Home |
Search |
Today's Posts |
|
#1
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
I have a 7 question survery with the following possible responses
1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! |
#2
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
Hi,
Try it this way =SUM(A1:A7)/COUNT(A1:A7) Mike "Erinayn" wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! |
#3
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
I assumed you N/A is a text input and not an excel #N/A error. If it is the
Excel error then use this instead =SUMIF(A1:A7,"<#N/A")/COUNT(A1:A7) Mike "Erinayn" wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! |
#4
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
If you really have N/A (plain old text) in that cell, then
=average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson |
#5
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
It is just an alpha N/A. Now that I'm looking at it only one person has
responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . |
#6
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
As you stated, you did get a response which addressed the "alpha N/A". The
response was that Average will ignore text. Now you have to explain why that doesn't work for you, and what you mean by "the average is of nothing". Regards, Fred "Erinayn" wrote in message ... It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . |
#7
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
We poll about 8 people total per event and 1 received only 1 survey total for
that event. The person answered N/A to the first 3 questions and 5s for the rest. I am trying to get an average across the survey - which would be 5 since it's 20/4 and that worked. But we also get an average for each question across all of the responses. So in this instance since I got only one reply and that answer was N/A I am getting the #DIV! error as there is really nothing to divide with as the denominator is 0. HTH "Fred Smith" wrote: As you stated, you did get a response which addressed the "alpha N/A". The response was that Average will ignore text. Now you have to explain why that doesn't work for you, and what you mean by "the average is of nothing". Regards, Fred "Erinayn" wrote in message ... It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . . |
#8
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
Then you want something like:
=if(count(a1:a7)=0,"N/A",average(a1:a7)) Regards Fred "Erinayn" wrote in message ... We poll about 8 people total per event and 1 received only 1 survey total for that event. The person answered N/A to the first 3 questions and 5s for the rest. I am trying to get an average across the survey - which would be 5 since it's 20/4 and that worked. But we also get an average for each question across all of the responses. So in this instance since I got only one reply and that answer was N/A I am getting the #DIV! error as there is really nothing to divide with as the denominator is 0. HTH "Fred Smith" wrote: As you stated, you did get a response which addressed the "alpha N/A". The response was that Average will ignore text. Now you have to explain why that doesn't work for you, and what you mean by "the average is of nothing". Regards, Fred "Erinayn" wrote in message ... It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . . |
#9
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
=if(count(a1:a7)=0,"No numbers",average(a1:a7))
Check for at least one number in that range first. Erinayn wrote: It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . -- Dave Peterson |
#10
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
PERFECT!! That worked just how I needed it. I knew I needed an if plus the
concept from before but didn't know how to make it work. Thanks to everyone for their input. "Dave Peterson" wrote: =if(count(a1:a7)=0,"No numbers",average(a1:a7)) Check for at least one number in that range first. Erinayn wrote: It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . -- Dave Peterson . |
#11
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
=IF(COUNT(A1:A6),AVERAGE(A1:A6),"")
-- David Biddulph "Erinayn" wrote in message ... It is just an alpha N/A. Now that I'm looking at it only one person has responded as such and the kicker is that no other responses for the area were received so I'm getting an error since the average is of nothing. Any thoughts on that? "Dave Peterson" wrote: If you really have N/A (plain old text) in that cell, then =average(a1:a6) should work fine. =average() ignores non-numeric (but not errors) values. On the other hand, if you have the error #N/A (resulting from a formula???) in that range, then =average() will return an error. You could use: =SUMIF(A1:A6,"<1e37")/COUNT(A1:A6) 1E37 in scientific notation for a giant number. 1 followed by 37 zeros. Erinayn wrote: I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! -- Dave Peterson . |
#12
![]()
Posted to microsoft.public.excel.worksheet.functions
|
|||
|
|||
![]()
If N/A has been entered as text (not from a formula), the =AVERAGE(A1:A7)
will ignore it If you have NA() as part of a formula that is returning N/A: Option 1 =AVERAGE(IF(ISNUMBER(C1:C7),C1:C7)) Array formula: you must commit it with Shift+Ctrl+Enter not just Enter Option 2 Non array formula =SUMIF(D1:D7,"0")/COUNTIF(D1:D7,"0") best wishes -- Bernard Liengme Microsoft Excel MVP http://people.stfx.ca/bliengme "Erinayn" wrote in message ... I have a 7 question survery with the following possible responses 1 2 3 4 5 N/A We then take ther responses and average them. I'd like to eliminate the N/A responses from the denominator and from the total sum. I currently have 1 cell for this operation - which doesn't work when the response is N/A since it's a simple (sum)/7 equation. Thanks for any help! |
Reply |
Thread Tools | Search this Thread |
Display Modes | |
|
|
![]() |
||||
Thread | Forum | |||
Summing all numeric values only in cells in a column | Excel Worksheet Functions | |||
averages for cells | Excel Worksheet Functions | |||
Summing first 2 characters of a cell (some non-numeric) | Excel Discussion (Misc queries) | |||
Replacing characters in numeric text strings and SUMming | Excel Discussion (Misc queries) | |||
Summing part of an Alpha Numeric String | Excel Worksheet Functions |