Counting Unique Values with Functions

Subscriber David Keys has a worksheet in which there is a list of countries. This list, in cells A1:A100, can contain duplicates. David is seeking a way to determine the number of unique countries in the list.

There are several ways you can go about deriving a count, without resorting to using a macro. The method you should use depends on the characteristics of the data in the list. A good place to start, however, is to define a named range that represents the list of countries. In the following examples, it is assumed that the range is named Countries. (Catchy name, huh?)

If the list contains only text entries, and does not contain any blank cells, then the following will provide a count:

=SUM(1/COUNTIF(Countries,Countries))

This should be entered as an array formula, by pressing Ctrl+Shift+Enter. If the list contains blank cells, then the formula becomes a bit more complex. The following long array formula will work if there are blanks:

=SUM(IF(FREQUENCY(IF(LEN(Countries)>0,MATCH(Countries,Countries,0),""), IF(LEN(Countries)>0,MATCH(Countries,Countries,0),""))>0,1))

This approach--using the FREQUENCY function--is fully recounted in the Microsoft Knowledge Base:

http://support.microsoft.com/support/kb/articles/Q90/4/00.asp

If you prefer to not use array formulas (for whatever reason), you can utilize a blank column to the right of your list. This column will contain regular formulas that indicate if the value to its left is unique in the list or not. The first time a value appears, the formula returns the number 1. On each subsequent appearance of the same value, the formula returns a 0. Start by sorting your list, then place the following formula in cell B1:

=IF(ISNA(VLOOKUP(A2,$A$1:A1,1,)),1,0)

Just copy the formula from B1 to the range B2:B100. With these results in place, you can easily sum column B and have a count of the unique values in the list.