thanks - will try

"Roger Govier" wrote in message
...
Hi

With your data on Sheet1, in a range from say A1:F100, with A1:F1 holding
the headers, set up a table on Sheet2 with unique values from Sheet1
column
A in cells A2:An, unique values from Column B of Sheet1 in cells B1:?1
(n represents the last row number, ? represents the last column letter)

then in cell B2 of sheet2 enter

=SUMPRODUCT(--(Sheet1!$A$2:$A$100=$A2),--(Sheet1$B$2:$B$100=B$1),--(Sheet1!C
$2:C$100<""))
Copy across for the width of the column headings created.
Copy the whole row of formulae down for the extent of the entries in
column
A of Sheet2

You would then need to repeat the block from column B to column ?, 3 more
times to represent each of the remaining 4 columns of data from Sheet1,
but
changing the last part of the sumproduct formula to Sheet1!$D$2:$D$100,
E2:E100, and F2:F100 respectively.

If you wanted to make the table smaller and aggregate the count for all 4
columns of data, then change the formula to

=SUMPRODUCT(--(Sheet1!$A$2:$A$100=$A2),--(Sheet1$B$2:$B$100=B$1),--(Sheet1!C
$2:F$100<""))

Regards

Roger Govier


teepee wrote:
Hello - may I prevail on your collective knowledge please to solve a
problem?

I currently have a pivot table which takes 6 data entry columns, tables
the
first entry vertical, the second horizontal, and puts the last 4 in the
main
body of the table, counting their incidence in relation to the first 2.
A
pretty bog standard pivot table in fact.

My problem is this. In order to use a certain procedure, I have to
rewrite
this process to achieve the same end without using a pivot to get there.
The
authors of the procedure assure me it can be done using other data
filtering
functions within Excel but they don't say how. Anyone got any idea?

Thank in advance

teepee