Our current President likes to give individuals he dislikes nicknames: “Rocket Man,” “Liddle,” “Pocahantas,” etc. I’d like to suggest one for him, one that considers where his militaristic motivation may come from. Anyone who, like Senator McCain, truly experienced war would not have the same aggressive stance as Donald Trump. So I think that Mr. Trump should be known as “President Bone Spurs,” in recognition of how he personally dealt with his possible involvement with the military. Other suggestions?
What statisticians call Type 1 errors (incorrectly rejecting the null hypothesis) and Type 2 errors (incorrectly accepting the null hypothesis) initially arose from signal detection theory: is that blip on the radar screen a signal or just noise? The two errors were known to us engineers (my former life) as either a false alarm or a missed detection.
But these are not the only statistical errors that can occur. Andrew Gelman proposed two additional statistical errors, Type S (confidently stating that a value is positive when it is negative, or vice versa) and Type M (confidently stating that a value is small in magnitude when it is large, or vice versa). They have less to do with the actual statistics than with interpretation of those statistics.
In furtherance of Gelman’s extension of statistical errors, I’d like to propose a new one, the Type K error. This is in recognition of the attempt by Kris Kobach (Kansas Secretary of State and vice chair of a federal voter fraud commission) to deny the vote to (at least) tens of thousands of US citizens in order to prevent the two or three improper votes (out of millions cast) from occurring. [My numbers may be off, but you get my meaning.]
There have been other manifestations of this “error” in recent days. A report detailing the economic consequences of admitting refugees did not include the overwhelming financial benefits they provide over the long haul. In other words, the Type K error might be defined as “the deliberate and wrongful act associated with a statistical evaluation of the effect of only one side of a policy.”
Although the media have been getting better in dealing with numbers, the statistician in me sometimes recoils at such reports. An example is in today’s NY Times op-ed piece on Vietnam, in which the author writes that now is “an appropriate time to honor the suffering and the sacrifice of all those who served, including the 58,000 American service members, the estimated 1.3 million North and South Vietnamese fighters and the two million civilians who were killed during the conflict.”
I’m sure that it follows strict Times guidelines as to how to report numbers, but it gives one a visual impression that is akin to the quote apocryphally attributed to Stalin: “the death of one person is a tragedy; the death of one million is a statistic.” I would think that a better way to report this would be to write of “the 58,000 American service members, the estimated 1,300,000 North and South Vietnamese fighters and the 2,000,000 civilians who were killed during the conflict.” It provides a more accurate indication of the extent of the tragic consequences of the Vietnam war.
There has been a host of recent articles and books decrying the use of “big data” to make decisions about individual behaviors. This is true in commerce (Amazon, Facebook, etc.), but also true in criminal justice, my field of research. Moreover, some of the algorithms that forecast dangerousness are proprietary, making it all but impossible to determine the basis for challenging a sentence based on the algorithm’s outcome. Recent books, such as Weapons of Math Destruction and The Rise of Big Data Policing, underscore the dangers of such activity. This is the essence of an autopilot approach to forecasting behavior – hands off the wheel, leave the driving to us.
There is some research that supports this type of algorithmic decision-making. In particular, Paul Meehl, in Clinical versus Statistical Prediction, showed that, overall, clinicians were not as good as statistical methods in forecasting failure on parole, as well as the efficacy of various mental health treatments. True, this book was written over fifty years ago, but it seems to have stood the test of time.
It is dangerous, however, to relegate to the algorithm the last word, which all too many decision-makers are wont to do (and against which Meehl cautioned). All too often the algorithms, often based on so-so (i.e., same-old, same-old) variables – age, race, sex, income, prior record – are used to “predict” future conduct, ignoring other variables that may be more meaningful on the individual level. And the algorithms may not be sufficiently sensitive to real differences: two people may have the same score even though one person may have started out doing violent crime and then moved on to petty theft, while the other may have started out with petty crime and graduated to violent crime.
That is, the fact that a person has a high recidivist score based on the so-so variables should be seen as a threshold issue, a potential barrier to stopping criminal activity. It should be followed by a more nuanced look at the individual’s additional life experiences (which do not fit into simple categories, and therefore cannot be included as “variables” in the algorithms). That is, everyone has an age and a race, etc., but not everyone was abused as a child, was born in another country, or spent their teen years shuffling through foster homes. Therefore, these factors (and as important, the timing and sequence of these factors) are not part of the algorithm but may be as determinative of future behavior as the aforementioned variables. This is the essence of a power steering approach to forecasting behavior – you crunch the data, but I decide how to use it and where to go.
Regarding power steering, I’m sure that many of you would rather look at an animated map of weather heading your way than to base your decisions (umbrella or not?) on a static (autopilot) weather forecast (BTW, does a 30 percent chance of rain refer to the likelihood of my getting wet in a given time period or to the fact that 30% of the area will be rainy and may skip me entirely?). The same issues are there in crime analysis. A few years ago I coauthored a book on crime mapping, which introduced the term that heads this post. In that book we described the benefit of giving the crime analyst the steering wheel, to guide the analysis based on his/her knowledge of the unique time and space characteristics of the areas in which the crime patterns developed.
In summary, there’s nothing wrong with using big data to assist with decision-making. The mistake comes in when using such data to forecast individual behavior, to the exclusion of information that is not amenable to data-crunching because it is highly individualistic – and may be as important in assessing behavior than the aforementioned variables.
Now that the Senate-blessed Neil Gorsuch has donned the Robes of the Righteous, we have to reconsider the way we live our lives. Not very religious myself, I am now thinking that I should be more active in this area.
So here’s how one might fight fire with fire (and a touch of brimstone). As far as I know, there are no hard-and-fast criteria for defining a religion. For example, Scientology was started by the science fiction writer L. Ron Hubbard (who knows, maybe it started out as a joke?), and has been able to gull some of Hollywood’s dimmest stars into joining it.
So what might it take to become a religion recognized by the Supreme Court? My brilliant idea: turn tennis into a religion. Let’s call it Tênis, using the Portuguese spelling to make it more exotic.
Here are some of my p-baked thoughts (hopefully, p > 0.5):
- Shrines: we have Forest Hills, Wimbledon, Roland-Garros, and Melbourne Park, to which we can make our pilgrimages.
- Matriarchs and patriarchs: the Grand Slam tournament winners would surely qualify.
- Club fees: since you are contributing to a religious endeavor, you should be able to include your fees as a charitable deduction on your income taxes.
- Government grants: as per the Trinity Lutheran decision, if the court surfaces need to be redone, a government grant is not out of the question.
- Rackets (an obvious double entendre) and balls: they could be purchased tax-free.
I don’t mean to imply that all religions are as shallow as the one I’m suggesting; it’s just that if we are going to remove the barrier between church and state, as Gorsuch, Alito, and Thomas seem to want to do, we should consider how to leverage it to our advantage, or at least to point out the inconsistencies in their arguments.
PS: I originally entitled this post “What Does It Take to Start a Religion?“ but felt that “Found” would be more …. profound.
PPS: I considered focusing on golf instead of tennis, but thought that it might give someone in high office an idea.
PPPS: In adding comments to this post, be thoughtful. After all, this could be the founding Testament for a new religion. I don’t want it to include a shopping list, as in “A Canticle for Leibowitz.”
“The reason this is such a big deal has — we have this big, messy, wonderful country where we fight with each other all the time, but nobody tells us what to think, what to fight about, what to vote for, except other Americans, and that’s wonderful and often painful.
“But we’re talking about a foreign government that, using technical intrusion, lots of other methods, tried to shape the way we think, we vote, we act. That is a big deal. And people need to recognize it.
“It’s not about Republicans or Democrats. They’re coming after America, which I hope we all love equally. They want to undermine our credibility in the face of the world. They think that this great experiment of ours is a threat to them, and so they’re going to try to run it down and dirty it up as much as possible.
“That’s what this is about. And they will be back, because we remain — as difficult as we can be with each other, we remain that shining city on the hill, and they don’t like it.”
–James Comey, 8 June 2017
We have a few years in which to find a presidential candidate. I’ve found mine, if he can be convinced to run. On either party’s ticket, or on a new one, as Emmanuel Macron has done in France.
This is a technical issue more than a policy issue, but it has some policy implications, in keeping with this website’s slogan. It is prompted by so many recent polls that purport to show that Republicans seem to be backing Trump with no diminution in fervor. Many of these polls are cited by Charles Blow in his op-ed pieces in the New York Times. And today’s Washington Post-ABC poll shows a similar split between Republicans (67% support exiting the climate agreement) and Independents (22%) and Democrats (8%).
My question is, do those statistics reflect no change in Trump support, or is something else going on as well? In particular, is the percentage of people who self-identify as Republicans going down, leaving only the most fervent supporters in those retaining allegiance to the party?
I have a similar question about the data from the Energy Information Administration (EIA) posted by James Wimberly. In the EIA example, the percent of energy use is depicted over time, showing, for example, that hydropower has gone from generating 30 percent of total US electricity in 1950 to six percent in 2016.
Now it may be that some dams were decommissioned over the past sixty-odd years. I would guess, however, that it’s more likely that the total amount of hydropower-generated electricity has not dropped very much, but that its share has dropped because we are using more electricity overall. A better graphic would be one that shows the actual output of each source instead of its share of the total. [And an even better graphic would be one that stacked the contributions one on top of the other, adding up to the total energy usage, which would show the actual contributions rather than the relative contributions.]
Consider where we’d be if Hillary Clinton had won the presidency. Benghazi would be resurrected; the email scandal would have been the subject of at least two congressional investigations; any progress in terms of the policies she and the Democratic Party had espoused would not only have been ignored, but would have been scathingly addressed – and the Donald would have been shouting “Fraud!” from the hilltops.
True, we’re in a parlous situation with our current administration, but look at what has been taking place throughout the country. If anything, the republic is in better shape for having this cartoon character “running” the country. The Republican Party is in a real quandary, with essentially every one of its priorities (the wall, immigration, health care, Social Security, tax “reform”) unable to get any traction. With a Clinton administration in power, they would probably have been able to pass their legislative agenda, but it would have been subject to veto after veto, hardly endearing Clinton to the country. As it now stands, we will have to suffer through a crazy time, at least until November 2018, at which time (from my lips …) the Democrats will take back at least the Senate, and Trump will throw in the towel.
During most of the recent “debates” about (you name it) health care, Russia, size of crowds, vote counts – one side says one thing and the other side denies or refutes or obfuscates. Then it just boils down to a pissing contest whose takeaway is, for most people, “a pox on both your houses.” And it gets filed away in most minds as the same-old same-old political infighting, forgotten after an hour or so.
But what if one side says to the other, “You just said X; I said Y. not only do I believe that Y is correct and X is wrong, but I’m willing to back up my belief with money. I will pledge $Z to your favorite charity if I’m proved wrong; are you willing to pledge the same amount to my favorite charity if I’m right?”
Not only does this call the liar’s bluff and bluster, it also increases the length of time that the supposed controversy is in front of the public. “Why isn’t Congressperson PR (for example) willing to put his money where his mouth is?”
Today’s New York Times has an op-ed piece extolling some of the virtues of the Republican plan for health insurance; one take-away from it (featured by the NYT) is that “5 percent of Americans generate more than 50 percent of health care expenses.”
So what? Before I retired in 2002, my medical expenses were minimal. Since then, however, I have had a number of medical problems. In other words, the smug feeling I used to have about others who populated the health care system has given way to the reality of (what I should have known, as a statistically savvy person) the difference between cross-sectional and longitudinal analyses. Cross-sectionally, 50 percent is pretty scary, unless you realize that that 50 percent is primarily populated by the likes (and age) of me. Longitudinally, however, the data may show a different story, with perhaps 10 percent of the population never having major problems throughout their life, and have paid (as insurance should) for the difficulties that they luckily never experienced.
The author of the op-ed noted that his 93-year-old father “just received a $50,000 catheter-inserted aortic valve, which was covered by Medicare.” Is he suggesting that his father should have just sucked it up and lived in pain or in a wheelchair for the next few years of his life? Doesn’t he realize that Medicare is just what he recommends, that his father and those like him are using Medicare to “save their own money for just this sort of rainy day,” with the proviso that we may not all need that umbrella? Insurance, whether for cars or homes or health, is meant to spread the risk.