Author: justinmungal

Justin Mungal

Lately, I find that data visualizations are more eye-catching than attention-grabbing.  Complex color schemes on maps draw in the eye, but there is nothing to maintain attention for the purposes of learning something new, as the informative details of the data cannot be distilled from the image alone.  My favorite example of visualization eye-candy is the red and blue mapping of American states according to their political leaning. 

The image is attention grabbing, but the real learning takes place at the county level, and still, the true understanding of county level data needs to be coupled with county population levels for an accurate understanding of voter geography.  In essence, the graphic distorts the reality of voters’ geographical dispersion and therefore impedes my trust of the source and any other data they may present.

My slightly jaded attitude towards eye-catching images was unhinged by a recent attention-grabbing visualization of the world’s most popular brands by country.

The visualization depicts a smattering of the most valuable brands across the globe by country.  The color scheme represents brand strength on a scale of 0 – 100 and the size of the country corresponds with the valuation of the most valuable company for that country.

What I love about the visualization is that at first glance, my attention was grabbed and I immediately started learning.  We are so used to hearing about major U.S. brands, that few Americans, including myself, could claim meaningful knowledge of major non-U.S. brands.  This image provides a pallet for the eye of major global brands in relation to what the viewer is likely familiar with – American brands.  Thusly, the viewer is immediately immersed in a visual exploration of international brand valuation.  Note here, that I am presuming the viewer to be American since the chart was produced by an American news media station (MarketWatch).  By comparing international brands to familiar U.S. brands one realizes just how dominant the American market is – our companies dwarf those of other countries by fractions of dollar value.  The sizing of the countries instantaneously captures this major discrepancy in valuations, although given the size of the image as a whole, it is a bit difficult to fully appreciate the sizing imbalances.

What I most dislike about the graphic is its size.  For example, the size of the image not only makes the differences in country size difficult to distinguish, but even makes reading the company names labeled on smaller countries rather unintelligible.  Furthermore, the size of the image means that the authors were forced to create a cutoff for which countries would show up at all.  I presume that a country needed to have a company worth over $3 billion since the lowest valued company listed is $3.7 billion.  Accordingly, no African and few Latin American countries made the cut to be depicted in the visualization.  Personally, learning about non-US companies was what grabbed my attention, and so missing out on large chunks of the world was rather disappointing.  In this regard, the author seems to have forgotten the audience, Americans, and what would be most insightful to them, namely a thorough exposé of unknown global brands in juxtaposition to well-known U.S. brands.

A redeeming factor for the shortcomings of this visualization is its pairing with a listing of the world’s top ten most valuable companies. Interestingly, only two non-U.S. companies make it on to (the lower end of) this list.  Here, the giants of globalization are named and their dominance by valuation is most well pronounced.  Given that the learning point for me was regarding non-U.S. companies, I would have liked to see either an extended list (say top 20) and/or a list of highest valued non-U.S. companies.  Again, the authors should have been more sensitive to the fact that their audience is Americans who are well versed in local business but less so and more interested in foreign business affairs (as far as learning something new goes).  To that end, it would have been easy for the authors to have allowed the user to scroll down a longer list of companies or to even select a listing of companies by country to provide a deeper dive and visual confirmation of the suggestions of the mapping (that many global brands are small in comparison to top US brands and that many of those non-US top brands are largely unknown to the general American public) for an audience captivated by their first stunning mapping of global company valuations.

 

References:

http://www.marketwatch.com/story/the-most-valuable-brands-in-the-world-in-one-chart-2017-02-08

http://illinoisentertainer.com/wp-content/uploads/2012/10/Red-States-vs.-Blue-States.png

 

Justin Mungal

Let’s face it, climate change is not just a debate, but a fierce fight for survival.  Those who reject 97% of scientists’ claims that human-driven carbon emissions are causing significant and catastrophic global warming are not merely “on the fence,” but rather vehemently opposed to what they categorize as conspiracy theory.  Any evidence that has more than a zero percent chance of speculation is utterly rejected by the group.  It is time to move the debate from scientific claim to scientific fact, where no ounce of speculation can be contorted into an “alternative fact.”

The journalists at Futurism have created “Five Graphics to Start a Conversation About Climate Change.” The first set of images depict what the majority of people believe, and that is that combined atmospheric temperatures are exponentially rising at alarming rates.  

While, this should be convincing enough, I no longer find it useful as climate change deniers claim that it represents normal fluctuations and that the correlation of rising CO2 emissions with these temperatures does not imply causation. This shred of speculation between correlation and causation collapses the debate into a stalemate, and so I believe that we must move the discussion into the realm of absolutely non-contentious facts.

The second set of images simply shows the rising rates of C02 emissions due to cement, gas, oil, and coal and its corresponding partitioning into the atmosphere and oceans.

This graphic is powerful because there is absolutely zero speculation and thus no room for debate.  Correlation and the (high) likelihood of causation are left off the table, and what remains is an impressive graphical representation of the massive amounts of human generated CO2 and its absorption into land, air and water.

The third graphic simply explains the science of how CO2 in the ocean can acidify its waters.

Again, there is no speculation of causation, it is merely a statement of scientific fact around the chemistry of water and CO2.  Effectively, this builds upon the second graphic, to raise the question of what impact our CO2 emissions can have on earth’s oceans.  The question is answered in the third graphic by the description of a chemical reaction summed up in the equation Dissolved C02 + Water à Carbonic Acid.  Again, there is no ambiguity or speculation – just solid scientific fact.

The fourth graphic shows how long C02 stays in the atmosphere after its initial pulse.

Seventy percent of pulsed C02 remains after one hundred years and forty percent remains after one thousand years.  Again, the graphic is not intended to show speculative correlation or causation, but rather clearly and accurately represents the fact of C02’s lifetime in the atmosphere.  Essentially, once CO2 is pulsed, it stays in the atmosphere for a very long time.  Coupled with graphics two and three, we see a picture of the massive amounts of C02 being emitted threatening the pH balance of our oceans for a long time.

The fifth graphic depicts projections of what earth’s climate might look like in the years 2081-2100.

https://futurism.com/wp-content/uploads/2017/05/5-climate-future-ipcc.jpg

The projections depict a boiling world of monsoon rains and ocean covered lands.  I could imagine that such a world would be barely, if at all, inhabitable by humans.  However, even I as a climate change believer, find this image difficult to digest.  I know that scientists find it difficult to predict the weather for a given week, so how accurate can hundred year projections be?  I do not deny that such predictions are possible considering our current global warming trends, but forecasting that far into the future seems spurious and, given the rash skepticism of the opposing side, non-provocative.

For the majority of us who believe in the work of climate scientists, graphics one and five are panic inducing.  We believe that the rising global temperatures are abnormal and humanly caused.  Moreover, the idea that the earth may not be humanly inhabitable for long if we continue on our current trajectory is not farfetched.  These two images capture that scientific consensus frighteningly well.  However, the purpose of this article, and I believe the necessary debate, is with those who reject such claims for the improbable claim that the overwhelming evidence of correlation does not imply causation.  That is why I am far fonder of graphics two, three, and four in terms of moving the debate forward.  They clearly and simply depict measurements and chemical reactions.  They cannot be negated as they do not contain any speculation over causation, are repeatable, and provable.  Thus, they can be starting points for serious debate over the current state of climate affairs and the ominous threats they imply.

What I would do to improve the graphics is include more visualizations like graphics two, three, and four.  Those engaged in the fight of their lives – the debate of climate change – need more ammo that cannot be misconstrued and discarded as circumstantial.  To spend more time creating images like graphics one and five are futile, as we know that the counterargument is to negate them as speculation and propaganda.    Unfortunately, we do not have time for visualizations whose validity is debatable – time is of the essence.

References:

<https://futurism.com/five-graphics-start-conversation-climate-change/>

Justin Mungal

Tax is a powerful tool for implementing effective public policy.  Few legislative mandates share its efficacy in shaping, seemingly overnight, corporate behavior.  Inextricably, it is tied to the notion of the common good insofar as it pools society’s financial resources for funding that vision of social welfare and human well-being.  Aside from technological innovation, it stands as one of the greatest formators of our modern economy.  For that reason, there is large vested interest in shaping tax code and many a think-tank has arisen around the D.C. metropolitan in order to have a voice at that table of national discussion.

The Tax Foundation recently released its 2017 State Business Tax Climate Index.  Their visualization shows a map of the fifty United States of America color coded as blue for the ten worst business tax climates, orange for the ten best business tax climates, and grey otherwise.  Also, the individual rankings (1-50) are printed in white on each individual state.  The visualization’s goal appears to be to create a KPI based on the results of their study in which they rank states according to 100 variables grouped into the five categories of: corporate taxes, individual income taxes, sales taxes, unemployment insurance taxes, and property taxes.  The stated goal of the KPI is to enable tax policy makers to compare their state’s tax system to other American states.  The rationale for comparing state tax systems is that most business decisions to move based on tax incentives are intrastate decisions rather than international ones.  Thus, the ability to retain business stakeholders is based on the relative favorability of one’s state tax structure to another state’s.  Furthermore, by ranking every single state according to 100 variables, states can build themselves a roadmap to improvement based on the differences of tax structure in higher ranking states.

While the visualization is the poster child of the Tax Foundation’s report, I find it exceptionally uninformative.  Directly below the visualization they have printed the numerical rankings of each state.  This tabular representation of the same data is much more straight forward and easier to digest.  For example if I were a tax policy maker from North Dakota, ranked #29, I would have difficulty finding the next best state (i.e. #28) from whom I could learn how to improve my state tax structure.  Indeed, one must scour the map until finally locating #28 on the state of Mississippi.  Contrast this to the tabular data with its column of overall ranking, where the next best state is easily spotted (loading the table into Excel and filtering the data according to overall rank would make it even easier).  Indeed, I find no benefit to studying the visualization over the table, as the mapping of the data essentially scatters the physical location of data whereas the table organizes it.  The only benefit of the mapping is that it adds eye-catching color to the eighty-page report.

What I find most disappointing about the Tax Foundation’s visualization is that the report itself is very well done and informative.  However, skimming the internet for similar visualizations, I find the Pew Foundation’s:and Wallet Hub’s: maps of state tax data.

 

The Pew Trust Foundation’s map has more interesting bins by which states are colored and Wallet Hub’s visualization delivers a heatmap; both maps working interactively to show the individual state ranking when the cursor is place over the state.  While the Pew and Wall Hub reports cover different domains of data, they point out that a unique perspective on visualizing state tax data rankings is possible.  Comparatively, the Tax Foundation’s visualization falls short as it does not offer any new perspective on the table of data immediately below but rather obfuscates those same results for the purpose of soliciting “eye candy.”

Given that the Tax Foundation’s report is high quality, I believe there is room for optimistic hope that their visualization can be improved.  Moreover, I personally think that the table of results below the visualization is well organized and sufficiently summarizes their findings.  That said, I would add on top of that data another data set that would make the visualization illuminating.  For example, the map could be color coded to indicate the hottest states to which businesses relocated to due to tax incentives, with the original report rankings either being numerically printed as they have now or interactively projected as in the Pew map.  This layering of data in the visualization would build upon the table and create a convincing argument as to why a state may want to change their tax code and which state’s tax code they should be modeling theirs after.  Such a visualization would give state tax policy makers a clearer roadmap to economic success.

Resources:

<https://taxfoundation.org/2017-state-business-tax-climate-index-released-today/>

http://www.pewtrusts.org/en/multimedia/data-visualizations/2014/fiscal-50#ind0 and

https://wallethub.com/edu/best-worst-states-to-be-a-taxpayer/2416/

Justin Mungal

http://www.nytimes.com/elections/forecast/president

I remember the night of the 2016 presidential election hauntingly well.  Several friends of mine and I gathered at The Hut in honor of its historic closing and in celebration of Hillary Clinton’s imminent win.  Beers in one hand and phones in the other, we starred with our eyes glued to the bouncing needle on the NY Times live presidential forecast.  We could hardly steal a second to sip our beers because the needle on the dashboard kept bouncing left and right, edging towards Donald then retreating to Hilary, back and forth on “Chance of Winning the Presidency” scale, keeping us in endless suspense.   Little did I know that the previous polls indicating Hillary’s soon-to-be landslide victory were as false as the wavering needle of NY Times exit poll data.

Looking back on that night I realize I was being duped and deprived of enjoying my frosty brew, instead stuck to an incessantly moving  and meaningless needle.  Inspecting the data graphic now, I have no idea what I was even looking at that night of the election.  The categories the needle teetered between (very likely, likely, leaning, and tossup) are inherently vague.  The needles to the right depicting popular vote margin and electoral votes made sense – they showed numerical data, the needles moved according to incoming county data, and visually account for each incoming vote and their actual impact on the election results.  So what then does the category “tossup” mean, how many votes does it take to move from “likely” to “unlikely”, and on what basis does the needle move within a given category?  Later in an NPR podcast I would find out that the NY Times does not receive a steady stream of election polls data, which the continuously moving needle would indicate, but rather receives chunks of data from counties at various times throughout the night as their ballot casting centers close and compile the exit poll data.  The endlessly wavering needle was false and was programmed to keep viewers in suspense and glued to their app.  The true data, the two needles to the right (“Popular Vote Margin” and “Electoral Votes”), moved infrequently and did not require constant attention – offering an opportunity to close the app and put our phones down.

In the dawning era of “fake news” and “alternative facts” the American public needs, now more than ever, media stalwarts of integrity.  The New York Times has been known to be one of America’s most prestigious sources of news and must fight to maintain this privileged seat of honor.  Ironically, we are also entering an era of data driven news.  We have the capacity to collect, analyze, and visualize vast quantities of data quickly, informatively, and entertainingly.  In a world where each news corporation has its own spin on every story, data must be reverenced for its undeniable link to truth.  For the NY Times to report false and sensationalized data for the purposes of maintaining viewership is reprehensible and unfairly places the burden of discerning the integrity of data on ordinary citizens rather than on highly trained data scientists.  Given the dilution of truth in our modern Twitter chatter and Facebook news feeds of lies it is necessary that the historic bastions of reporting integrity recommit themselves to the noble profession of raising the bar for what it means to be an “informed citizen.”  In the advent of big data and visualization, the possibility of raising that bar to new heights previously unimaginable has become a reality.  We must not turn that reality into a fantasy by creating false data and misleading visualizations.  The consequences of such behavior – a president with ties to Russia, the disintegration of universal healthcare, the launching of fifty-nine cruise missiles on Syria, etc. – are indeed grave.