM, sheeri kritzer wrote:
One particular bit of SQL you may find helpful is this:
concat(ifnull(a_id,""),ifnull(a_text,""))
concat with anything and a null value will produce a null value. That
snippet of sql code will help you get one answer from the 2
[posted only]
On Oct 19, 2005, at 4:07 PM, Jon Frisby wrote:
Create an Excel spreadsheet. Import the raw data, structured as-is,
into a worksheet. Select all the relevant columns. Go to Data ->
"Pivot Table and Pivot Chart Report". Click "Finish". From the
"PivotTable Field List", drag the
eled
"Drop Row Fields Here", then drag question ID into the box labeled "Drop
Column Fields Here".
Voila.
-JF
> -Original Message-
> From: Jeffrey Goldberg [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, October 19, 2005 1:44 PM
> To: Brent Baisley
> C
You want a Pivot Table. Excel will do this nicely (assuming you have
65536 rows or less), but SQL does not provide a mechanism to do this.
If you want a web based interface you can look at Jtable. (I *think*
that's what it's called -- it's a Java web app that provides an HTML
pivot table interfac
I agree with Brent.
One particular bit of SQL you may find helpful is this:
concat(ifnull(a_id,""),ifnull(a_text,""))
concat with anything and a null value will produce a null value. That
snippet of sql code will help you get one answer from the 2 the
original database had. Unless there's ever
[mailed and posted]
On Oct 19, 2005, at 3:34 PM, Brent Baisley wrote:
The person you inherited from formatted the data correctly in my
opinion.
I agree.
What you are trying to do is store the data as you see it, which is
rarely a normalized data model. Your presentation layer should
han
The person you inherited from formatted the data correctly in my
opinion. With the existing format, you can index all the data with a
minimum number of indexes and quickly compile results. It can scale
to any number of questions without having to modify the underlying
data structure. It can
