|
|
Not a nice thing to say to a coworker, but was he right. I had a discussion with a male collegue about the difference between and empty cell and a cell with a zero. He called me a young, inexperienced, idiot woman because I insist his workbook is wrong since he did not account for empty cells that are awaiting data. He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and C16 and C17 dont have data yet - he insists his result is the G-d given answer. I suggest that =Average (C14:C17) is correct and that he is getting the wrong answer. It seems that his favorite feature, Conditional Formatting does not distinguish between Null and 0. Please settle the argument - I should complain to HR but will get far better satisfaction if I can site some reference and call him a #$&^ IDIOT! Thanks
|
|
I assume that you are a woman. Your idiocy can only be verified with an IQ test.
Using =AVERAGE() is better because it ignores missing data.
(b.t.w. - hiding a 0 with conditional formatting does not make it blank, it is still a 0) -- Gary''s Student - gsnu200823
"Pepper" wrote:
[Quoted Text] > Not a nice thing to say to a coworker, but was he right. > I had a discussion with a male collegue about the difference between and > empty cell and a cell with a zero. He called me a young, inexperienced, > idiot woman because I insist his workbook is wrong since he did not account > for empty cells that are awaiting data. > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > C16 and C17 dont have data yet - he insists his result is the G-d given > answer. > I suggest that =Average (C14:C17) is correct and that he is getting the > wrong answer. > It seems that his favorite feature, Conditional Formatting does not > distinguish between Null and 0. > Please settle the argument - I should complain to HR but will get far better > satisfaction if I can site some reference and call him a #$&^ IDIOT! > Thanks
|
|
You are both right, unfortunately. There is nothing wrong with EITHER
formula depending on your desired result.
His formula stipulates the zero is an acceptable answer and should
impact the average. This is true, for instance, if you were averaging
test scores, and you actually got zeros on some tests.
But your formula stipulates an average of existing data entries only,
which is also a correct approach.
You don't need to ask your HR which of you is correct, since that also
doesn't build team morale, you need to ask them their policy on
unanswered data points. Do they want unentered data to still count as
zero and be averaged, or do they want the average to only count against
entered data.
Even if they pick your solution, which I would, that still doesn't make
him "wrong", so tread gently.
--
JBeaucaire
------------------------------------------------------------------------
JBeaucaire's Profile: http://www.thecodecage.com/forumz/member.php?userid=73
View this thread: http://www.thecodecage.com/forumz/showthread.php?t=45512
|
|
If he wants to divide by 4 regardless then =(C14+C15+C16+C17)/4 is the same as =sum(c14:c17)/4 is the same as =average(c14:c17) if there is a 0 in one cell. But if one blank it will divide by 3 instead, etc
-- Don Guillett Microsoft MVP Excel SalesAid Software dguillett1[ at ]austin.rr.com "Pepper" <Pepper[ at ]discussions.microsoft.com> wrote in message news:F3D62D5C-B965-45C9-8629-295320A79A69[ at ]microsoft.com...
[Quoted Text] > Not a nice thing to say to a coworker, but was he right. > I had a discussion with a male collegue about the difference between and > empty cell and a cell with a zero. He called me a young, inexperienced, > idiot woman because I insist his workbook is wrong since he did not > account > for empty cells that are awaiting data. > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > C16 and C17 dont have data yet - he insists his result is the G-d given > answer. > I suggest that =Average (C14:C17) is correct and that he is getting the > wrong answer. > It seems that his favorite feature, Conditional Formatting does not > distinguish between Null and 0. > Please settle the argument - I should complain to HR but will get far > better > satisfaction if I can site some reference and call him a #$&^ IDIOT! > Thanks
|
|
My point is, his formula always divides by 4. If the range indicates four fiscal quarters, and the last quarter has not yet been entered, then averaging by his formula divides the sum of three quarters by 4 - and that is not the average for the first three quarters, it should be divided by 3 (like the Average function does)
"Don Guillett" wrote:
[Quoted Text] > If he wants to divide by 4 regardless then > =(C14+C15+C16+C17)/4 > is the same as > =sum(c14:c17)/4 > is the same as > =average(c14:c17) if there is a 0 in one cell. But if one blank it will > divide by 3 instead, etc > > > -- > Don Guillett > Microsoft MVP Excel > SalesAid Software > dguillett1[ at ]austin.rr.com > "Pepper" <Pepper[ at ]discussions.microsoft.com> wrote in message > news:F3D62D5C-B965-45C9-8629-295320A79A69[ at ]microsoft.com... > > Not a nice thing to say to a coworker, but was he right. > > I had a discussion with a male collegue about the difference between and > > empty cell and a cell with a zero. He called me a young, inexperienced, > > idiot woman because I insist his workbook is wrong since he did not > > account > > for empty cells that are awaiting data. > > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > > C16 and C17 dont have data yet - he insists his result is the G-d given > > answer. > > I suggest that =Average (C14:C17) is correct and that he is getting the > > wrong answer. > > It seems that his favorite feature, Conditional Formatting does not > > distinguish between Null and 0. > > Please settle the argument - I should complain to HR but will get far > > better > > satisfaction if I can site some reference and call him a #$&^ IDIOT! > > Thanks > >
|
|
|
[Quoted Text] >He called me a young, inexperienced, idiot woman
I think you have grounds for a sexual harassment lawsuit!
-- Biff Microsoft Excel MVP
"Pepper" <Pepper[ at ]discussions.microsoft.com> wrote in message news:F3D62D5C-B965-45C9-8629-295320A79A69[ at ]microsoft.com... > Not a nice thing to say to a coworker, but was he right. > I had a discussion with a male collegue about the difference between and > empty cell and a cell with a zero. He called me a young, inexperienced, > idiot woman because I insist his workbook is wrong since he did not > account > for empty cells that are awaiting data. > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > C16 and C17 dont have data yet - he insists his result is the G-d given > answer. > I suggest that =Average (C14:C17) is correct and that he is getting the > wrong answer. > It seems that his favorite feature, Conditional Formatting does not > distinguish between Null and 0. > Please settle the argument - I should complain to HR but will get far > better > satisfaction if I can site some reference and call him a #$&^ IDIOT! > Thanks
|
|
I already indicated I would use your approach since YTD functionality in
your formulas is GOOD thing. His formula is useless until the last day
of the last quarter. Maybe he LIKES knowing he can't be held accountable
for anything until the end of the year, who knows?
The question you're asking is for your superiors, though. They need to
answer the question: -"Should 'averaged' data present accurately all
throughout the year, or always average against 4 quarters...even during
Q1, Q2 and Q3."-
Whatever THEY answer is what you and your colleague must follow. I'll
bet you dollars to donuts they tell you that your cumulative average is
the correct one.
--
JBeaucaire
------------------------------------------------------------------------
JBeaucaire's Profile: http://www.thecodecage.com/forumz/member.php?userid=73
View this thread: http://www.thecodecage.com/forumz/showthread.php?t=45512
|
|
So.... Ask him which he wants.
-- Don Guillett Microsoft MVP Excel SalesAid Software dguillett1[ at ]austin.rr.com "Pepper" <Pepper[ at ]discussions.microsoft.com> wrote in message news:104AD18C-D47A-4ED4-A0BA-3512896381CC[ at ]microsoft.com...
[Quoted Text] > My point is, his formula always divides by 4. If the range indicates four > fiscal quarters, and the last quarter has not yet been entered, then > averaging by his formula divides the sum of three quarters by 4 - and that > is > not the average for the first three quarters, it should be divided by 3 > (like > the Average function does) > > "Don Guillett" wrote: > >> If he wants to divide by 4 regardless then >> =(C14+C15+C16+C17)/4 >> is the same as >> =sum(c14:c17)/4 >> is the same as >> =average(c14:c17) if there is a 0 in one cell. But if one blank it will >> divide by 3 instead, etc >> >> >> -- >> Don Guillett >> Microsoft MVP Excel >> SalesAid Software >> dguillett1[ at ]austin.rr.com >> "Pepper" <Pepper[ at ]discussions.microsoft.com> wrote in message >> news:F3D62D5C-B965-45C9-8629-295320A79A69[ at ]microsoft.com... >> > Not a nice thing to say to a coworker, but was he right. >> > I had a discussion with a male collegue about the difference between >> > and >> > empty cell and a cell with a zero. He called me a young, >> > inexperienced, >> > idiot woman because I insist his workbook is wrong since he did not >> > account >> > for empty cells that are awaiting data. >> > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, >> > and >> > C16 and C17 dont have data yet - he insists his result is the G-d given >> > answer. >> > I suggest that =Average (C14:C17) is correct and that he is getting the >> > wrong answer. >> > It seems that his favorite feature, Conditional Formatting does not >> > distinguish between Null and 0. >> > Please settle the argument - I should complain to HR but will get far >> > better >> > satisfaction if I can site some reference and call him a #$&^ IDIOT! >> > Thanks >> >>
|
|
Firstly, Biff I do have grounds, but am not that thin skinned. He is just crotchedy older "gentleman".
Secondly, the sum is supposed to be annual-to-date (in my bosses words) therefore the Average should be divided by the number of quarters that data exists.
If we have 100, 200 and 250 for the first three quarters, then the total to date is 550 and the average is 183.333.
Thank you all for your feedback.
"JBeaucaire" wrote:
[Quoted Text] > > I already indicated I would use your approach since YTD functionality in > your formulas is GOOD thing. His formula is useless until the last day > of the last quarter. Maybe he LIKES knowing he can't be held accountable > for anything until the end of the year, who knows? > > The question you're asking is for your superiors, though. They need to > answer the question: -"Should 'averaged' data present accurately all > throughout the year, or always average against 4 quarters...even during > Q1, Q2 and Q3."- > > Whatever THEY answer is what you and your colleague must follow. I'll > bet you dollars to donuts they tell you that your cumulative average is > the correct one. > > > -- > JBeaucaire > ------------------------------------------------------------------------ > JBeaucaire's Profile: http://www.thecodecage.com/forumz/member.php?userid=73> View this thread: http://www.thecodecage.com/forumz/showthread.php?t=45512> >
|
|
Very good response.
The epitome of tact and diplomacy.
Gord Dibben MS Excel MVP
On Wed, 31 Dec 2008 17:04:45 +0000, JBeaucaire <JBeaucaire.3laltu[ at ]thecodecage.com> wrote:
[Quoted Text] > >You are both right, unfortunately. There is nothing wrong with EITHER >formula depending on your desired result. > >His formula stipulates the zero is an acceptable answer and should >impact the average. This is true, for instance, if you were averaging >test scores, and you actually got zeros on some tests. > >But your formula stipulates an average of existing data entries only, >which is also a correct approach. > >You don't need to ask your HR which of you is correct, since that also >doesn't build team morale, you need to ask them their policy on >unanswered data points. Do they want unentered data to still count as >zero and be averaged, or do they want the average to only count against >entered data. > >Even if they pick your solution, which I would, that still doesn't make >him "wrong", so tread gently.
|
|
First, you should ask him to differentiate between an "Idiot Woman" and an "Idiot Man" - perhaps there is some basis in gender for a difference between the two, although as far as I know, generally accepted IQ tests make no such distinction.
Now, as to how the formula should work - I think Don Guillett's first post in the thread covers it nicely - either you count zeros/missing entries as zero with a division by 4 always, or you take the sum of non-zero entries and divide by the number of non-zero entries to obtain a true average for data that you actually have in hand.
As for the crotchedy older gentleman, please tell him for me that I sincerely hope he is happily married at the moment, because with his attitude, if he ain't already, he probably is never going to be :). P.S. I'm grumpy, I'm older, and hopefully most would consider me a gentleman; and I'd never make such a statement to a member of the alternate gender [assuming female is the alternative to male, without regard for other potential alternatives :)]. He would appear to be a moron, studying to become an idiot, and failing miserably!
"Pepper" wrote:
[Quoted Text] > Firstly, Biff I do have grounds, but am not that thin skinned. He is just > crotchedy older "gentleman". > > Secondly, the sum is supposed to be annual-to-date (in my bosses words) > therefore the Average should be divided by the number of quarters that data > exists. > > If we have 100, 200 and 250 for the first three quarters, then the total to > date is 550 and the average is 183.333. > > Thank you all for your feedback. > > "JBeaucaire" wrote: > > > > > I already indicated I would use your approach since YTD functionality in > > your formulas is GOOD thing. His formula is useless until the last day > > of the last quarter. Maybe he LIKES knowing he can't be held accountable > > for anything until the end of the year, who knows? > > > > The question you're asking is for your superiors, though. They need to > > answer the question: -"Should 'averaged' data present accurately all > > throughout the year, or always average against 4 quarters...even during > > Q1, Q2 and Q3."- > > > > Whatever THEY answer is what you and your colleague must follow. I'll > > bet you dollars to donuts they tell you that your cumulative average is > > the correct one. > > > > > > -- > > JBeaucaire > > ------------------------------------------------------------------------ > > JBeaucaire's Profile: http://www.thecodecage.com/forumz/member.php?userid=73> > View this thread: http://www.thecodecage.com/forumz/showthread.php?t=45512> > > >
|
|
Thanks for the kind words. I like the part about the crotchety older gentleman which fits this grumpy 72 yr old.
-- Don Guillett Microsoft MVP Excel SalesAid Software dguillett1[ at ]austin.rr.com "JLatham" <HelpFrom [ at ] Jlathamsite.com.(removethis)> wrote in message news:BC22CF20-9C1E-4BF6-AFD3-F39495B8920F[ at ]microsoft.com...
[Quoted Text] > First, you should ask him to differentiate between an "Idiot Woman" and an > "Idiot Man" - perhaps there is some basis in gender for a difference > between > the two, although as far as I know, generally accepted IQ tests make no > such > distinction. > > Now, as to how the formula should work - I think Don Guillett's first post > in the thread covers it nicely - either you count zeros/missing entries as > zero with a division by 4 always, or you take the sum of non-zero entries > and > divide by the number of non-zero entries to obtain a true average for data > that you actually have in hand. > > As for the crotchedy older gentleman, please tell him for me that I > sincerely hope he is happily married at the moment, because with his > attitude, if he ain't already, he probably is never going to be :). P.S. > I'm > grumpy, I'm older, and hopefully most would consider me a gentleman; and > I'd > never make such a statement to a member of the alternate gender [assuming > female is the alternative to male, without regard for other potential > alternatives :)]. He would appear to be a moron, studying to become an > idiot, and failing miserably! > > "Pepper" wrote: > >> Firstly, Biff I do have grounds, but am not that thin skinned. He is >> just >> crotchedy older "gentleman". >> >> Secondly, the sum is supposed to be annual-to-date (in my bosses words) >> therefore the Average should be divided by the number of quarters that >> data >> exists. >> >> If we have 100, 200 and 250 for the first three quarters, then the total >> to >> date is 550 and the average is 183.333. >> >> Thank you all for your feedback. >> >> "JBeaucaire" wrote: >> >> > >> > I already indicated I would use your approach since YTD functionality >> > in >> > your formulas is GOOD thing. His formula is useless until the last day >> > of the last quarter. Maybe he LIKES knowing he can't be held >> > accountable >> > for anything until the end of the year, who knows? >> > >> > The question you're asking is for your superiors, though. They need to >> > answer the question: -"Should 'averaged' data present accurately all >> > throughout the year, or always average against 4 quarters...even during >> > Q1, Q2 and Q3."- >> > >> > Whatever THEY answer is what you and your colleague must follow. I'll >> > bet you dollars to donuts they tell you that your cumulative average is >> > the correct one. >> > >> > >> > -- >> > JBeaucaire >> > ------------------------------------------------------------------------ >> > JBeaucaire's Profile: >> > http://www.thecodecage.com/forumz/member.php?userid=73>> > View this thread: >> > http://www.thecodecage.com/forumz/showthread.php?t=45512>> > >> >
|
|
I could not resist adding to the discussion...
Well, he probably got two out of three right... you must be young (since you did not dispute this) and inexperienced (arguing with old men and not sure about your stand) :-)
Also if not being an expert on some aspect of Excel makes one an idiot then I would guess more than 95% of the world's population would qualify (including me.)
If I were in your place I would have given him three numbers and asked him to calculate the average on paper. I would have walked away if he divided by four . If he divided by three then I would have asked him to enter the three numbers on his spreadsheet and compare the result with his manual calucation.
No one should call any one else an idiot under any circumstances.
Now my two cents; You were right and he was wrong.
HAPPY NEW YEAR.
PS: Can you guess my gender from my response? My age?
This reminds me of an incident when an young woman told an old man that he did not have to open the door just because she was a woman. His response - "I am doing it because I am a gentleman"
"Pepper" wrote:
[Quoted Text] > Not a nice thing to say to a coworker, but was he right. > I had a discussion with a male collegue about the difference between and > empty cell and a cell with a zero. He called me a young, inexperienced, > idiot woman because I insist his workbook is wrong since he did not account > for empty cells that are awaiting data. > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > C16 and C17 dont have data yet - he insists his result is the G-d given > answer. > I suggest that =Average (C14:C17) is correct and that he is getting the > wrong answer. > It seems that his favorite feature, Conditional Formatting does not > distinguish between Null and 0. > Please settle the argument - I should complain to HR but will get far better > satisfaction if I can site some reference and call him a #$&^ IDIOT! > Thanks
|
|
BTW: I don't think anyone directly addressed the issue of a difference between a cell with a zero in it and a cell than actually has not had any entry (or formula) entered into it: YES, Virginia, er Pepper, there IS a difference, and Excel treats empty cells differently at times than cells with either a zero length string ( "" ) or a zero in them.
Here's one simple proof: In cell A1 put "abc" (as abc) skip over A2, never putting anything in it in A3 put the formula = "" in A4 type the value 42 now, in A5 put this formula: =CountA(A1:A4) and the result should be 3 even though only cells A1 and A4 have a visible value in them, the zero length string in A3 is counted, but the EMPTY cell at A2 is not.
For that matter, put a 0 into A1, A3 and A4 and formula = COUNT(A1:A4) in A5 and again the result is 3, showing that an EMPTY cell is not treated as a zero for this purpose.
There is a similar difference when programming in VBA, where we are given specific tests for an empty cell vs one with a zero length string or a zero value, as IF IsEmpty(Range("A2")) Then MsgBox"Cell A2 is actually EMPTY" and if Range("A2")=Empty then MsgBox"Cell A2 is still actually EMPTY, not zero!"
"Pepper" wrote:
[Quoted Text] > Not a nice thing to say to a coworker, but was he right. > I had a discussion with a male collegue about the difference between and > empty cell and a cell with a zero. He called me a young, inexperienced, > idiot woman because I insist his workbook is wrong since he did not account > for empty cells that are awaiting data. > He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > C16 and C17 dont have data yet - he insists his result is the G-d given > answer. > I suggest that =Average (C14:C17) is correct and that he is getting the > wrong answer. > It seems that his favorite feature, Conditional Formatting does not > distinguish between Null and 0. > Please settle the argument - I should complain to HR but will get far better > satisfaction if I can site some reference and call him a #$&^ IDIOT! > Thanks
|
|
Not so for AVERAGE.
However, I fully subscribe to that and take it a step further if(len(trim(a1))>0,1,2)
and I even use in code where vba trim is not as good as if len(application.trim(range("a1")))>0 then
-- Don Guillett Microsoft MVP Excel SalesAid Software dguillett1[ at ]austin.rr.com "JLatham" <HelpFrom [ at ] Jlathamsite.com.(removethis)> wrote in message news:AB0EF7BD-CBCF-4D6C-9C76-5E6FA89A5519[ at ]microsoft.com...
[Quoted Text] > BTW: I don't think anyone directly addressed the issue of a difference > between a cell with a zero in it and a cell than actually has not had any > entry (or formula) entered into it: > YES, Virginia, er Pepper, there IS a difference, and Excel treats empty > cells differently at times than cells with either a zero length string ( > "" ) > or a zero in them. > > Here's one simple proof: > In cell A1 put "abc" (as abc) > skip over A2, never putting anything in it > in A3 put the formula = "" > in A4 type the value 42 > now, in A5 put this formula: =CountA(A1:A4) > and the result should be 3 even though only cells A1 and A4 have a visible > value in them, the zero length string in A3 is counted, but the EMPTY cell > at > A2 is not. > > For that matter, put a 0 into A1, A3 and A4 and formula = COUNT(A1:A4) in > A5 > and again the result is 3, showing that an EMPTY cell is not treated as a > zero for this purpose. > > There is a similar difference when programming in VBA, where we are given > specific tests for an empty cell vs one with a zero length string or a > zero > value, as > IF IsEmpty(Range("A2")) Then MsgBox"Cell A2 is actually EMPTY" > and > if Range("A2")=Empty then MsgBox"Cell A2 is still actually EMPTY, not > zero!" > > > > "Pepper" wrote: > >> Not a nice thing to say to a coworker, but was he right. >> I had a discussion with a male collegue about the difference between and >> empty cell and a cell with a zero. He called me a young, inexperienced, >> idiot woman because I insist his workbook is wrong since he did not >> account >> for empty cells that are awaiting data. >> He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and >> C16 and C17 dont have data yet - he insists his result is the G-d given >> answer. >> I suggest that =Average (C14:C17) is correct and that he is getting the >> wrong answer. >> It seems that his favorite feature, Conditional Formatting does not >> distinguish between Null and 0. >> Please settle the argument - I should complain to HR but will get far >> better >> satisfaction if I can site some reference and call him a #$&^ IDIOT! >> Thanks
|
|
No argument from me on any of that, with the AVERAGE() function being one of the exceptions to the general rule.
"Don Guillett" wrote:
[Quoted Text] > Not so for AVERAGE. > > However, I fully subscribe to that and take it a step further > if(len(trim(a1))>0,1,2) > > and I even use in code where vba trim is not as good as > if len(application.trim(range("a1")))>0 then > > -- > Don Guillett > Microsoft MVP Excel > SalesAid Software > dguillett1[ at ]austin.rr.com > "JLatham" <HelpFrom [ at ] Jlathamsite.com.(removethis)> wrote in message > news:AB0EF7BD-CBCF-4D6C-9C76-5E6FA89A5519[ at ]microsoft.com... > > BTW: I don't think anyone directly addressed the issue of a difference > > between a cell with a zero in it and a cell than actually has not had any > > entry (or formula) entered into it: > > YES, Virginia, er Pepper, there IS a difference, and Excel treats empty > > cells differently at times than cells with either a zero length string ( > > "" ) > > or a zero in them. > > > > Here's one simple proof: > > In cell A1 put "abc" (as abc) > > skip over A2, never putting anything in it > > in A3 put the formula = "" > > in A4 type the value 42 > > now, in A5 put this formula: =CountA(A1:A4) > > and the result should be 3 even though only cells A1 and A4 have a visible > > value in them, the zero length string in A3 is counted, but the EMPTY cell > > at > > A2 is not. > > > > For that matter, put a 0 into A1, A3 and A4 and formula = COUNT(A1:A4) in > > A5 > > and again the result is 3, showing that an EMPTY cell is not treated as a > > zero for this purpose. > > > > There is a similar difference when programming in VBA, where we are given > > specific tests for an empty cell vs one with a zero length string or a > > zero > > value, as > > IF IsEmpty(Range("A2")) Then MsgBox"Cell A2 is actually EMPTY" > > and > > if Range("A2")=Empty then MsgBox"Cell A2 is still actually EMPTY, not > > zero!" > > > > > > > > "Pepper" wrote: > > > >> Not a nice thing to say to a coworker, but was he right. > >> I had a discussion with a male collegue about the difference between and > >> empty cell and a cell with a zero. He called me a young, inexperienced, > >> idiot woman because I insist his workbook is wrong since he did not > >> account > >> for empty cells that are awaiting data. > >> He uses, for example, =C14+C15+C16+C17)/4 to calculate average value, and > >> C16 and C17 dont have data yet - he insists his result is the G-d given > >> answer. > >> I suggest that =Average (C14:C17) is correct and that he is getting the > >> wrong answer. > >> It seems that his favorite feature, Conditional Formatting does not > >> distinguish between Null and 0. > >> Please settle the argument - I should complain to HR but will get far > >> better > >> satisfaction if I can site some reference and call him a #$&^ IDIOT! > >> Thanks > >
|
|
|