REPORTING unique values

[jane]

here is my data:
Mary
Sally
Sally
Mary
John
John
David

I used the following to COUNT the unique values:
=SUMPRODUCT((B5:B15<>"")/COUNTIF(B5:B15,B5:B15))

What can I do to get a list that REPORTS the unique data - my result should
look like this:
Mary
Sally
John
David

thanks in advance! jane

 

[Pete_UK]

You need to have a heading (like Name), then you can highlight the
data plus header and then click on Data | Filter | Advanced Filter. In
the pop-up you should click on Unique Records Only as well as Copy to
another location - put a suitable cell reference in the box, such as
F1.

When you click OK then you will have your unique list in column F.

Hope this helps.

Pete

 

[Peo Sjoblom]

Use this instead

=SUMPRODUCT((B5:B15<>"")/COUNTIF(B5:B15,B5:B15&""))


that will not return an error if all cells are empty


You can use advanced filter which has a built in way, select the list, then
select unique records only then copy to another location. Excel 2007 has a
built in method for this called remove duplicates



--


Regards,


Peo Sjoblom

 

[Alan Beban]

Use this instead

=SUMPRODUCT((B5:B15<>"")/COUNTIF(B5:B15,B5:B15&""))


that will not return an error if all cells are empty


You can use advanced filter which has a built in way, select the list, then
select unique records only then copy to another location. Excel 2007 has a
built in method for this called remove duplicates

Can you post the syntax for using the remove duplicates method?

Thanks,
Alan Beban

 

[jane]

HI Peo and Pete,
Thank you for responding so quickly!

My result using Peo's formula was 6. My formula gave me 6 also but I needed
a list of the actual names without doing the auto-filter as Pete suggested.

thoughts? jane

 

[jane]

Hi Peo and Pete,
The result with Peo's formula was 6. My formula also returned 6 but I
really need a list with the actual names without using the auto-filter as
Pete suggested...

thoughts? jane

 

[Peo Sjoblom]

I corrected your formula that counts it, I never said it would return unique
values
If you test your formula and remove all values in your range it will return
a DIV error

To use a formula is rather complicated, here's a link that will do that

http://tinyurl.com/2fwou2


you can download a sample from the same page


http://nwexcelsolutions.com/Download/Sort_alphabetical_and_then_extract_unique_records.xls


it's definitely easier to use advanced filter


--


Regards,


Peo Sjoblom

 

[T. Valko]

Making a couple of assumptions.

If you're using this formula to count distinct entries:

=SUMPRODUCT((B5:B15<>"")/COUNTIF(B5:B15,B5:B15))

Then there are no empty cells within your range. Otherwise, that formula
would return an error.

I'm also assuming all the data is TEXT (like your posted sample).

So, assuming the above formula is entered in cell C5:

Enter this formula in D5:

=B5

Enter this array formula** in D6 and copy down until you get blanks:

=IF(ROWS(D$56)<=C$5,INDEX(B$5:B$15,MATCH(TRUE,COUNTIF(D$55,B$5:B$15)=0,0)),"")

** array formulas need to be entered using the key combination of
CTRL,SHIFT,ENTER (not just ENTER)

 

[Peo Sjoblom]

You can't read? I told her to use advanced filter


"You can use advanced filter which has a built in way, select the list, then
select unique records only then copy to another location. Excel 2007 has a
built in method for this called remove duplicates?"


I thought perhaps it would be easier to select a range and apply filter to
get the values but of course there are ways using a formula

=IF(ISNA(MATCH(0,COUNTIF($C$1:C1,$B$5:$B$15),0)),"",INDEX($B$5:$B$15,MATCH(0,COUNTIF($C$1:C1,$B$5:$B$15),0)))

entered with ctrl + shift & enter

Where $C$1:C1 is the cell above the cell with the formula, adapt accordingly


HTH



--


Regards,


Peo Sjoblom

 

[Teethless mama]

=IF(ISERR(SMALL(IF(MATCH(Name,Name,0)=ROW(INDIRECT("1:"&ROWS(Name))),MATCH(Name,Name,0)),ROWS($1:1))),"",INDEX(Name,SMALL(IF(MATCH(Name,Name,0)=ROW(INDIRECT("1:"&ROWS(Name))),MATCH(Name,Name,0)),ROWS($1:1))))

ctrl+shift+enter, not just enter
copy down as far as needed

 

[Alan Beban]

You can't read? I told her to use advanced filter


"You can use advanced filter which has a built in way, select the list, then
select unique records only then copy to another location. Excel 2007 has a
built in method for this called remove duplicates?"


I thought perhaps it would be easier to select a range and apply filter to
get the values but of course there are ways using a formula

=IF(ISNA(MATCH(0,COUNTIF($C$1:C1,$B$5:$B$15),0)),"",INDEX($B$5:$B$15,MATCH(0,COUNTIF($C$1:C1,$B$5:$B$15),0)))

entered with ctrl + shift & enter

Where $C$1:C1 is the cell above the cell with the formula, adapt accordingly


HTH

Oh, I can read. I'd hazard a guess that I've been getting paid for doing
it for longer than you've been putting meaningful sentences together.

The mistake I made was trying to make sense out of your statment "Excel
2007 has a built in method for this called remove duplicates." I now
take it that you don't really have in mind anything that's called
"remove duplicates" (except, perhaps, by you).

Thanks for your courtesy.

Alan Beban

 

[Peo Sjoblom]

Oh, I can read. I'd hazard a guess that I've been getting paid for doing

it for longer than you've been putting meaningful sentences together.

The mistake I made was trying to make sense out of your statment "Excel
2007 has a built in method for this called remove duplicates." I now take
it that you don't really have in mind anything that's called "remove
duplicates" (except, perhaps, by you).

Thanks for your courtesy.

So what's your contribution to this thread except being a troll?

 

[Alan Beban]

So what's your contribution to this thread except being a troll?

I thought from your comment that there was some new feature in Excel
2007, some new "built-in method" that would deal with an issue that has
been around for years and still comes up, and I was trying to get some
clarification of it. Silly me for taking your nonsense talk to be
meaningful.

Alan Beban

 

[Roger Govier]

Now boys, stop sqabbling!!!

Alan, Peo is quite correct when he says that XL2007 has a built in method
called Remove Duplicates. There is no formula as such. The dialogue is very
similar to the start of the Sort dialogue.

If the OP made a copy of here column of data, then selected Data tab>Data
Tools>Remove Duplicates
the dialogue asks if you want to continue with the selection or expand it.
You then get asked whether the column has headers, and after proceeding the
result is shown, which in the OP's example would say 2 duplicate values
found and removed, 5 unique values remain.

It is a useful new feature in XL2007 which does deal with what you say is an
issue that has been around for years.

 

[Bernd P]

Hello Jane,

I suggest to use my UDF lfreq (http://www.sulprobil.com/html/
listfreq.html) and to use its first output column only. Keep in mind
that it sorts the output, though...

Regards,
Bernd

 

[jane]

WOW everyone!
thank you so much for the help today. I corrected my original formula and
am working on applying the other. Will check back in if there are any more
questions.

Very much appreciate everyone's time and thoughts!
take care, jane

 

[Alan Beban]

Now boys, stop sqabbling!!!

Alan, Peo is quite correct when he says that XL2007 has a built in method
called Remove Duplicates. There is no formula as such. The dialogue is very
similar to the start of the Sort dialogue.

If the OP made a copy of here column of data, then selected Data tab>Data
Tools>Remove Duplicates
the dialogue asks if you want to continue with the selection or expand it.
You then get asked whether the column has headers, and after proceeding the
result is shown, which in the OP's example would say 2 duplicate values
found and removed, 5 unique values remain.

It is a useful new feature in XL2007 which does deal with what you say is an
issue that has been around for years.

Thanks, Roger. That's exactly the sort of clarification I was looking
for when I posted the seemingly, to me, innocuous, non-challenging,
civil question:

"Can you post the syntax for using the remove duplicates method?

Thanks,
Alan Beban"

and got back an off-the-wall snotty "You can't read?" response.

The reference to a "built-in method . . . called remove duplicates"
sounded very much to me (and still does, though I now recognize the
intended meaning) like a reference to a built-in REMOVEDUPLICATES
function, which I thought was an interesting development.

Thanks again for your clarification,

Alan Beban

 

[ryguy7272]

There are MANY ways to do this!

Count Uniques:
=SUMPRODUCT((A1:A1001<>"")/(COUNTIF(A1:A1001,A1:A1001&"")))
=SUMPRODUCT((A1:A1001<>"")/(COUNTIF(A1:A1001,A1:A1001)+(A1:A1001="")))
=SUMPRODUCT((A1:A1001<>"")/COUNTIF(A2:A1001,A2:A1001&"")*(A1:A105<>""))
=SUMPRODUCT(--(A1:A1001<>""),1/COUNTIF(A1:A1001,A1:A1001&"")
=SUM(IF(FREQUENCY(IF(LEN(A1:A1001)>0,MATCH(A1:A1001,A1:A1001,0),""),IF(LEN(A1:A1001)>0,MATCH(A1:A1001,A2:A1001,0),""))>0,1)
=SUM(--(FREQUENCY(IF(A1:A1001<>"",MATCH(A1:A1001,A1:A1001,0)),ROW(INDIRECT("1:"&ROWS(A1:A1001))))>0))
=SUM(IF(A1:A1001<>"",1/COUNTIF(A1:A1001,A1:A1001))
=SUM(IF(FREQUENCY(IF(A1:A1001<>"",MATCH("~"&A1:A1001,A1:A1001&"",0)),ROW(A1:A1001)-ROW(A1)+1),1))
=COUNT(1/FREQUENCY(A1:A1001,A1:A1001))

List Uniques:
Cell C1 will have the result of "Count Uniques" from above
=IF(ROWS(C$1:C1)<=$C$1,INDEX($A$1:$A$1000,SMALL(IF(FREQUENCY(IF($A$1:$A$1000<>"",MATCH("~"&$A$1:$A$1000,$A$1:$A$1000&"",0)),ROW($A$1:$A$1000)-ROW($A$1)+1),ROW($A$1:$A$1000)-ROW($A$1)+1),ROWS(C$1:C1))),"")
(this does the same thing as TM's Function).


Regards,
Ryan---

 

[Roger Govier]

You're very welcome, Alan

 

[Alan Beban]

There are MANY ways to do this!

Count Uniques:
=SUMPRODUCT((A1:A1001<>"")/(COUNTIF(A1:A1001,A1:A1001&"")))
=SUMPRODUCT((A1:A1001<>"")/(COUNTIF(A1:A1001,A1:A1001)+(A1:A1001="")))
=SUMPRODUCT((A1:A1001<>"")/COUNTIF(A2:A1001,A2:A1001&"")*(A1:A105<>""))
=SUMPRODUCT(--(A1:A1001<>""),1/COUNTIF(A1:A1001,A1:A1001&""))
=SUM(IF(FREQUENCY(IF(LEN(A1:A1001)>0,MATCH(A1:A1001,A1:A1001,0),""),IF(LEN(A1:A1001)>0,MATCH(A1:A1001,A2:A1001,0),""))>0,1))
=SUM(--(FREQUENCY(IF(A1:A1001<>"",MATCH(A1:A1001,A1:A1001,0)),ROW(INDIRECT("1:"&ROWS(A1:A1001))))>0))
=SUM(IF(A1:A1001<>"",1/COUNTIF(A1:A1001,A1:A1001)))
=SUM(IF(FREQUENCY(IF(A1:A1001<>"",MATCH("~"&A1:A1001,A1:A1001&"",0)),ROW(A1:A1001)-ROW(A1)+1),1))
=COUNT(1/FREQUENCY(A1:A1001,A1:A1001))

List Uniques:
Cell C1 will have the result of "Count Uniques" from above:
=IF(ROWS(C$1:C1)<=$C$1,INDEX($A$1:$A$1000,SMALL(IF(FREQUENCY(IF($A$1:$A$1000<>"",MATCH("~"&$A$1:$A$1000,$A$1:$A$1000&"",0)),ROW($A$1:$A$1000)-ROW($A$1)+1),ROW($A$1:$A$1000)-ROW($A$1)+1),ROWS(C$1:C1))),"")
(this does the same thing as TM's Function).


Regards,
Ryan---

And if the functions in the freely downloadable file at
http://home.pacbell.net/beban are available to your workbook

=ArrayUniques(A1:A1001), array entered, will list 'em vertically,

and

=ArrayUniques(A1:A1001,,"1horiz"), array entered, will list 'em
horizontally.

Or, if array entering is inconvenient, you can use

=INDEX(ArrayUniques($A$31:$A$1001),ROW(A1),1) filled down, or

=INDEX(ArrayUniques($A$31:$A$1001,,"1horiz"),1,COLUMN(A1)), filled across.

And

=ArrayCount(ArrayUniques(A1:A1001)) will count 'em.

Alan Beban