In this ironically titled “Let’s not get too excited” chart, David shows us how the twitter community is divided.
(Original image uploaded on Flickr here by McCandless)
It looks stunning. According to this chart, Out of every 100 people (or accounts) on twitter,
- 20 are completely inactive
- 50 are lazy or not tweeted in last one week
- 5 have more than 100 followers
- 5 are loud mouths and hence tweet heavily and contribute to 75% of twitter volume.
- 20 are gray colored
Out of curiosity I went to the source of these statistics from which the info-graphic is made.
If my reading comprehension skills are to be trusted, the article in no way conveys that all the above 5 types of people are different. That means there can be overlaps. So the chart is trying to mislead us.
In plain speak, it is possible that the lazy group can have more than 100 followers.
Also, the entire “completely inactive” group is a subset of “lazy” group. Point #6 in the original source for this graph confirms this:
Half of all Twitter users are not “active.” If you take a general description of being “active” on Twitter to mean that you have posted a tweet at some point in the last 7 days (1 week), then the survey learned that 50.4% of all Twitter users fit this category. If you remove the 21% from point #1, this leaves about 30% of users who have an account and have tweeted before, but happen to be inactive now.
What is the most important lesson this can teach us:
- Next time you plot percentages (either in a pie chart or square pie) make sure there are no overlaps. Each block is complete in itself. Otherwise, just list them down as a bunch of stats and don’t bother making a chart.
I am not too excited about this chart. What about you ?
PS: Follow me on twitter. I am not lazy, inactive, loud mouthed or gray.
PPS: Browse more examples of bad charts and learn from others mistakes.
4 Responses
Are the gray ones the tweet spammers?
Jon…no, the grey are either old folk, or younger folk with small children.
I know this because I noticed my hair colour in the mirror this morning, after the kids woke me up at some ungodly hour. If I wasn’t so green, they would be in danger of becoming pink. (I probably shouldn’t be so dark blue when discussing this stuff, or the boys in blue might knock on my door and take the kids away).
Chandoo…nice to see someone delving into the numbers, rather than just oohing and aahing over pointless prettyness. I laugh at the description of things like this as ‘info graphics’. Where’s the info? Without any info, why is this graphic interesting?
The other thing wrong with this picture is that there are no clowns (see http://peltiertech.com/WordPress/bad-bar-chart-practices-or-send-in-the-clowns/). A recent survey I carried out tells me that 3 out of 4 people are clowns . Even worse, no statisticians – that same survey I carried out tells me that 1 in 4 people are grey statisticians (sample size= me, the wife, and two inconsiderate kids. Note that the kids may be statisticians, but they did not indicate this on the sample form. But they are definately clowns)
Surely, the most important lesson this tells us is that Twitter is totally pointless.
Any bandwagon that politicians climb aboard and advocate as an importantway to reach out to their constituency should be avoided like the plague.
The thing that scares me most is that sooner or later, really important info will only be available on reprobate sources such as this.