Excel build bar graphs with oddly formatted data - excel

I am currently working in Microsoft Excel 2011 on Mac OS X. I am given a large amount of data in different tables and need to make 2 variable bar graphs with the data.
I understand that the usual way of orienting bar graphs in excel involves placing the data like so:
a b
A 1 2
B 3 4
However, currently I am going to be working with a lot of data of the format:
A a 1
b 2
B a 3
b 4
Is there a quick way to either 1) Convert the data from my format to the correct format, or 2) Build a bar graph (without having to go in and do a ton of customization of the data) with the current format.
Currently the only solution I have working is to manipulate and move the data around and rename everything once the bar graph is made, but this is very cumbersome.

I would do this in three steps:
Fill in the missing values for your first column. This is a bit manual. If the data is huge you consider writing a quick vba subroutine. At any rate, make the data look something like:
Highlight this data and turn it into a pivot table.
Structure the pivot table to match your bar graph data format and add a pivot chart:

Related

Excel chart's data is in Days, can I show the Series Data Lables in Months

My chart's range data ("A1:E6") looks is in the screen-shot below:
And this is my current Projects Timeline chart (Type Stacked-Bar):
As you can see, my Series Data Labels are in the same format as my Range, which is days.
Question: is there a way (without VBA) to format the Data Labels to be in months ?
For instance, for Project 3, Stage 1 (brown bar), instead of 90, it will display 3.
Is there a way to "out-smart" the screen below:
You can use helper columns in your data to calculate the desired value you want to show in the labels, i.e. divide by 30 to arrive at the approximate month value for any given number of days.
Then, depending on your Excel version and/or preference and/or need for backwards compatibility, you can either
manually edit each data label, hit F2 to edit the data label, type a = sign and then click the cell that has the calculated month value. Repeat for each data label, or
if that sounds too tedious, download and install the free XY Chart Labeler tool by Rob Bovey, which automates the steps above. You can share the file with people who don't have the tool, since all label references will end up hard-coded. The tool just helps reduce the manual labour to link labels to the cells. Or
with Excel 2013 or higher, use the formatting option in the Data Labels side panel to use "Values from cells" and select the cells where the converted values are. This feature is not backwards compatible and will show placeholder text instead of the labels if the file is opened in Excel versions earlier than XL 2013.
Edit: By the way, the formatting section that you highlighted in your screenshot is for formatting the numbers that the label shows. A format can only change the display of a value, but not perform a calculation, like convert number of days into number of months. So, the format approach is a dead end from the get-go.

Is such a chart possible in Excel?

I wanted to know if creating such a chart was possible using the functions available in Excel (inc. VBA)?
In summary:
The outer-bars update based on the extreme values of the data
The inner blue shaded bar represents one standard deviation around the
mean
The current value is the latest recorded value
The arrow represents the 3 month trend (either up or down)
I'd ideally want the above to be auto-refreshing with the source data coming in from an external feed.
Is the above possible using the charts + functions in excel? I'm not looking for a blow-by-blow description, just broadly the tools/methods you would use to achieve such a display of data.

How to change columns in a plot with shortcuts

I am working on Microsoft Excel 2013 scatter plots and I'd like to know if there is a more productive way to do this:
Let's say I have a big database with several columns and rows, i.e:
A B C D
1 Length Width Volume Area
2 2 1 8 4
3 3 2 7 1
4 1 5 3 5
5 7 3 12 6
I create a scatter plot for Volume vs Length. After, I want to create another scatter plot for Area vs Width. As I have a lot of rows and columns, I copy the first plot and then, in select data, I change the range of data (A to B and C to D). Is there an efficient way to do this? (shortcuts with arrows keys, maybe? I didn't find them) or Do I have to change the letters manually?
Thank you.
J.
If this is a repetitive task in the sense that every say Friday you create the same 4 graphs based on new information in the same columns then you could record a macro of what you are doing and associate it with keyboard short cuts.
You can do the same thing with updating the graph information but it starts getting a little more complex when you start changing which column number you want as the series. But its definitely possible. I would recommend recording a macro of you editing the 2-3 graphs so we can sheet patterns and we can probably help you edit the macro.
Now something simpler that may, and I stress may, suit your needs. Select A1 to D5, then insert you XY scatter plot. You X axis which should be common to all needs to be in the left most column. When you go to edit your source data you should see each series listed. You can simply check and uncheck which series you would like to have active. the ones with a check mark will display. If you needed three graphs at the same time, you could then copy and paste the graphs twice and set up each one to your desire without having to reselect the data range, you would just check/uncheck which series to display

How to show only months with data in a bar chart excel

Cash Comp
Jun-13 $121,112 $123,022
Jul-13 $116,289 $89,661
Aug-13 $94,718 $75,898
Jun-14 $108,699 $135,293
Jul-14 $109,685 $122,862
Aug-14 $84,161 $87,540
So I have data that shows like this in excel I want to do a bar chart that shows the difference but for some reason excel includes months that I'm not including in my data ex:(Sept-13, Oct-13, etc .....) My question is how do I stop excel from doing that and only showing the months I will like to include? Thanks
If your Column A looks like a date to Excel then it will format the x-axis of a graph to show time linearly. That's a sensible assumption. If you want to undermine that assumption by plotting just the months in the data range then reformat the "dates" as strings.
One way to do that is insert a new Col B with a formula like =text(a2,"mmm-yy") and use that as the series for the x-axis.

how to get excel graphs to use the same custom date format as a cell mmm-yyyy

I applied a custom date format to a column in excel (the format yyyy-mm e.g. 2014-01)
But when I create a graph out of three columns of data (date-in above format |heading|value) the graph shows the date as 2014-01-31 2014-02-28.
How do I get the graph to display the date as I have formatted it 2014-01 only without changing the data type to text and pasting values?
I use Excel 2013 but as far as I know this aspect of charting has not changed much since long before Excel 2010. Select your horizontal axis (click on one of the date labels in the chart), select Format Axis... and you should be able to:
(a) format the values there independently (ie in your case reapply the same custom format as applied to the raw data) and/or
(b) tick Linked to source.
The latter approach may not work quite as well as the former. Linked to source may mean something a little different from just "take the formatting from the raw data" (well, for the axis labels the format has to be common to all, whereas in the raw data it may differ between data points) and with a quick look I have not found details (for Excel 2010) of exactly what that is supposed to do.
So the former approach may be more reliable. For this you may need a two-step process. First make your custom format available to the chart and then apply it. That way the results are the same, chart and raw data formatting, though the process is not fully automatic.
One thing to watch out for is the linearity of the chart for time values. For example, if the dates are strings and the data points Jan 2012, Feb 2012 and March 2013 there is some risk the resulting labels will be equally spaced whereas the gap between the second two points should normally be thirteen times that of the gap between the first two points. Where data is approximately at regular time intervals anyway this is not always immediately obvious.

Resources