# Tag: statistics

## Visualizing Numbers Effectively

Most people have trouble with numbers. They easily visualize up to twelve. Once beyond 100 numbers kind of blur together. There is a difference between a one in 500 chance of something happening and a one in a million chance, but most people do not really understand it. Or the difference between a million and a billion.

“Making Numbers Count: The Art and Science of Communicating Numbers,” by Chip Heath and Karla Starr, offers a solution to that problem. It presents tools to understand numbers and effectively communicate the meaning of numbers to others. The authors provide a step-by-step process to give readers mastery of numbers using a few simple rules.

Take the difference between a million and a billion. Heath and Starr have readers visualize the difference this way: a million seconds is twelve days; a billion seconds is 32 years.  You suddenly appreciate the scale of the difference. Twelve days is two workweeks linked by a weekend. Thirty-two years? Depending on your age, it could be twice your lifespan, your lifespan, or half your lifespan. Regardless of your yardstick, you know it is a whole lot more than two workweeks linked by a weekend.

## Zero Risk is Never, Ever Possible

Over the last many weeks, many of you know that I have been going through chemotherapy. (You can read one of my posts here.) When I learned that chemotherapy was recommended after having a single mastectomy, I was stunned; we’d detected the tumors early, they’d removed the breast, and the cancer had not infiltrated my lymph nodes. But according to one of the latest, most sophisticated tests (oncotype), I was at high risk of recurrence of breast cancer at my age. The oncologist explained that the “ideal” number of rounds of chemo would be six (every three weeks), but four would be “acceptable.”

At this point, I have completed three rounds. I know that there are many women who find the treatment very debilitating, and although I was miserable and unwell, it could have been worse. But my husband and I began to rethink the decision to go more than four rounds. What was the source of our hesitation?

## The World is Getting Better – Honest!

Is the world getting worse or better?  Given the constant barrage of bad news, it is easy to think things are going from bad to worse.  You would be wrong, though.

“Ten Global Trends Every Smart Person Should Know (And Many Others You Will Find Interesting),” by Ronald Bailey and Marian L. Tupy, explains why. They show, using objective data, the different ways in which the world is improving.

They wrote the book because “You can’t fix what’s wrong in the world if you don’t know what is actually happening.” Using straightforward data and graphs they demonstrate why and how the world has improved, especially over the last 72 years.

## Member Post

One of the more discouraging things about modern life is the tendency of the media to take every little thing to DEFCON 1* without investigation or thought. That is certainly true of things connected to politics, but the tendency has flowed far beyond that realm. Take for example the revelation that the Houston Astros had […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## American Architectural Geography: Part 1, Timing

Last winter, Ricochet’s own @thelostdutchman published a great series about Pennsylvania political geography. Being something of a geography geek myself (the map-loving kind, not the critical-theory-spouting kind), I thought I’d try my hand at writing something a tad less detailed and a tad more ambitious — a brief description of American architectural geography.

Finding data which says something meaningful about architecture is not an easy task, perhaps because architecture is an art, and art isn’t quantifiable. But, still, the statistical gods have smiled upon us Americans. In 1940, the Census Bureau decided, for the first time, to ask detailed questions about American housing. As the libertarians winced, homeowners and renters filled out a questionnaire inquiring about such subjects as property values, housing size, mechanical systems (like heating, plumbing, and electricity), and, best of all, housing age. The data is aggregated by county and city (and farm and non-farm), and it’s organized, roughly, by decade — with a category for houses built before 1860, one for houses built in the 1860s and 1870s, one for houses built in the 1880s, and so on. This means that the interested obsessive (like me) can gain some understanding of any one county’s architectural chronology. Is the data accurate? Not entirely. Self-reported data is seldom accurate. But it’s accurate enough to show trends. I’ve done plenty of spot-checking, and the data usually aligns with what I’ve observed. The picture it paints is a meaningful one.

## The Perils and Pleasures of Modeling

The term ‘model’ is much in the news, and I’m not talking about @RightAngles trade. It’s the term apparently favored by the media to describe a general area that may also go by: cybernetics, system dynamics, advanced statistics, simulation, control theory, and others. Having some academic and professional background in the domain, this is my (inevitably simplified) attempt to sketch its limits, so you can be smarter than the average journalist.

So, simplifying, as warned: There are two types of models. One is broadly statistical in approach. The other attempts to be more mechanistic.

## Member Post

Democrats have long been warm to the idea of eliminating the Electoral College simply due to their preference for centralization of power over granting power to the individual states. The outcomes of the 2000 and 2016 elections, wherein a GOP candidate won the Electoral College, and, accordingly, the election, despite running a deficit to the […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## Member Post

Like many around here, I was surprised at Trump’s victory. Sure, he didn’t win the popular vote, but he brought out enough disaffected rural whites to win in the states that mattered. This is his base now, and I expect that they want him to deliver some goodies. In delivering these goodies, I fear that […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## Member Post

From Harry Enten at Five-Thirty-Eight: There’s a belief, which I don’t share, that the growing share of nonwhite voters in the population, particularly Latinos, is giving Democrats an enduring advantage in winning elections. The theory — known to some as the “Emerging Democratic Majority” — works only if voting patterns stay the same and Republicans […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## Member Post

As part of another project, I’ve compiled county-level socio-economic data for the American West (the 12 mainland states on or westward of the Rocky Mountain line). This includes things like employment levels, per capita income sources, ethnic mix and religious affiliation, derived from the US Census, Bureau of Economic Affairs, and the Association of Religion […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## Guns Don’t Kill Children … Swimming Pools and Cars Do.

This piece from Reason is a good primer on the lack of a market for “smart-guns,” and covers both the technical challenges in making them and — more interestingly — the lack of demand for them. Is this because gun owners are callous, child-hating fanatics? No: it’s just that firearms don’t kill that many kids.

Inspired by the piece, I took a gander through some of the CDC data for fatal injuries to children between the ages of 0 and 14 years in the United States between 2004 and 2010 (the most recent period listed). Here are some relevant data for the an average year during that period:

• 6,327 children were killed through injury (all causes, both intentional and non-intentional).
• 1,890 were killed through unintentional cars accidents  (30 percent of total).
• 749 were killed by unintentional drowning  (12 percent of total).
• 45 were killed by unintentional use of firearms (less than 1 percent of total).
• 378 were killed by all uses of firearms (6 percent of total). This would include all child suicides and homicides, as well as accidents.

(It should go without saying — though I’ll say it regardless — that every one of those deaths is a tragedy and that I can only imagine what the parents must be going through.)

## Lies, Damned Lies and the Washington Post’s Omitted Statistics

To its significant credit, the Washington Post has devoted much time and energy over the last year to assembling a database of fatal police shootings. By their tally, some 998 Americans were shot to death by police under all variety of circumstances in 2015. That is double the previous high total reported by the FBI, a fact that unveils an unquestionable gap in government statistics management. It is somewhat remarkable that no government entity accurately tracks this data. However, inasmuch as such statistics come partnered with Disraeli’s lies and damned lies, the reluctance of law enforcement to provide unethical activists with a tool chest of numbers to twist is not unsurprising.

And, as if on cue, the Post has proven that fear well founded. A tool that could have shed light on (arguably) the most crucial aspect of the relationship between government and governed was instead (though not unexpectedly) obfuscated and sullied the conversation with misleading spin and blatant omission.

When it comes to judging police use of force, the most important factor is it’s reasonableness: that is, the context of the use of force and the perceptions of all involved. Was the suspect armed or did he appear to be armed? How far away was he? Did the officer give the suspect a chance to comply? Was that even possible? Were there other options available? Even with nearly a thousand lethal police shootings last year, the number of shootings (lethal or otherwise) by officers is a miniscule fraction of all encounters police have with citizens. Thus, these factors are crucial to understanding what sets a given use-of-force encounter apart from all the others.

## Member Post

From a very compelling article about the statistics of gun ownership, homicide, and other crimes in the United States, which has way too much good stuff in it to repeat here, one tidbit stuck out for me: The United States’ homicide rate of 3.8 is clearly higher than that of eg France (1.0), Germany (0.8), […]

This is a members-only post on Ricochet's Member Feed. Want to read it? Join Ricochet’s community of conservatives and be part of the conversation. Join Ricochet for Free.

## How to Lie with Statistics

“F.B.I. Treating Attack in San Bernardino as Terrorism” reads the New York Times headline, implying that perhaps the Times demurs. The sense of the paper being dragged, reluctantly, from a preferred narrative is accentuated midway through the article by a curious graphical island, appearing in splendid isolation from the actual text: