I have raw data in my spreadsheet that comes from a Google Form that looks like the following:
(Cost) (Source) (Frivolous) (Medium) (Comments)
A B C D E
1 15.94 McDonalds Yes Credit was hungry
2 98.32 School No Check Paid for textbooks
3 843.00 Hospital No Check Surgery
4 0 asdff Yes N/A Ignore this one woops
5
6 23.99 Dentist No Credit Check up
I want this data to always be copied to a different sheet, but ONLY the data that matches a condition. That condition in this case is if Frivolous is No, meaning I only want on this separate page to track valid important spending.
My second page I want them to look like the following:
(Cost) (Source) (Frivolous) (Medium) (Comments)
A B C D E
1 98.32 School No Check Paid for textbooks
2 843.00 Hospital No Check Surgery
3 23.99 Dentist No Credit Check up
Notice how empty entries are ignored and also entries with Yes under Frivolous are ignored as well.
How would I achieve this? I have absolutely no idea how that would work since I've only been able to achieve this through filter which will not work for this.
I would like to say a few words in defense of Google Spreadsheets and show some great functions that will work, but they are not supported by [excel].
Query
First you may use simple query:
=QUERY(sheet1!A:E,"select * where C = 'No'")
This single short formula will give the desired result, there's no need to fill right and down.
Filter
Actually you may use filter too. This function seems to work too:
=FILTER(sheet1!A:E,sheet1!C:C="No")
Please, read more info about this functions:
Filter
Query and full Query Language Reference
You'll find many exciting things that could be done in Google spreadsheets.
Actually, I was having some trouble with [google-sheets] ArrayFormula function so I used an old-school formula with SMALL and INDEX function in its array form. In A2,
=iferror(index(Sheet13!A$1:A$99, small(index(row($1:$99)+(Sheet13!$C$1:$C$99<>"no")*1E+99, 0, 0), row(1:1))), "")
Fill both right and down.
So you were in fact correct that this could be solved in [excel] with an identical solution as [google-spreadsheet]. However, there are superior methods in newer [exce] (2010+) using the AGGREGATE function that [google-spreadsheet] does not support and I'm sure that [google-sheets] has more elegant functions that I am not recalling right this moment.
Look to Sheet13 and Sheet14 here for the working sample.
Related
I have a spreadsheet that looks like this
A B C D E
-------------------------------------------------------------------------
Mercedes A Class C Class E Class G Wagon
BMW 1 Series 3 Series 4 Series
and I wish to create a spreadsheet that looks like this
A B
------------------------------
Merecedes A class
Merecedes C class
Merecedes E class
Merecedes G Wagon
BMW 1 Series
BMW 3 Series
BMW 4 Series
I have used the transpose function, but it doesn't create a new record for each value, but rather just rearranges the columns.
How might I achieve this?
I have been looking for this same answer.
What you are trying to do it called unpivoting but it doesn't look like there is a built in function in google sheets to do it.
I found this fix in the google product forums. https://productforums.google.com/forum/#!topic/docs/5UAxnQKMFm8
At first glance it looks a little complicated but victor tabori's code worked like a charm.
copy and paste his code into tools script editor in your google sheet and then use the =unpivot function in your spreadsheet. Only limitation I ran into was it failed trying to reference data in a different page of the sheet.
For the sake of anyone who's search brought them to this old question on this perennial need; this type of rearranging is a bit easier than it used to be, with Google's addition of the FLATTEN function. Example formula:
=ARRAYFORMULA(QUERY(SPLIT(FLATTEN(A2:A&"|"&B2:E),"|"),"select * where Col2 is not null"))
From the inside out, here we work on a temporary array of makes and models,
with a symbol | concatenated between makes from A and models found in B:E,
FLATTEN then puts it all into a single column,
then SPLIT uses the temporary | characters to split it to a two-column array,
and finally, a quick QUERY lets us filter out blanks after the fact (like the row E3 would become), without having to do all the calculations twice.
Oh, and ARRAYFORMULA lets it work across the whole array of makes and models we input.
Good Evening and thanks in advance for taking the time to read and help.
I have a 3 column excel file which I am trying to populate the 3rd column with a return value found next to the row its found in.
so for example I want to look at column MANAGERSFULLNAME for value
Cheryl Rommelfanger and find the match in column FULLNAME. Once the match is found I want to populate MANAGERSX2FULLNAME but not with the value found in FULLNAME but with the value next to in column MANAGERSFULLNAME
So for this example we look in MANAGERSFULLNAME for Cheryl Rommelfanger and find the match in FULLNAME Cheryl Rommelfanger then populate MANAGERSX2FULLNAME with
William Dearth
FULLNAME MANAGERSFULLNAME MANAGERSX2FULLNAME
Dena Peters Cheryl Rommelfanger
Kyle Marsh Melissa Hall
Cheryl Rommelfanger William Dearth
ive tried a few things and can only get a count not the value next to it.
=MATCH($E2&$F2,INDEX($B2:B4000&$C2:C4000,),)
=IF(ISERROR(MATCH(E2,F2,$B$2:B$4000,$C$2:C$4000,0)),"",E2)
=IF(ISERROR(MATCH(L2,$K$2:K$4000,0)),"",L20)
any help would be greatly appreciated.
So I apologize but I am having a bit of trouble understanding your columns, but the general idea is clear.
Your attempts are really close. You want to use index(match) as opposed to match(index). The link below describes how to do this.
Index match formula
If I'm understanding you correctly it sounds like you're trying to find and list the bosses boss so-to-speak to display a hierarchy of sorts. I'm using just columns A, B, and C (C being the managerx2fullname) this formula should work fine:
=index(B$2:B$4000,match(B2,A$2:A$4000,0))
You will of course need to change the columns to fit your needs. Don't include a dollar sign in B2 because you want this to increment as you drag the formula down the column. The link below shows a screen shot from my test. In it we see that in row 2 John is Adams boss, who in turn is Joe's boss. I think that's what you're shooting for here.
Screen shot
I am trying to find a closest absolute value with index match. I looked at several other posts like here but what i am trying to do is a bit different as i want to add multiple search criterias.
As you can see , I am trying to get the absolute closest time for a specific person.
I am using the formula =index(C2:C21,match(F4,B2:B21,-1),match(E4,A2:A21,0)) and I had to copy column B in column C to make my 1st match work. The result is shown in G4. Unfortunately I am struggling to get the correct result.
Effectively I would like use the formula that was posted in the previous post (see link at the top) =INDEX(E2:E21,MATCH(TRUE,INDEX(ABS(D1:D21-I4)=MIN(INDEX(ABS(D2:D21-I4),,)),,),0))
with with a search criteria (the name of the person).
Any help would be much appreciated
Thank you
Thanks #avram
I still end up with some cases where the formula does not work. See below. in G6 and G7 i should get 10:25. (You can ignore column A)
Try this formula in G4,
=index(C$2:C$21, match(min(index(abs(index(C$2:C$21+(B$2:B$21<>E4)*1E+99, , )-F4), , )), if(B$2:B$21=E4, abs(C$2:C$21-F4), 1E+99), 0))
This will work in either google-sheets as a standard (non-array/non-CSE) formula or excel as an array (CSE) formula.
If anyone else wants to tackle this problem with a more elegant formula, you can copy the sample data from this publicly shared google-sheet.
Index match find closest value with multiple search criteria
Perhaps this may exempt a fourth person from retyping the same tired data that the op delivered in image(s).
A very simple approach using a "helper" column with data like:
We want the closest absolute match for larry to 10:15 AM. We enter larry in E1 and 10:15 AM in F1
Then in D2 we enter:
=IF(A2=$E$1,ABS(B2-$F$1),"")
and copy downward. (this is the absolute difference for larry) Finally in E2:
=INDEX(B:B,MATCH(MIN(D:D),D:D,0))
With bigger tables having more columns, it is very easy to add additional criteria if needed.
This answer uses Array Formulas which must be entered using CTRL+SHIFT+ENTER. It's kind of complicated, so I'll do my best to explain and will revise if necessary. Here's a screenshot:
Here is the formula in its raw form; names are entered in column A, Times in Column B.
=INDEX(B1:B7,MATCH(MIN(IF(A1:A7=D2,ABS(E2-B1:B7),"")),IF(A1:A7=D2,ABS(E2-B1:B7),"")))
As you might suspect, it uses INDEX/MATCH to get the job done, but the key is using an IF statement to generate both the search criteria and the array that the MATCH function searches within. Let's break it down.
Sec 1, Match Search Array
IF(A1:A7=D2,ABS(E2-B1:B7),"")
This creates the Search array for the match function. If the name in D2 (our criteria) is equal to the name in the search array, it return the absolute value of the difference between the criteria time and the time in the array we're searching. Otherwise it returns a blank value. Do not use 0 for this as it will skew the match result.
Sec 2, Match Search Criteria
MIN(IF(A1:A7=D2,ABS(E2-B1:B7),""))
This tells us the smallest value in the above array. We use this value as the search criteria in the MATCH function.
Sec 3, putting 1 & 2 Together
MATCH(MIN(IF(A1:A7=D2,ABS(E2-B1:B7),"")),IF(A1:A7=D2,ABS(E2-B1:B7),"")) This searches for the smallest abs difference defined in Section 2 within the array created in Section 1 and returns the row number.
Sec 4, Indexing the times
=INDEX(B1:B7,MATCH(MIN(IF(A1:A7=D2,ABS(E2-B1:B7),"")),IF(A1:A7=D2,ABS(E2-B1:B7),"")))
This returns the time value from column B in whatever row is identified by the Match function above.
Hopefully this all makes sense. Remember to enter it as an array formula.
Thank you ahead of time for anyone who can help me with this, I think I am close, but it still isn't working.
I have a simple sheet activity reporting sheet that I am asking staff to complete over the upcoming year - It has 5 columns:
Column A: Date -In format (4/4/2013 13:30:00)
Column B: Title -In format (text string)
Column C: Attendance -In format (Numbers)V
Column D: Vol led - In format (text string)
Column E: Staff Led - In format (text string)
Using this data I am 90 % positive that I can aggregate on a different summary sheet that contains some static data like months (in the B column) to aggregate on. I am having trouble configuring the criteria in the filters though to cause the correct output to either sum or count .
Quantity of events ed by either staff or vol, if neither box is checked the event should not be counted) Right now I am trying this but it is not working
=SUM(FILTER('Hostel Activities'!A:A,MONTH('Hostel Activities'!A:A)=$B3, NOT(AND(ISBLANK('Hostel Activities'!D:D),ISBLANK('Hostel Activities'!E:E)))
Total number of attendance in a month for activities led by staff or volunteers Right now I am trying this but it is not working
=SUM(FILTER('Hostel Activities'!A:A,MONTH('Hostel Activities'!A:A)=$B3, NOT(AND(ISBLANK('Hostel Activities'!D:D),ISBLANK('Hostel Activities'!E:E)))
THIE WORKS! ## Heading ##Total number of volunteer led activities in a month for activities Right now I am using this and it IS working
=COUNT(FILTER('Hostel Activities'!A:A,month('Hostel Activities'!A:A)=B3,not(isblank('Hostel Activities'!E:E))))
Thank you for any assistance and/or guidance
Danny
The first problem I see with your first two formulas is that you're calling SUM on your FILTER result. But the FILTER is returning the column A, which are dates. So, your basically summing dates, which will surely not yield the result you're looking for. Why are you not using COUNT, as you did on your last formula?
Second, the first two formulas you pasted are identical, how do you expect them to return different results?
It seems that for the first two want to sum an OR condition. You can do this two ways (that I can think of now). The simpler to understand is just to sum two COUNT(FILTER(... formulas, one for each criteria, e.g.
=COUNT(FILTER('Hostel Activities'!A:A,month('Hostel Activities'!A:A)=B3,not(isblank('Hostel Activities'!D:D)))) + B6
Assuming that on B6 is the other COUNT formula (the 3rd one, that already works).
Another option would be to use an OR function as criteria for the FILTER. Like this:
=COUNT(FILTER('Hostel Activities'!A:A,month('Hostel Activities'!A:A)=B3, OR(NOT(ISBLANK('Hostel Activities'!E:E)), NOT(ISBLANK('Hostel Activities1!D:D))) ))
I believe I have figured out a method that works by making some adjustments in the formulas and the source data.
Basically
IN THE SOURCE REPORTING DATA:
I combined columns D and E into the same column and added data validation so the coordinator has to enter if the activity is led by staff,volunteer, or neither.
IN THE MONTHLY AGGREGATION REPORT:
To count the number of activities led by either staff or volunteers I used this :
=COUNT(FILTER('Hostel Activities'!A:A,month('Hostel Activities'!A:A)=B3,'Hostel Activities'!D:D="Staff"))+E3
*E3 is the count of volunteer led activities which is found using this formula:
=COUNT(FILTER('Hostel Activities'!A:A,month('Hostel Activities'!A:A)=B3,'Hostel Activities'!D:D="Volunteer"))
Adding up the number of participants in activities run by either staff or volunteers was a little more difficult, but I was able to do it by adding up 2 unique equations. I would prefer using an OR statement in the filter criteria, but I just couldn't get that to work. This is how I was able to make it happen:
=SUM(FILTER('Hostel Activities'!C:C,month('Hostel Activities'!A:A)=B3,'Hostel Activities'!D:D="Staff")) + SUM(FILTER('Hostel Activities'!C:C,month('Hostel Activities'!A:A)=B3,'Hostel Activities'!D:D="Volunteer"))
Thank you all for your assistance
I am analyzing an electronic survey I made using Google Forms and I have the following problem.
One of the questions can take multiple answers in the form of Checkboxes as shown in the picture below. The question is in Greek so I have added some Choice1, Choice2, Choice3 etc next to each answer in order to facilitate my question.
In my data when someone chose lets say Choice1 and Choice2,
I will have an answer which is the concatenation of the strings he checked seperated with commas.
In this case it would be:
Choice1, Choice2
If someone else checked Choice1, Choice2 and Choice4
his answer in my data would be:
Choice1, Choice2, Choice4
The problem is SPSS has no way of seperating the substrings (seperated by commas) and understanding which Choices each case has in common. Or maybe there is a way but I don't know it :)
When I, for example, do a simple frequency analysis for this question it produces a table that perceives
Choice1, Choice2
as a completely different case from
Choice1, Choice2, Choice4
Ideally I would like to somehow tell SPSS to count the frequency of each unique Choice (Choice1, Choice2, Choice3 etc etc) rather than each unique combination of those Choices.
Is that possible? And if it is can you point me to the documentation I need to study to make it happen?
Thx a lot!
Imagine you are working with the following data, which is a CSV file you have downloaded from your online form. Copy and paste the text below and save it to a text file named "CourseInterestSurvey.CSV".
Timestamp,Which courses are you interested in?,What software do you use?
12/28/2012 11:57:56,"Research Methods, Data Visualization","Gnumeric, SPSS, R"
12/28/2012 11:58:09,Data Visualization,"SPSS, Stata, R"
12/28/2012 11:59:09,"Research Dissemination, Graphic Design",Adobe InDesign
12/28/2012 11:59:27,"Data Analysis, Data Visualization, Graphic Design","Excel, OpenOffice.org/Libre Office, Stata"
12/28/2012 11:59:44,Data Visualization,"R, Adobe Illustrator"
Read it into SPSS using the following syntax:
GET DATA
/TYPE=TXT
/FILE="path\to\CourseInterestSurvey.CSV"
/DELCASE=LINE
/DELIMITERS=","
/QUALIFIER='"'
/ARRANGEMENT=DELIMITED
/FIRSTCASE=2
/IMPORTCASE=ALL
/VARIABLES=
Timestamp A19
CourseInterest A49
Software A41.
CACHE.
EXECUTE.
DATASET NAME DataSet2 WINDOW=FRONT.
LIST.
It currently looks like the image below--three columns (one timestamp, and two with the data we want):
Working with some syntax from here, we can split the cells up as follows:
* We know the string does not excede 50 characters.
* We got that information while we were reading our data in.
STRING #temp(a50).
* We're going to work on the "CourseInterest" variable.
COMPUTE #temp=CourseInterest.
* We're going to create 3 new variables with the prefix "CourseInterest".
* You should modify this according to the actual number of options your data has
* and the maximum length of one of the strings in your data.
VECTOR CourseInterest(3, a25).
* Here's where the actual variable creation takes place.
LOOP #i = 1 TO 3.
. COMPUTE #index=index(#temp,",").
. DO IF #index GT 0.
. COMPUTE CourseInterest(#i)=LTRIM(substr(#temp,1, #index-1)).
. COMPUTE #temp=substr(#temp, #index+1).
. ELSE.
. COMPUTE CourseInterest(#i)=LTRIM(#temp).
. COMPUTE #temp=''.
. END IF.
END LOOP IF #index EQ 0.
LIST.
The result:
This only addresses one column at a time, and I'm not familiar enough to modify it to work over multiple columns. However, if you were to switch over to R, I already have some readymade functions to help deal with exactly these kinds of situations.
Unfortunately there is no easy "built-in" way to achieve this, but it is certainly achievable with spreadsheet formulae, or Google Apps Script.
Using formulae, assuming your check box question lands in column D, this will produce a "normalised" list:
=ArrayFormula(TRANSPOSE(SPLIT(CONCAENATE(D2:D&",");",")))
and you can turn that into a two-column list and QUERY it to return a table of frequencies:
=ArrayFormula(QUERY(TRANSPOSE(SPLIT(CONCATENATE(D2:D&",");","))&{"",""};"select Col1, count(Col2) group by Col1 label Col1 'Item', count(Col2) 'Frequency'";0))
If your locale uses a comma as a decimal separator, replace {"",""} with {""\""}.
It is easy to split the fields into separate variables as described above. Now define these variables as a multiple response set (Analyze > Tables > Multiple Response Sets), and you can analyze these with the CTABLES or MULT REPONSE procedures and graph them using the Chart Builder