Charting

3 Charts from Ubuntu

I’ve just upgraded to the latest and greatest Ubuntu release, and I thought I’d take a look at what the various spread sheets install with it can do. I’ve got 3 on my system;  KSpread, oOO, and Gnumeric , although I think I may have installed Gnumeric myself. Any way here are some screen shots!

First up, Kspread, wow does this suck! It’s almost impossible to format the chart!

Screenshot-KSpread

Next, Gnumeric, which was much better, but still rubbish if I’m honest (I’m actually a big fan of Gnumeric, I think it has some really useful features).

Screenshot-gnumeric chart

And finally, oOO, easily the best of the bunch, and the only one you could even mention in the same breath as Office, but it’s still absolutely MILES behind!

Screenshot-Untitled 1 - OpenOffice.org Calc

So there you go, if you want to do charting stay well clear of open source spread sheets! Or have I been unfair, I’m not a skilled user of these products, am I showing that in this analysis?

Also if you know of any killer graphic (charting) apps, let us know by posting a comment. And I dont mean Tableau, or map grinder or we are human (or what ever it’s called), actual charts, that my boss stands an chance of understanding!

The Good and Bad of Bullet Graphs – again!

Mike Alexander said something interesting the other day and I thought I should let everyone know! Mike was taking about bullet graphs, not only how to make them in Excel, but actually about the really important thing, which is weather or not they’re any good! I’ve never used them in anger myself, but I have always liked the idea, and thought that they looked kinda nice – much neater and more logical then the dreaded Speedo gauge!

Mike suggested that because they showed so much data, senior managers find them hard to understand and you need to spend time explaining the chart – which is never ideal!
This surprised me as they are really quite simple little charts, but then I remembered how stupid, different intelligent, most senior managers are and I thought how plausible it actually would be. Anyway, I had a little think and here my suggestions to make them easier to read – I have no way of knowing if it will work, but it’s just a thought!

Bullet Charts

What do you think, would that make any difference?
As a bonus here’s a list of links to bullet chart stuff, to make these charts I used Mike’s Video, and it was super easy, good work Mike!

Steven Few original spec for this type of chat.

A nice post with some examples of more complext versions over at clearly and simply

The godfather of Excel reporting Charley Kyd has his method

Chandoo shows a cell based method!

PHD Pareto Analysis – agian

In PDH Pareto Analysis, Chandoo [Congrulations mate;-)))], shares with us a chart showing some pareto charting he was doing. This interested me because I didn’t get the chart. I’d like to open this up for debate, and hopefully learn something.
Here Chandoo’s Chart:
PHDPareto
Here are my thoughts on Chandoo’s chart – there probanly wrong!

  • The chart title doesn’t  provide us with much information about the chart.  I thought you where suppose to use titles that gave some insight into what the chart is telling us?
  • I don’t get what the bars show, for pareto I don’t care about the count of each X category, I just want to know the %, of X to Y, right?
  • Over all I don’t get what the chart is telling me. The 80% line is crossed at “Dashboards” or “100 Excel Tips”, which get about 5k’s each – but what’s that got to do with pareto?

Here’s how I show pareto relationships, which address the issues above
MIEPareto
I often use pareto for analysis stock profiles etc., where there are 1000’s of items, and these are 3 things I really want to know:

1.    Does this profile conform to 80/20 rule?
2.    If not, what’s 80% of the X by Y.
3.    How long is the 20% tail?

In practice I often add the lines, but don’t normally add the text, I think it’s clear what the lines are showing. I’m not so worried about the data in Chandoos chart, because it’s a bit misleading anyway (only 10 pages?), but what I hope my chart shows better is the 2 important data point 20% of X and 80% of Y and also how the tail looks.

Now that’s just my take on it and I’m almost certainly wrong! What’s your view?

DDOE Golf Charts, agian, agian

In Golf Charts Dick K posted some examples of charts he’d knocked up for his golf league, then in Golf Charts – Another Take, Tushar Mehta posted his take on the charts. In What would you do if a co-worker makes ugly charts? Chandoo asks the question, so I guess my answer is I’d do a blog post about them and try and do a better job [although I'd also guess I’ll fail and Jon P or Chandoo, will have to do it properly ;-)]
Tushars’ charts look OK, but Dicks’, (and there just no nice way to say this Dick, sorry) looks horrid, but also I think I have a few ideas that might be better at a showing the data. Of course they might not!!!!

Here we go.

Chart 1, Position Each Week.
Tushars’ step chart is a good way to show the data, in fact I think it’s probably the best way. I might have just bunged in a bar chart though, I think it works OK when there’s not many data points. I’d be tempted to also change the Y axis so that it was as long as the total number of positions in the league, that’s a bit more absolute.

Golf Chart Chart1

Chart 2. Best and Worst.
Again Tushars’ chart is an improvement on Dicks’, but maybe the Y axis could be reversed? I like the idea on the range, but an average would be useful, and how about a standard deviation; – a average without a deviation is like a King with out a crown! Showing the date (week number) is easy too. I also sorted the data by lowest score, that’s just good practice I think?

Having made this chart, I can’t say I love it. I’d like to scale up the error bars for the SD, but that would mean their size would not be proportionate to the Y axis scale. Adding a second Y axis would be rubbish too. I’m not sure this is a win, hummmm.

Golf Chart Chart2

Chart 3. Over Par, Over Handicap.
Dicks’ chart’ actually not that bad, I think it shows the data well, but Tushar brings up a very good point about positives. The issues with Tushars chart is that because the starting point is not the same its hard to get a handle on the over all performance of each player, which you can do in Dicks. I tried a load of combinations for this, but in the end settled on something simple. I think it works ok, maybe the colours could be better, but there’s quite a lot of information in this chart.

There is a bit of a problem here though and that’s with the actual data. If you play off 20, and drop 2 shots, thats not really as bad as if you play off 5 and drop 2 shots. A percentage I hear you say? Maybe…

Golf Chart Chart3

I lost a bit of interest in the rest of the charts, but then the last one spiked it a bit.

Chart X.X
This is simple, show the average an total scores. But when I looked at both Tushar’s and Dick’s charts 2 things struck me. Firstly the bars are meaning less, they show who has the best scores, but there is not reason they should be as long or a short as they are. Especially for first one the resolution could be higher.  Second thing was that I could not compare all the data at the same time, the 2 sets of bars next to each other are too much (admittedly this might just be because I’m limited!)

I was not sure how to make this better, I thought about a panel chart, but I couldn’t see how that would plan out. In the end I went for a bit of in cell charting, but in the end again, I had to use a real charts because in cells charts could not give me the resolution I needed. I never solved the second issue!

Golf Chart Chart4
On reflection, I don’t think I’ve brought much to the party, all my charts are bar charts(!), and one looks a bit like a box plot. I think I’ve got some way to go with charting!

Sparklines in Excel 2010 – why?

Don’t get me wrong. I’m looking forward to having spark/spike lines out of the box. In fact, I even asked for it!(See 3rd comment), and I think MS have done a bang up job (from what I’ve seen).

But this is what’s playing on my mind. How many Excel users even know what sparklines are? I would say with some confidence that it’s less than 0.5%.

Sparklines in excel 2010

How then does this fit with the idea of the Ribbon UI, which is designed for the (apparent) 90% of Excel users that can’t read and click a lists of words! – (menus). I know this arguments flaky, but I hope you can see my point, I know that the UI team is from a different part of MS than the Excel guys. But in my head it doesn’t stack up. Has the charting engine been fixed?

I’ve not actually touched 2010, but I just feel that after the relative disaster of 2007, it would not have been the first thing I’d have started working on. I hope they’ve ironed out all the over issues, they’ve had 13 years to work on 3D charts!

BTW, sparklines for Excel are not new, there are a number of addins that can make backwards compatible spike lines, relatively easily.

Free ;-)

http://www.spreadsheetml.com/products.html [bit pants]

http://www.bloggpro.com/nanocharts-milestone-release-v051/ [my fav]

http://sparklines-excel.blogspot.com/ [most comprehensive]

Paid

http://www.bonavistasystems.com/

http://www.bissantz.com/sparkmaker/index_en.asp