Dumbing Down The SOTU


State of The Union Speeches Reading Levels from SocProf on Vimeo.

The interactive graphic from the Guardian is here. The Flesch-Kincaid reading level measurement is here.

The real question is why this is happening, especially considering the fact that Americans are more educated now than they were in the days of Washington or Madison. One suggestion might be that the SOTU is now a TV spectacle, written by communication specialists, not for an educated audience. Early SOTU were probably heard and read only by a few. The spectacle dimension directly alters the content.

The Visual Du Jour – Foreign-Borns: Then and Now

Check out this great infographic from The Census Bureau on migration:

Foreign Born infographic image

[Source: U.S. Census Bureau]

There is a lot that is interesting here. First of all, the percentage of US population that is foreign-born is lower now than it was in the 19th century so, even though the foreign-born population is larger in numbers, because the US population has also grown larger, that percentage has varied in a different direction than the raw numbers.

What is not surprising is the shift in countries / regions of origin. The 19th century was the time of European migration while the more recent trends have shifted to Central America and Asia. That migrating population is also younger and distributed more widely across the US rather than concentrated on the coastal areas.

[As a side note: thanks to the Census for providing full embedding code for this infographic.]

The Visual Du Jour – Open Governance (Or Not)

I found this interesting set of data via The World Bank Dataviz Tumblr: some good interactive visualizations based on the Open Budget Survey (full report here):

Open Budget Survey from SocProf on Vimeo.

Here are some static images.

The map:

The rankings:

The overall view:

The categorized rankings from the report:

The infographic:

These are all good examples of the changes brought about by global governance mechanisms that allow different international organizations, non-profit, civil society groups, etc., to demand accountability from governments, in this case, on budgetary issues and processes.

The Visual Du Jour – Global Nip/Tuck

A while back, I reviewed a book on the sociology of plastic surgery, Making The Cut. That book was prescient and the trends it discussed have not abated in the context of the cosmetic surgical culture:

“In the new economy nothing is more sexy than surgery. From Botox to lipo to tummy tucks and mini-facelifts, the number of cosmetic surgery operations undertaken around the globe has soared recently, as consumers spend more and more on themselves in the search for sex appeal and artificial beauty. In a society in which celebrity is divine, information technology rules, new ways of working predominate and people increasingly judge each other on first impressions, cosmetic enhancements of the body have become all the rage.” (7)”

And so, the Economist has a chart with more recent data on the trend:

And this cosmetic surgical culture has gone global.

As Elliott stated in his book:

“My argument is that the new economy spawned by globalization intrudes traumatically in the emotional lives of people – with many scrambling to adjust to today’s routine corporate redundancies. (…) The dramatic changes now occurring in the global electronic economy and on the ways in which corporate layoffs, downsizings and offshorings are affecting people’s sense of identity, life and work. (…) Many have reacted to this sense of social dislocation and economic insecurity – what I term today’s pervasive sense of ambient fear – by turning to forms of extreme reinvention in general and cosmetic surgical culture in particular. Many are calculating that a freshly purchased face-lift or suctioning of fat through liposuction is the best route to improved live, careers and relationships.” (9)

As I wrote in my review, the cosmetic surgical culture is an individual response to a social-structural issue (C. Wright Mills, anyone?), that is, the pressures of corporate life and the global economy. In the context of general economic insecurity, even for social classes that not so long ago considered themselves secure (after all, the 1980s layoffs affected mostly industrial workers, but, as conventional wisdom went, it is just an upgrading of the economy. Once the labor structure moves away from union-heavy industrial labor towards a more education service-trained workforce, then, everything will be fine… because brown people will never be able to do the educated, technological jobs of the service economy… how did that turn out?).

In the cosmetic surgical culture, the personal  / subjective responds to the structural / objective. As the global conditions trickle down to individual societies and ultimately to individuals, they generate uncertainties (and Elliott does not mention the risk society but I think this theory is relevant here) regarding work, relationships and life in general to which the cosmetic surgical culture is a response.

Visualizing The Transnational Capitalist Class

Via Nathan Yau (who did write the book on visualization and has a great website to go along with it), this very interesting and interactive visualization of the world’s billionaires:

Billionaires 2013 from SocProf on Vimeo.

Here are a few static images.

The rankings:

The bar charts:

The ranking shifts over one year:

The plots:

The map:

In case you wondered to whom I was referring in the video on the transnational capitalist class, it was, of course, Leslie Sklair who also wrote the book The Transnational Capitalist Class. This book is a must-read for anyone interested in the sociological study of the dominant class in the global system. Then, you should read William I. Robinson’s A Theory of Global Capitalism.

The Visual Du Jour – Where The Brown Shirts Are

Here is an interesting data visualization from Der Spiegel, on the rise of the Nationaldemokratische Partei Deutschlands (NPD), the German far-right party.

As you can see from the map and the post title, I don’t think the choice of brown dots as color scheme is random. It is a rather simple data visualization but it clearly shows the areas of greater influence of the NPD, as measured through voting rates. It shows rather clearly where the NPD has gotten some popularity (i.e.: the former East Germany).

That being said, I am not a big fan of dots because they make proportions / rates hard to tell. I know there is the legend on the left but once you start working on the map, can you really tell, beyond the areas of greater aggregations, exact percentages (when those are not given in the textual notations on the side?).

And if the brown is designed to underline some political ugliness, it succeeds.

It is a bit of a shame though that the article does not provide any explanation for this. Maybe the reasons are obvious to Germans, but I got this as part of the international, English-language edition, and not all readers (including me) may be aware of the subtleties of German party politics. Although I was aware that the former DDR is now the hotbed of far-right politics (for reasons of downward mobility, economic dislocations, and precarization), but I was hoping for more.

The Visual Du Jour – GOOOOOOAAAAL

I love soccer, or as we Europeans call it, football. I also like network visualizations. So, how can I not love this network visualization of Ballon D’Or votes. If you don’t know what the Ballon D’Or is, check this out.

Quick explanation:

“The above visualization shows the network of votes of the Ballon d’Or 2012. Voters and voted for players make up the 524 nodes of the graph. Node size is based on indegree. The 1513 edges are based on the given votes, with each of the voters having three votes: 1st place 5 points (thickest line), 2nd place 3 points, and 3rd place 1 point (thinnest line). Node color indicates either being a captain (red), coach (violet), journalist (blue), or player who did not vote (green).”

There really seems to be two players and everybody else: Messi and Cristiano Ronaldo just dominates the whole thing.

The Visual Du Jour: The Power Elite – Chinese Style

This is an interactive infographic, so, here is a quick video:

The Power Elite – Chinese Style from SocProf on Vimeo.

I initially found this in a blog post in Le Monde but the original article is here.

Note the consistency in color schemes in the different visualizations.

A few snapshots (not interactive, of course):


A family tree (well… family bubbles):

What privilege is:

Still The Gun Thing…

A few publications (here and here, for instance) decided to do some raw data collection (which would need to be refined and correlated with other variables to be truly useful). Slate, especially, has produced a simple interactive graphic regarding gun deaths since Newtown, which has received much visibility:

Less spectacular but as important is this article on the NRA’s (successful) lobbying efforts to suppress research on this topic:

“One aspect of the political effort to turn the US into a gun culture was laid bare just before Christmas inan editorial published in JAMA by Arthur Kellerman and Fred Rivara, two public health physicians. They present a shocking and well-described perspective not available elsewhere — a story of how politics, funding, and sociopathic profiteering have combined to thwart public health research, ultimately creating a smoother path for corporate interests that exploit citizens and their lives just as cigarette manufacturers did a few decades ago — minimizing risks and dismissing deaths in order to make their money. By tying their business to freedom, gun manufacturers and their shills have been able to make incredible inroads into our political system. How much so? They’ve been able to stifle research into gun violence for more than 15 years.

Kellerman and Rivara write that in 1996, pro-gun members of Congress succeeded in eliminating the National Center for Injury Prevention and Control at the Centers for Disease Control and Prevention (CDC). As if defunding this center weren’t enough, the following language was added to the appropriations bill:

. . . none of the funds made available for injury prevention and control at the Centers for Disease Control and Prevention may be used to advocate or promote gun control.

So, the CDC lost a center devoted to injury prevention, and lost the ability to shift funds to study gun violence. Later, when other agencies tried to fund high-quality research on injury prevention, which naturally touches on firearms, Congress extended the restrictive language, ultimately applying it to all the Department of Health and Human Services agencies, including the National Institutes of Health (NIH).”

Go read the whole thing. But this tells you everything you need to know about an organization that refuses discussion of an issue, will have swarms of trolls derailing discussion threads all over the Internet, and using its lobbying power to get rid of research (which means, they know what the research might show, because the data otherwise available points to a clear direction). This is bully behavior, unsurprisingly.

The MOOC Experience

So, the MOOC is over. It has been a very interesting six week but I made it. I completed all the projects and I am now waiting for my completion certificate. So, what has this MOOC experience been like?

To recap, the MOOC I took was Introduction to Infographics and Data Visualization, offered through the Knight Center for Journalism in the Americas, University of Texas at Austin, and taught by Alberto Cairo. I wanted to take a MOOC because it seems to be the thing right now. You can find almost every day an item about MOOCs in the Chronicle of Higher Education or Inside Higher Ed. I tend to be skeptical of buzzwords. So, I thought, at the very least, and before passing final judgment in MOOCs, I should take at least one. And I did not want to take a sociology course because *yawn*. I thought I might as well try to learn something in the process of testing out the format. After all, I have already let it be known at my college that I wanted to develop MOOCs there.

So, I randomly bumped into a mention about the data visualization course on Twitter (who said Twitter was useless!) and decided to jump. This turned out to be the most interesting experience I have had in a while and it turned out to be the right decision. Since I had registered early for the course, I decided to read Alberto Cairo’s book, The Functional Art, to get a head start. So, by the time the course started, I was ready.

The Course

The course itself was 6-week long, structured week-by-week, increasing in difficulty and complexity as weeks passed. Because it was the first MOOC for the Knight Center, we could tell that a few course corrections were made along the way. Also, since the course was offered in-house (as opposed to through Coursera or Udacity), enrollment was limited to 2,000 students (small number for a MOOC).

Every week, we had some reading to do. All the materials were provided, we didn’t have to buy anything, books or software. We were also provided with video lectures by Alberto Cairo, based largely on his book, as a series of Youtube or Vimeo videos on the specific topic of the week. On weeks 1 and 2, we had to complete quizzes. Those were a bit of a joke, I have to say: 2 quizzes, 5 questions each, 30 minutes to complete. It is probably why those disappeared after week 2. I did not miss them.

In weeks 1 and 2, our work on data visualization was mostly critique. Alberto Cairo gave us infographics and we were supposed to analyze them and determine what worked, what did not, and what we would do to improve it. We were supposed to post our analysis on a message board, and then, post twice more to respond to other students’ analyses. The first week, there was only one message board and that was a mess. I posted early (I think I was second), then waited for other people that I could respond to. Very quickly, I was overwhelmed when the floodgates opened. Not only that but for the late posters, there really wasn’t much left to say about the original infographic. There is only so much critique that can be done. So, I posted my two other responses and left that first discussion board. There was just too much going on there.

The course organizers must have realized that as well because by Week 2, they had created 10 message boards and we could pick whichever we wanted. I picked #10 and stuck with it. I was not the only one because I saw the same names cropping up again week after week (Board 10 FTW!). We are a gregarious species. Apart from that, Week 2 went by pretty much as Week 1 had: readings, video lectures, quiz, infographic critique.

Things got really interested in Week 3 (gone were the quizzes) where we were asked to start designing, even if a basic sketch of an alternative infographic based on some data Alberto Cairo gave us. This was mine (don’t laugh):

Initially, this was truly daunting for me. I had never really done that. But that is where I realized how this course was: a whole bunch of us had to jump in the pool and do something we hadn’t done before. In addition, we had to comment on two other projects from our classmates. Everybody was pretty thoughtful about their comments but, obviously, the skills gap between students was pretty immense. Some people were obviously already professional designers. Others like me are pretty decent with data but less so with the design part. And still others were truly novices at the whole thing.

Weeks 4, 5 and 6 were designed the same way and I have already posted my work on these weeks herehere and here. I will also be posting all my data visualization work, beyond this MOOC, in this Flickr set. So, for these three weeks, most of work was watching the video lectures, project design, and then spend some time in the message board looking at other people’s work and offering comments.  Theses discussions were very stimulating and it was exciting to see other people’s work and ideas, irrespective of skills level. Overall, the project took most of my concentrated time while working on the message board was more scattered. For the last three weeks, I would estimate that I spent about 6 hours a week on this course (including one all-nighter!). I spent less than that in weeks 1 and 2. It was fine, I never felt rushed but I always did want to finish my project work early because of my own full-time job. But it was all very manageable.

The Students

This was a major surprise to me: about 2,000 students, and, according to the Knight Center, people from over 100 different countries. I am not surprised. You could tell, through the discussion boards, that this was an extremely diverse crowd, a lot of  non-native-English speaker people (like me). I don’t think many of them were college students either. It really felt that we were all adults, working professionals in a variety of fields, like journalism, design, academia and non-profits. This was truly a global course and I hope it stays that way because it was fascinating, especially on week 6 where we could pick our own project, to see what topic people selected. The diversity of interests and approaches was breathtaking.

The Instruction

Alberto Cairo was the official instructor for the course and from what I could tell, one or two assistants helped him manage the course, especially the message boards (although I did not notice any trolling or inappropriate behavior at all in there, which is a fear I had read about before… not here, everyone was focused on the work). However, Alberto was in constant contact with us either through general messages and posts or by popping in different message boards and offering feedback and advice. Some of us reached out to him on Twitter and he is very approachable and kind. That certainly helped and will, I’m sure, contribute to making the course a popular one. He initially stated, after Week 2, that he might pulled back a bit but he did not, as far as I could tell.

His video lectures were great and very much reinforced what is in his book. The quality of the Youtube videos was a bit BLEH but that is probably more a Youtube crappy compression format issue than anything else. These videos were pretty short (under 15 minutes), so I would watch a couple at a time at different times of day and week. They pretty much set the “agenda” and focus of the week’s work.

Alberto also provided optional Illustrator lectures. I have not watched them, but I have them saved for later, once I get over my Adobe products phobia (is there a name for that? A support group?).

What Worked

The course itself is more powerful than I ever imagined it could be. I had no idea when I started six weeks ago that I would get all excited about plunging into massive databases (Calc, FTW!) to extract information and that I would learn Tableau, and get all fancy in Piktochart. I still have some learning to do for Tableau and the dreaded Illustrator. Heck, I’m even considering going into R and Python (that’s for later). It would actually be great if Alberto Cairo were to create MOOCs for that too.

But this course really pushed me to just jump in and start working on my own visualizations, no matter how amateurish they look. I think many of us will leave the course with enough skills to start working our own stuff. Those of us who were not designers when we started are still not designers now but we get a little bit of the skills that can help us integrate some design into our regular work. We all learned a lot. A lot.

The atmosphere of the course, especially through the message boards, was relaxed but serious, professional but friendly. As I said above, there was no trolling. Everybody was focused so, I’m sure we all felt confident to share our work, knowing that it would be evaluated thoughtfully. We encouraged each other a lot while providing good critiques.

Six weeks was the perfect duration. Enough to get some depth into our work, but no burnout and no repetitive work. The course objectives were limited as well as the scope of the course. I think this course was perfectly sized in organization, substance and workload.

What Could Be Improved

I have always been ambivalent about using a lot discussion boards and peer review as pedagogical tools. The first week of the course showed that it can quickly get messy and counterproductive. so, sure, you can create a whole bunch of smaller message boards but then you have to manage them all. At the same time, it’s a MOOC, so, you won’t have everybody’s work evaluated and graded by an instructor, so peer review is the main option. It works, I think, when working on skills. It is more difficult when working on more “written” work.

Quite a few students have complained that they were receiving too many emails from the message boards and could not keep up (of course, they could have shut that down) but the point is it is tricky to keep up with a message board activity, even a smaller one.

So, what frequently happened is that people like me posted our work early and we discussed each other’s work. Then, we would wait for other people to start posting, responding mostly to the first ones to post. But then, our own work would sit there, largely ignored as the board postings multiplied and other students started talking to each other. As the week progressed, I would pull back and so, mostly did not respond to latest project postings. Actually, since the boards remained opened after the end of the week, in week 3 and 4, I still received emails from postings from the Week 2 board, but I could not be bothered as that point. I had moved on. In the end, I think a lot of people got their work remain unevaluated. The requirement was for us to evaluate at least two other projects. Maybe that requirement should be increased to 4 or 5.

Maybe there should have been some way to have some ability grouping. But then, I also really enjoyed seeing the work of more expert people so, I don’t know what the right balance is here. In the end, I think diversity should prevail. After all, even novices can offer good critiques from a strict user perspective.

Final Thoughts

I do hope Alberto Cairo offers an Infographics II, especially in interactive graphics. I am so nowhere near that level. I also hope there is a way for “alumni” to stay in touch (although there is a Facebook group but I wonder whether this is the right format). Maybe we could create our own Tumblr to share our dataviz work.

This MOOC, I think, defeats a lot of the negative stereotypes about the format: no one is going to take this MOOC and test out of anything. It is not going to steal students away from colleges and universities. But it certainly has imparted skills to people who wanted and needed them. The fact that the course was not populated by college students should also appease some fears. After all, this was more continuing education than anything else: a bunch a professionals from all over the world (who will not go study at the Knight Center anyway) who wanted a new educational experience who fit their busy schedules and who had access to technology. The course is not a substitute for a full-fledged curriculum on the subject. It is a bite-sized introduction, and a very good one at that, but a limited experience nonetheless.

I don’t even know if this course would work with college-age students. Although it was only for six weeks, this course required some commitment and persistence that not all younger students might have (especially with no credit at the end). I wonder what percentage of the 2,000 class completed the course.

Now, the million dollar question is whether this MOOC is way better than the average MOOC, and now I am completely spoiled and other MOOCs will feel lame and mediocre by comparison.

And I really, REALLY want Alberto Cairo to teach a MOOC on dataviz for social scientists.

But again, this has been a wonderful experience and I will miss it. One of my colleagues (a librarian) will be taking it next time around.

More Visualization of Social Mobility by Gender

One of my classmates suggested I unstack the bars in my previous visualization of international comparison of social mobility, by gender and class of origin. He was right, the pattern I noted becomes very clear once the bars were unstacked. Click on the images for their full sizes (or click here and here):

The bar chart on the left represents social mobility for sons of low-earning fathers. The green bars measure the percentage of these sons who end up in the bottom 40% (close to no upward mobility), the red bars represent the percentage of these sons who end up in the top 40% (upward mobility). As you can see, these sons are more likely to be stuck in the bottom 40% in the US and UK than in Scandinavian countries.

The bar char on the right represents social mobility for daughters of low-earning fathers. The blue bars measure the percentage of these daughters who end up in the bottom 40% (close to no upward mobility), the brown bars represent the percentage of these daughters who end up in the top 40% (upward mobility).

At first glance, it definitely looks like the daughters move up in larger proportion than the sons, with still less mobility in the US, but the UK looks a lot more like the Scandinavian pattern for daughters.

Now, if we compare those who end up in the bottom 40% (sons + daughters) and those who end up in the top 40% (sons + daughter), we get this (big image here and here), keeping the same color palette, for convenience:


The bar chart on the left represents the sons and daughters of low-earning fathers who end up in the bottom 40%, and the one on the right represents the sons and daughters of low-earning fathers who end up in the top 40%. Here again, you can clearly see the greater mobility in Scandinavian countries and especially for daughters, less likely to be stuck at the bottom 40%, more likely to move up to the top 40%, with less mobility in the US and UK.

As the scatterplots in my previous post showed, there is no simple and straightforward explanation for this. More likely, there are multiple variables at work that would require more elaborate statistical testing. But there is no doubt about the lower level of social mobility in the US, compared to Scandinavian countries. At this point, one can assert that the Horatio Alger narrative is a myth. I would taking it further and assert that it is a damaging myth because it presents false expectations, and does not allow for the very serious lack of social mobility to ever make it into the public discourse outside of individualistic, moralistic or racist narratives (those who don’t succeed do so because of their individual / moral failings or their different “values”… i.e. they’re lazy). The issues of structural obstacles to mobility in the context of increasing inequalities and the consolidation of class privilege never get discussed and therefore never enter the arena of public policy.

Literature Visualization

Visualizations are not just for social scientists with data on their hands. They can be used to analyze literature too. After all, most fiction writing is writing about social relationships. Take this example from a Shakespeare play (click on the image or here for the full thing):

You can clearly see where the big drama / plot points are, the messier entanglements between acts 2 and 3, and the resolutions in act 5 (compare where the lines start in act 1 and the different groupings at the end of act 5).

This is not the only representation one could envision. Relationships are of different nature: conflict, cooperation, competition, etc. That could be visualized too. One could also group characters into set based on common characteristics, such as family, religion, social class, etc. I really like the idea of mapping literature work that way. I might actually try once I get more technical know-how.

Exploring Social Mobility Through Visualizations

Ok, so, I am getting towards the end of my MOOC and for our last project, we were asked to come up with our own idea and topics and think about how to visualize them and basically have these visualizations tell a story. Since I had just received my brand new copy of the Economic Policy Institute‘s State of Working America (one of my favorite sources of socio-economic data on the American society from one of my favorite US think tank), I decided to dive in and see if I could find something that interested me and that I could explore more in details.

I found this (click here or on the image for a much larger view):

So, ok, no surprise here, Scandinavian countries have higher social mobility than the US, UK and a few other countries, with continental Europe and Oceanic countries somewhere in the middle. Then, I decided to look a bit more in details in the data to see if I could detect more precise information. I found this interesting thing (click here or on the image for larger view):

Hmm, a gender aspect to this. So, that is the angle I decided to pursue by correlating a series of gender-related variables (such as education spending on tertiary education or gender enrollment in tertiary education) to the mobility coefficient, with a series of scatterplots. I did it in Tableau, then transferred the results into a Piktochart infographic (definitely click here or on the image for a larger view):

As you can see, some variables worked better than others and produced stronger correlations. Social mobility is, of course, a complex and multi-layered issue, so, one would have to chase down more variables to correlate and maybe run a few regressions and other statistical tests, but I thought this was an intriguing gender pattern.

So, as far as software, I processed all the data and correlation coefficients through a Calc spreadsheet in OpenOffice. I constructed the scatterplots in Tableau. And I did some stuff with Simple Diagrams again.

For the scatterplots, I went looking for data at the World Bank and OECD websites where you can download databases on pretty much every topic under the sun.

I’ll have a general reflection post on this MOOC experience when it’s over next week.