Analysis: If the graph makes a single point, it’s that all states regulate whether a weapon is carried in public or not, but the Northeast is the most highly regulated area for guns, while Southeast is the least. Beyond that, it’s not clear what the point of the graph is — how does the user learn more by interacting with the graphic? What are the subtle points of understanding does the user get from the diagram?

Looking closer at the graphic, I find a lot of problems with the execution:

• The categories are heterogeneous: carry in public is one type of regulation vs. what type of gun requires a permit vs. data points on what is regulated (missing firearms, locking devices and points of sales (gun shows, private sales). For regulation, these seem like a jumble of unrelated issues stuffed into one category. The categorization is the most sloppy part of this graph and does not add to the understanding of the issues.
• The underlying data is poorly documented, although there is a companion post as to how the creators struggled to summarize the data (ie fit the data to the visualization they were attempting. There is also a lot missing to annotate methodology: when the graph is scaled by population, is the outermost ring proportional? or entire slice? what’s the relevance of “population”? Wouldn’t it have made more sense to have the slices proportional to gun owners? What is the time frame for the data? Is there a data table and a clear methodology?
• Color choice seems arbitrary and changes for each data category. Does red mean “unrestricted” when referring to regulation but “concealed carry” permitted?

How the chart could be improved:

• Clearly define what the point of the visualization is. Is it to show that there are geographic differences in gun laws? Show how gun regulation impacts states with higher populations? Show that states with fewer gun laws have more/less violent crimes committed with guns?
• Consistent categorization scheme and color scheme. a two tiered category of gun types (hand guns, rifles, etc) and channels (gun shows, gun shops, and private sales)  with a simple scale (white= unregulated, light blue = partial regulation, dark blue = full regulation)
• Add a data table and a clear description of methodology used for categorization and how the data fits the categories.

Overall, I like the idea for this graph to display multiple dimensions, but think it is too muddled to provide any insights.

This shows three pyramids: Male on left, right hand side females from 1950, 2005, and 2050.  (2050 is projection based on current trends). From 1950 to 2005, the pyramid turned into a vase shape until 2050 when it is projected to be an inverted pyramid or wedge. What this visualizes is the aging of Japanese society — in a pyramid, the society has a large base of younger people. For the wedge, the base is inverted as the society ages, lives longer, and has lower fertility rates.

A variant of the chart was produced by the National Institute of Population and Social Security Research that shows three different population models for 2050, based on different fertility assumptions:

A time-lapse view of the chart is here.

An interesting view of the patterns of population growth is in this diagram:

This series of patterns shows population growth, stasis and decline.

One way I can think of improving the charts is to add a horizontal axis to show the male/female percentages by age band and also indicate average number of children by parental age by a color/shading within te band.

1) How Round isYour Circle (John Bryant and Chris Sangwin) [website] describes ways of physically representing mathematical equations. In one example, they model harmonic series divergence using stacks of dominoes which can be measured to show the proof. A harmonic series is 1+1/2+1/3+1/4+1/5.…

The Book, How round is Your Circle?, explores visualizing mathematical concepts using physical objects like dominoes, pantographs, and wooden models.

Dominoes can be used to model the divergence of the harmonic series .

2) Mathematical Models (Condy and Rollett) was a groundbreaking book in the early 1950s. Cundy in particular was instrumental in developing the “new math” curriculum in Britain in the 1960s. The book is filled with ideas on modeling mathematical concepts, including polyhedra, wire models, and quadratic surfaces. Bryant and Sangwin acknowledge the influence of this book on their work.

Polyhedra, from Cundy and Rollett’s Mathematical Models.

3) Architectural Models (Megan Warner) describes materials and methods of creating architectural models. While not strictly data modeling, the methods can be used to create models of 3d graphs and charts.

Model Making by Megn Werner gives methods an dmaterials for creating architectural models.

In a later post, I will create some Venn diagram models using both 2d and physical models as an experiment.

In this interactive graphic from Trefis it asl shows market size and share (on the right).

Flow maps are also called Sankey Diagrams and there is a site exploring visualizations .

Example Sankey diagram showing palm oil manufacture and derivative products

Example Sankey diagram showing palm oil manufacture and derivative products

Free tools:

Sankey Maker (includes a Sankey blog).

Flow Map Maker

Flow map tool and paper

Vernacular vs. Polite Architecture (from Brunskill, 1971)

Information-dense charts show multiple relationships in data in a single, chart. They take more time to read/review, but provide deeper insights. This week’s example from R.W Brunskill’Illustrated Handbook of Vernacular Architecture shows the evolution of “vernacular architecture”. Vernacular architecture ” will have been designed by an amateur, probably of the occupier of the intended building and one without training in design; he will have been guided by a series of conventions in his locality, paying little attention to what will have been fashionable on an international scale”.

This chart  shows multiple dimensions of data:

• Vernacular vs. Polite buildings over time (top chart line shows demarcation between vernaular and polite architecture)
• Proportion of Vernacular vs. Polite (Professioned ) over time
• Survival of vernacular buildings over time and size (lower line on chart)
• Number of vernacular buildings by suze and time period (century) (dots)

What can we learn from the chart?

• Most of the surviving buildings surveyed are are from 1700 — 1900 and small or large houses.  Great houses were built mainly before 1500
• By 1700, all the great houses were created by professional architects (“polite architecture”), after 1800, most large houses were done by professional architects also.
• Few cottages survive before 1800; in general there are few existing at all
• By 1900, most houses were built by architects and not vernacular architecture.
• The results need to be taken carefully as it is based on the surveys taken — how complete they are, what was missing or no  longer extant would impact the chart.
• The author modified the two curves , flattening the top (vernacular vs. polite) curve based on scholarly research available, even if the entire houses were not in existence. He deails them with the chart:

How could we improve the chart?

1. Add better definitions of the terms “polite”, “vernacular”, along with the criteria for the sizes (what is a great house). These are covered in the bo0k but a summary for the chart woul dbe useful
2. Add a data table with cross reference to identify which are the specific houses shown (if interactive, with mouseover, it could show the house information)
3. Better data labels on sections of the chart to make it self-evident.

Here’s an improved version:

Brief Take:
This chart combines a timeline, grouped scatterplot and trend line, and data table. It’s useful for showing how data segments change over time

Visualization using a modified polar area diagram in H. Wainer’s Understanding Unvertainty.

The author, Howard Wainer, comments that this chart has a flaw — while the length of the filaments is proportional to the size, there’s an implication that the reader is interpreting it as an area chart (as would be in the Nightingale rose which would use solid shapes, rather than filaments). Here each data point uses 9 filaments of varying lengths.       Here’s an alternative way to visualize the information using a simple bar chart:

Bar chart of sermon topics

Finally, a matrix view. Since I didn’t have the original data, I added 10 groups of respondents and also rounded to the nearest 10%. The rows contan the same data points as the original, but the columns show, for example, a group A which only preaches on two of the topics, while group B preaches on seven /10 topics. These lend themselves to additional grouping and insight. Perhaps goups B and C emphasize more social issues in their sermons, while group A focuses on other types of topics. This matrix view is more information dense (presents more data) and can yield more insights depending on the survey.

The Periodic Table, Eric Scerri, Oxford University Press 2007.

First, the ideas around periodicity evolved over time. Scientists like Lavoisier, John Dalton, Prout, and Dobereiner contributed ideas around the elements and their proerties. Dobereiner discovered that certain groups of elements — eg strontium oxide had approximately the same weight as the average of the weights of barium oxide and calcium oxide.

In formula:

SrO = CaO + BaO/2107 =  (59   +  155)/2

The three element groups (Sr, Ca, Ba) are called triads. Others are Br, Cl, I (Bromium, Clorine, Iodine). The German scientist Johann Gmelin developed an arrangement of elements by triads as follows:

Johann Gmelin’s arrangement of chemical elements based on triads.

A key factor in developing the periodic table was good data. In 1860, Stanislao Cannizzaro published a paper with a consistent table of atomic weights of the then known elements for the Karlsruhe Conference, the first worldwide conference on chemisty. The availability of good data (and a predictive method for new elements) facilitated the arrangement by periods.Meneleev’s creation of the periodic table has traditionally been ascribed to having written element and atomic weight on small cards (like index cards) that he arranged on  a desktop into periods. In 1869, he wrote a version of the periodic table on the back of an envelope, which he later refined.Looking at the process, having a physical manipulative (cards) and a physically written organization (diagram on the back of an envelope), facilitated discovering the order in the periodic table.At one point in my career I used to publish an annual data book with charts, data tables ‚and analysis around cloud services. To do it I created a prototype page and cut up each data table, graph, summary, and analysis section. We sat and reorganized the paper cutouts unti we found a pattern that told the story behind the data and uncovered the most insights. Something about shuffling paper and marking it up worked better than trying to do everything on the screen.

• The Periodic Table: It’s Story and Significance, Eric Scerri, Oxford University Press 2007.Eric Scerri’s book is a great summary of how the periodic table was created and what accelerated its development.
• Graphic Representations of the Periodic Table over 100 years, E.G. Mazurs, University of Alabama Press, 1974. Professor Mazur’s excellent collection of periodic table visualizations. Hard to find book — try a large university library.

Web Links:

• Wooden “manipulative” version  created  by Edward Mazurs.
• Chemical Heritage Museum  Mazurs’ collection of periodic tables. (Philadelphia PA)
• Periodic Table Catalog. Online collection of visualizations of the periodic table (circular, lemniscate, etc) organized by date.
• All Periodic Tables Great resource site with pictures of various periodic tables.

With a proportional Venn diagram, you add the dimension of size or count. For example, People who own a dog+cats (ab) would be much bigger than those owning a dog and fish (ac). Here are three examples using circles, ellipses, and rectangles to show proportional relationships:

Here’s an example of a proportional Venn Diagram using circles:

Here’s an example using Ellipses:

Here is a proportional Venn diagram with rectangles. To me it seems much easier to perceive relative proportions — note how easy it is to see the difference between AB and AC:

Tools:

• Euler APE  can produce both circles and ellipses.
• Draw Venn produces rectangles
• BioVenn  cicles — optimized for biological (genetic) series.