Blogstats: BLOG ABOUT STATS.
All about dissemination of Official Statistics.
The views expressed on this weblog are those of the authors alone and do not necessarily reflect the views of employers.
Header
The headers of this blog show pictures of places where the International Marketing and Output Database Conference IMAODBC was hosted. See "About"
Some days ago the Australian Bureau of Statistics ABS has published an e-Magazine about important statistical terms: Statistical language! This is an “educational resource designed to improve the reader’s understanding of some fundamental statistical concepts”.
There are not only definitions but also precious explanations about how to calculate and about benefits and downfalls related to these concepts.
Starting with opening speeches of Kjell Jansson, Director General of Statistics Sweden and Enrico Giovannini, Chief Statistician of OECD. Powerpoint-presentations are found at http://www.oecd.org/dataoecd/30/32/40293021.pdf
Staffan Landin’s comment on OECD Factbook is worth to be upgraded to a special post! In fact all the data from this pocket book have been integrated into Gapminder Graphs Community and are now visualized in trendalyzer style. A huge work as all the tables had to bee uploaded to Google spreadsheets; it would be very interesting to know if there exists (or is planned) a database connection to upload data into Google spreadsheets. I am sure Staffan can give an answer to this issue.
Another example bringing together Statistics and one’s personal situation is about life expectancy (!). It comes from a site called “Understanding Uncertainty“.
” This site is produced by the Winton programme for the public understanding of risk based in the Statistical Laboratory in the University of Cambridge. The aim is to help improve the way that uncertainty and risk are discussed in society, and show how probability and statistics can be both useful and entertaining! “
Statistical information is so close to everyday life and - at the same time - often so far from our personal experience. It’s all about averages, medians … .
Attempts to bridge this gap exist.
Inflation is one prominent example. Several National Statistical Offices have developped inflation calculators giving users a tool to adapt data to specific personal situations and to give better insight in the methods (see the explanations introducing the calculators).
Gapminder launches a community for visualising statistics by Trendalyzer and Google Motion Chart. Interested organisations who wants to share their international data are invited. More information at the Gapminder website and in their new Gapminder Charts Development Blog. Have also a look at this nice example created in Motion Chart with Google Spreadsheet.
Swivel, well known for giving the possibility to visualize and comment data, has introduced Swivel Business some weeks ago. Swivel Business adds more of Web 2.0 and lets companies and offices visualize their data in closed user groups. Every member can comment (parts of) the graphs and get feedback from other members. To discuss with others is perhaps the best way to a better understanding of what a graph means.
This interactivity could be used inside Statistical Offices. These offices could also organize groups with external members, for instance with schools. Learning to interprete data and gaining statistical literacy in such a way is quite attractive and also a good opportunity for statistical offices to stay in contact with users.
To get into Swivel Business is not very easy (for the moment) as you have to wait for a telephone call from Swivel before getting a password.
What is the Google Visualization API?
The Google Visualization API lets you access multiple sources of structured data that you can display, choosing from a large selection of visualizations. The Google Visualization API also provides a platform that can be used to create, share and reuse visualizations written by the developer community at large. Read more at http://code.google.com/apis/visualization/ and look at the video.
The free online economic statistics in the 2008 OECD Factbook (www.sourceoecd.org/factbook) have been transformed to include animated diagrams of the key indicators which bring the data alive visually. The interactive graphs enable users to clearly observe trends in the statistics and interact with the data to perform their chosen comparisons between countries. The Factbook also includes increased data for key non-OECD Member Countries, Brazil, Chile, China, Estonia, India, Israel, Russia, Slovenia and South Africa.
Disseminating statistics to the public is the one side, making statistics accessible the other one. Even if huge databases are actually free and many statistical websites offered to the public there remain many obstacles to get the data to the (right) people.
Issue and agenda setting is one point, metadata another topic in this context. And presentation issues still another.
For the moment let’s concentrate to the latter case. Visualizing is one of the most important strategies for statistics to gain momentum. But that’s not enough: There are countless graphs on the web not being looked at. Maybe there is no (clear) message, maybe that the presentation itself is not attractive enough.
Animating graphs can change this on the presentation side. There are lots of possibilities to do this.
One excellent example of both engaged message and perfect presentation is Hans Rosling with Gapminder/Trendalyzer. Here animated graphs and personal presentation are used together, it is the full monty!
Six minutes has analyzed one of his presentations in detail and gives Six Simple Techniques for Presenting Data:
Explain the data axes
Highlight subsets of data
Dig deeper to unwrap data
Place labels close to data points
Answer the “Why?” questions
Complement data with energetic delivery
Hans Rosling adds number 7. “I may add one aspect, I am lucky to talk about a topic where the actual knowledge is surprisingly low in spite of interest being high, the development of the world.” Some more background about this commitment can be found in H. Rosling’s interview with Discover Magazine.
More and more Web 2.0 applications give us more and more possibilities to manage and share our bookmarks, our RSS feeds, our photos and so on: iGoogle, netvibes, linkedin, del.icio.us, flickr, twitter … but let’s be honest: We don’t have time for all these applications, we risk to get lost.
Now a second generation of applications emerges integrating a lot of these functionalities. twine is such a tool shed worth to be looked at, especially because Web 3.0 or Semantic Web tools (RDF) are used.
Give your email to blogstats at gmx dot net if you want an invitation to use twine beta.
Google has activated the trendalyzer graphic style known and bought from Gapminder (thank you Marc Ringuette for the hint in your comment). Google integrates trendalyzer, now renamed Motion Chart, in its spreadsheet. There you can choose several types of visualizations, i.e. Motion Chart.
It’s like in EXCEL and more: The graphs produced from the data can also be exported as gadgets and introduced in iGoogle or in whatever website. The link to Google spreadsheet remains necessary and gadgets can be updated with new data. In fact you don’t get autonomous graphs as known from other trendalyzer clones (see trend graphs).
How to:
In Google spreadsheet data have to be highlighted and with “Insert -> Gadget” a Motion Chart can be produced from these data.
As a gadget choose Motion Chart
Give a title, check data fields … and it’s done
As with other templates it’s not easy to produce good graphical representations. Good concepts and clear messages are needed.
I have made two examples with data from Statistics Switzerland. These are first steps and first lessons learned. To see them in motion you have to leave this wordpress blog not allowing to insert iframes in the content section (is there a trick to do so ?).
Widgets or gadgets or however named are little programs that can be embedded in your website or your netvibes, iGoogle, pageflakes, Facebook etc. . They take the intelligence from other sites on the web and give it to your site. Popular examples are calendars, weather forecasts, tag clouds …. or soon trendalyzer ;). A complex widget is for instance yahoo.pipes. There are thousands of them out there on www.widgetbox.com or www.programmableweb.com.
Official statistics are using RSS feeds extensively, but not the more elaborate widgets (true?). Search widgets could be a first and very useful test case on huge statistics sites where informations are often hidden to impatiently searching users. A small test can be seen on Statistics Switzerland (German , French ). It uses a widget for Google and a widget, that introduces a search tool with user feedback (www.eurekster.com).
Analyzing trends in Gapminder style has become a legend. With Google trendalyzer we hoped to get a user friendly tool to produce such graphs but unfortunately this widget has disappeared and Yonatan B, too.
In a recent post Eng. Hisham Abdel Maguid from Epic Systems has shown us this Flash applications http://www.epicsyst.com/main3.swf or http://www.epicsyst.com/visual.swf
He told us that data input is in Excel, xml and Access format and that a program interface with edit menus will be available soon.
And there is another demo in this style by Boomerang
There are several tools that offer interactive comparisons between websites. It’s quite amusing playing with such tools. I did it with websitegrader and got results like these for some statistics websites … and for blogstats:
In general I think these comparisons give a first hint and are just interesting but should never be used or understood as scientific explanations. Deeper investigations are needed.
A test with Statistics France shows only very small differences in a second report. The tool seems to be coherent in itself.
Some explanations about the measured categories (partially from Websitegrader):
“A website grade of 97/100 means that of the hundreds of thousands of websites that have previously been evaluated, Websitegrader’s algorithm has calculated that this site scores higher than 97% of them in terms of its marketing effectiveness. The algorithm uses a proprietary blend of over 50 different variables, including search engine data, website structure, approximate traffic, site performance, and others.” - From Websitegrader. Not very transparent!
“Google PageRank relies on the uniquely democratic nature of the web by using its vast link structure as an indicator of an individual page’s value. In essence, Google interprets a link from page A to page B as a vote, by page A, for page B. But, Google looks at more than the sheer volume of votes, or links a page receives; it also analyzes the page that casts the vote. Votes cast by pages that are themselves important weigh more heavily and help to make other pages important.” - From Google
Traffic rank by Alexa . Alexa is an online service measuring millions of sites on the Internet and comparing them.
Inbound links: One of the most important measures for a website is how many other sites link to it. The more links the better.
Google inexed pages: This number is the approximate number of pages that have been stored in the Google index. The Google web crawler will visit the website periodically and look for new content for its index.
… : 
