问题描述:

Example data set

Right, above is a link to an image of a sub-segment of my data set. It is oriented in sets of 3 columns, with the first being a concentration, the second a qualifier, and the last an MDL - and continues for up to 95 samples (so a total of 285 columns making manual entry impractical) . How can i calculate the max or min of the concentration values for those that have a qualifier of "u" or vice versa have no qualifier?

I can't figure out anything, and unfortunately i don't have the time to re-orient the data. Anybody have an idea?

Perhaps something like this will do,

The 8 formulas in C7:J7 are,

```
=AGGREGATE(15, 6, $A2:$AY2/(($A$1:$AY$1=C$6)*($B2:$Z2="U")), 1)
=AGGREGATE(15, 6, $C2:$BA2/(($C$1:$BA$1=D$6)*($B2:$AZ2="U")), 1)
=AGGREGATE(14, 6, $A2:$AY2/(($A$1:$AY$1=E$6)*($B2:$Z2="U")), 1)
=AGGREGATE(14, 6, $C2:$BA2/(($C$1:$BA$1=F$6)*($B2:$AZ2="U")), 1)
=AGGREGATE(15, 6, $A2:$AY2/(($A$1:$AY$1=G$6)*($B2:$Z2<>"U")), 1)
=AGGREGATE(15, 6, $C2:$BA2/(($C$1:$BA$1=H$6)*($B2:$AZ2<>"U")), 1)
=AGGREGATE(14, 6, $A2:$AY2/(($A$1:$AY$1=I$6)*($B2:$Z2<>"U")), 1)
=AGGREGATE(14, 6, $C2:$BA2/(($C$1:$BA$1=J$6)*($B2:$AZ2<>"U")), 1)
```

Those cover both minimum and maximum values when either including or excluding the qualifier.

**Addendum: Excluding blank cells**

One more condition to check the LEN of the values can be added. To change the length of the value into a divide-by-1 (unchanged) or divide-by-0 (#DIV/)! error) wrap the LEN in the SIGN function.

```
=AGGREGATE(15, 6, $A2:$AY2/(SIGN(LEN($A2:$AY2))*($A$1:$AY$1=C$6)*($B2:$Z2="U")), 1)
```

I'm retaining the SMALL sub-function as only AGGREGATE's sub-functions 14 and up process as an array.