Tag Archives: new york times

More bad reporting on texting and driving, and new data

The New York Times‘ problem of misrepresenting the relationship between phones and traffic fatalities, which seems to have begun with Matt Richtel, has just gotten worse.

Richtel sells books on the fear of texting and driving (which, of course, is dangerous), and the website for his book still — despite my repeated entreaties, public and private — leads with the bad, false, unsourced Internet meme, that “the texting-while-driving epidemic continues to claim 11 teen lives per day.” (As a reporter, how could you sleep one night with that BS up under your name? Mind boggling.)

Anyway, the new entrant is David Leohnardt. At the heavy risk of jeopardizing future opportunities to publish on the Times op-ed page, I tweeted that his recent column included “one of the dumbest things I’ve ever read in the NYTimes.” Washington Post WonkBlog writer Jeff Guo pointed out Leonardt’s column, which claimed that, with regard to the recent spike in traffic deaths, “The only plausible cause is the texting, calling, watching, and posting that people now do while operating a large piece of machinery.” The column contained not a piece of evidence to support that claim (though there were some awful anecdotes), which is why I said it was dumb.

Which is too bad. But even though the spike in traffic deaths is concerning, reporting should not be wrong.

Early estimates from the National Safety Council (which uses a different method than the Federal NHTSA) show a 6% increase in traffic fatalities for 2016. Leonhardt, working really hard to make that absolutely as alarming as possible, produced this graph, showing percent change in fatalities over successive two year periods going back to 1980:

C6aFWA5U4AEzK8I

Because it’s hard to add up the pluses and minuses in your head, It would be really easy — really really easy — to look at Leonhardt’s chart and think fatalities are higher now than they were in 1980. But rather than pointing out that fatalities per person have fallen by half since 1980, he instead writes, “It’s the first significant rise in a half century,” which would be true except for the significant rise in every single decade of the last half century.

This is a lot like when Richtel described the 2015 rise as, “soaring at a rate not seen in 50 years.” Not that the rate was not seen in 50 years, of course, just that the soaring of the rate hadn’t been (or so the NYT Science Desk told me when I complained).

Adding 6% to the NHTSA numbers for 2015, I get the follow graph, showing the trends in deaths per person in the population, and deaths per mile traveled, as changes since 1970. (The deaths per mile haven’t been released for the whole year yet; click to enlarge.)

PercentWhite

That is a troubling spike, which takes us all the way back to 2009 fatality rates. We should make the roads safer, by using them less and using them more safely. But come on, NYTimes.

Read the whole, completely aggravating series, under the texting tag.

2 Comments

Filed under In the news

Is the New York Times trapped in an economics echo chamber?

Ask a stupid question.

When Justin Wolfers wrote about the dominance of economists in the pages of the New York Times, he concluded, “our popularity reflects the discerning tastes of our audience in the marketplace of ideas.” I discussed the evidence for that in this post, which focused on the particular organizational features of the NYT. At the time it didn’t occur to me that his data — relying on uses of “economist” in the paper — would be corrupted by false attributions. So this is a small data story and a larger point.

The small data story comes from a personal reflection by Dionne Searcey, who wrote about work-family conflict in her new post as West Africa Bureau Chief for the NYT. It was a perfectly reasonable piece, except for one thing:

Much has been written about work-life balance, about women getting ahead in their careers and trying to have it all. I often find that if you scratch beneath the surface of many successful working moms, they have husbands who work from home or have flexible schedules and possibly a trust fund. Or in many cases, you find a mom who does more than her fair share at home — or at least feels as if she does. Economists have a name for it, “the second shift.”

Wait, “economists”? The Second Shift is a classic work of sociology by Arlie Hochschild and Anne Machung first published in 1989 and revised twice. Why “economists”? The (very good) article that Searcey linked to was called, “The Second Shift: Men Do More at Home, but Not as Much as They Think,” written by journalist Claire Cain Miller, focusing principally on the research of several sociologists, led by Jill Yavorsky (a sociology PhD candidate at Ohio State with whom I have collaborated). There are no economists cited or quoted in the story.

The small data story is that this mention of economists will go into Wolfers’ count of the influence of economists in the marketplace of ideas, but it’s a false positive — it’s the influence of sociologists being falsely attributed to economists.

But why would Searcey say “economists”? The answer lies in the organizational culture of the NYT. Here’s why.

Here are my two tweets on the piece:

Considerately, Searcey replied:

How odd. When I pointed out again that the story she linked to was about sociologists talking about the second shift, she didn’t reply.

I recently wrote that economists don’t cite sociologists’ work as much as sociologists cite economists even when the two groups are working on the same questions with obvious implications for both. What about the second shift? A JSTOR search reveals 473 cases of “second shift” and “housework” in journals identified as sociology by the database. The same search in the realm of economics produces just 35 mentions (no fewer than 6 of which were written by sociologists).

So, why did Searcey think she “was referring to how economists talk about the second shift”? My only explanation is that it’s because the piece was published in the NYT section The Upshot. As I wrote in my Contexts post, Upshot

is edited by David Leonhardt, who was an economics columnist before he was promoted to Washington bureau chief in 2011. That promotion was a dramatic move, elevating an economics writer who hadn’t been a Washington political reporter. Upshot is a “data journalism” hub, which often (but not always) implies an economic focus. (On the opinion pages, economist Paul Krugman writes a column twice a week, and Joseph Stiglitz moderated a long series on inequality.) This can’t be the whole story, but in broad strokes it’s fair to say the paper as an organization moved in the direction of business and economics.

Upshot is, of course, where Wolfers was writing in praise of the idea-market power of economists. Is this just the free market of ideas allowing the most persuasive to rise to the top? Searcey’s errors suggests that it is not. Rather, the organizational status of economics has corrupted her perceptions so that if something appears there she simply believes it reflects economics (and no editor notices).

Incidentally David Leonhardt (whom I’ve written about several times) has been promoted to Op-Ed page columnist and associate editorial page editor.

4 Comments

Filed under In the news

NYT magazine infographic: not just dumb and annoying

This graphic from the New York Times magazine is bad data presented poorly (and reproduced poorly, by my camera phone):

nytspank

It’s presented poorly because those blood stains are impossible to compare since you can’t discern their edges, and it appears they don’t taper toward the edges at the same rate. Maybe they simply resized one of them to get the relative size, which would be wrong. Anyway, if they cared about communicating the data they probably would have used real data in the first place. (You could also complain that a red speckle-cloud is unfriendly to some color-blind people.)

It’s bad data because it’s an online NYT reader survey, which — although it’s from the “research and analytics” department (and no, I’m not going to add “analytics” to my Windows dictionary) — represents unknown sample selection effects on an undefined population. In other words, who cares what they think?

A survey like that would be a start if it was the only way you had to answer an important or hard-to-measure issue, and if you clearly stated that it was likely unreliable. But in this case there is good, nationally-representative data on this very question. So if NYT Magazine wanted to inform its readers of something, they could have used this.

Here’s the good data — from the General Social Survey — in a graph that is at least a lot better: this is good data in a chart that’s easier to read accurately, includes a breakout by strength of opinion, and uses more accessible colors (click to enlarge).

gss spank 2014.xlsx

I think the NYT Magazine graphics violations are not just dumb and annoying — here’s another post all about them — I think they harm the public good. Graphics like this spread ignorance and contribute to the perception that statistics – especially graphic statistics – are just an arbitrary way of manipulating people rather than a set of tools for exploring data and attempting to answer real questions. (If you want awesome real graphics, check out Healy and Moody’s Annual Review of Sociology paper.)

P.S., I wrote more about spanking here.

Leave a comment

Filed under Uncategorized

That Sunday New York Times Style section trend piece

Just folks trying to survive their divorces.

Just folks trying to survive their divorces.

Does it matter which one?

You know it from the opening paragraphs:

The women are architects, film industry executives, skin care consultants, product managers at tech companies, psychologists. They have worked in finance, publishing and television, though some had scaled back or left the work force when their children were born.

Divorce is what they have in common. Their stories are varied: the breadwinner wife whose husband’s career hadn’t quite taken off and who found comfort in an affair; the husband who never really adapted to parenthood; the wife with Ivy League degrees who stayed home with her child but lost her way in the marriage while the husband thrived in his international career.

Really. Divorce is what they have in common? How hard would it be to include a single mention of how rich and privileged these women are compared to the typical woman getting divorced? Penelope Green’s story never mentions the possibility.

Here is what a five-minute effort would have looked like:

top10divoccs

These 10 occupations account for 25% of all women age 40+ who reported getting divorced in the previous year.

In addition, 34% of those just-divorced, 40+ women are not non-Hispanic Whites (14% Black, 13% Hispanic, 4% Asian/Pacific Islander).

1 Comment

Filed under In the news

Is it me or is the liberal media elite getting more tone-deaf?

In two acts.

Act I: David Brooks

For whatever reason, one of the first things I read today was this David Brooks column, “The Nature of Poverty.” Brooks is selling character these days, so his own brand of vapid exhortation on the subject has reached a fever pitch in his columns. But in this case what struck me was not the pedantic “we’ve done so much and they are still not pulling their own weight” thing (not a quote), but the extreme blindness on the issue of the protests — which he didn’t even mention (after this sociology 101 fail of an opening: “Lately it seems as though every few months there’s another urban riot” — which is a quote).

What I mean is, the Black (and other) residents of the city rising up in protest is staring him in the face as a response to decades of exploitation, marginalization, and repression, and he just wants to talk about the same old problem of lacking “responsibility, future-oriented thinking, and practical ambition.” Amazingly, in the very next sentence, he offers this juxtaposition: “a healthy neighborhood is like a ballet, a series of intricate interactions in which people are regulating each other and encouraging certain behaviors.”

Really, ballet? Could he not think of any of the other millions of examples of social life that fit exactly the description, “a series of intricate interactions in which people are regulating each other and encouraging certain behaviors”? That basically describes any society (even poor ones), or any collective activity. Maybe it never crossed his mind that juxtaposing the character-deprived ghetto with the beautiful ballet might have a slightly elitist connotation? I know ballet is not — and should not be — all White, but it’s pretty White, especially in the popular imagination. Here’s the first page Google image search for “ballet child”:

balletchild

Of course, social organization is a big issue, and concentrating the efforts of poor communities for successful social change is vitally important. Brooks sees the dysfunction, and the riot, but he doesn’t see the beautifully unfolding protest, the real story of the day.

Act II: Nina Totenberg

After that aggravation, imagine my delight when, as my New York Times / NPR morning unfolded, I found out Nina Totenberg was one of the panelists on the Diane Rehm show’s weekly news roundup. Always rational and erudite, a keen legal analyst, she may be the favorite voice on NPR. But today her cluelessness reminded me a lot of Brooks’.

Seven minutes into the show, they turned from the details of the Baltimore situation to the social context, and Totenberg launched into this description of the city. This is her entire comment:

If you go to Baltimore – and they’re our neighbors here, and we go all the time – it is the quintessential example of a city that once was a thriving manufacturing, also a port, city. And now when you go there, they’ve built up the Harbor, and that’s lovely, but you drive through that city – it’s not that big – and half of it looks boarded up, people are standing on the street corners looking – just hanging out, looking like they’re probably drunk, and with nothing to do. It is sort of – the scene when you drive through the city, if you’re going to Johns Hopkins for something, and you drive through that city, it is reminiscent of what townships looked like in apartheid South Africa. It’s not a pretty sight.

This struck me as the nadir of the elitist media model that the show represents so well. She is telling us about Baltimore, based on her expertise which stems, apparently, from visiting the Inner Harbor and driving through “half the city” on her way to Johns Hopkins University. (Her use of the second person in “you’re going to Johns Hopkins for something” brings to mind Rebecca Skloot’s description in The Immortal Life of Henrietta Lacks, in which the poor Black population of Baltimore historically had no contact with the elite White campus in its midst.)

Of course, unemployment is a serious problem in the poor parts of Baltimore — and their addicts can’t as easily conceal their condition as can those with more wealth and real estate. But it’s the old, scared-White-person-driving-through-the-ghetto thing that Rand Paul illustrated the other day with, “I came through the train on Baltimore last night. I’m glad the train didn’t stop.” You would think Totenberg would have watched The Wire at least enough to get the point that behind the unemployed people on street corners — inside the homes that you can’t drive through — there are men, women, and children doing all the other things people try to do everywhere else in the country, but with less money and more police brutality to contend with.

Of course she’s sympathetic about poverty. So I would have let it go, if she hadn’t followed it up with this. When host Susan Page mentioned the controversy over “calling them riots or rebellions,” and Obama using the word “thugs.” She asked, “Nina, how much difference does the language that we journalists use matter in a case like this?” To which Totenberg responded:

Well, I suppose it does. But I don’t see that people who are setting fires and breaking windows and looting stores, that you can call it a rebellion. That’s a really – that’s a stretch for me. You can have terrible wrongs, at the same time that you have a reaction that’s – inappropriate – let’s put it that way. I mean, I read a story that even some of the gang members [chuckle] were helping to calm things down.

First, I’m trying to think of an example from history of something that you would call a rebellion — in anything but a metaphorical sense — that would not include setting fires, breaking windows, and looting stores. To make matters worse, her followup gang comment, rather than illustrating a heightened level of community solidarity that has appeared in Baltimore, is somehow supposed to make the rioters look even more savage, because they’re even offending gang members.

Toward the end of the hour Susan Page gave Totenberg a chance to help herself out a little:

Page: We’ve gotten several emails, directed at you Nina, and a comment that you made earlier. I’ll just read one of them from W. Middleton, who’s writing us from Ohio, this person writes, “I am troubled by the poor choice of words describing citizens of Baltimore. To say that people are just standing around and looking drunk is insulting and in many ways captures a racist portrayal of communities of color.” What would you say, Nina?

Totenberg: Well, I s— didn’t have in mind any particular race. What I had in mind was people who obviously have no job, nothing to do, are standing on the corner, not fully functional, perhaps I should’ve said “not fully functional,” and that is the saddest thing in a, in a country that prides itself on the ability to – uh, as a working country.

Completely missing the point that it’s not necessarily offensive to describe a drunk person on the street corner as a drunk person on a street corner, what’s offensive is describing half the city as a drunk person on a street corner.

Epilogue

The authenticity craze sweeping young people is accentuated by social media, which delivers “real time” information, and smartphone cameras, which are perceived as unedited windows into reality. That these old media elites would say things offensive to poor Black people is of course not news. But what strikes me more and more is that the out-of-touchness of their offensiveness is growing increasingly apparent in contrast to the wave of apparently (if not actually) unmediated, authentic information all around them.

8 Comments

Filed under In the news

Research on teen crashes confirms that reporters selling books on phone risks hype phone risks

Using phones while driving is dangerous and should stop. But the focus on this issue distracts us from other dangers in driving (which have — you’d never believe from the news — declined rapidly in recent decades). And it distracts us from the broader danger of relying on motor vehicle transportation.

I dwell on this subject because it offers lessons beyond its substantive importance (see all the posts under the texting tag). Today’s lesson is about conflicts of interest in the news media.

The AAA Foundation for Traffic Safety published a study of about 1,700 moderate or severe car crashes in which people ages 16-19 were driving. To identify possible causes of the crashes, they used cameras and motion sensors in the cars, and analyzed the seconds before each crash. The headline result probably should have been that 79% of the crashes occurred when teens were driving too fast. But that’s apparently not news, so the AAAFTS and all the news media reporting the story focused on the fact that 59% of the crashes showed distraction as a likely cause.

The report website highlighted the data on distractions, and that’s reasonable. One of their findings is that distractions in their survey account for a greater proportion of accidents than are reported officially — something we’ve assumed but have had trouble establishing empirically. So that’s useful. They used this graphic:

TeenCrashInfographic

By this accounting, phones were involved in 12% of crashes, second only to interacting with passengers. But this is an artifact of the way the categories are binned. And a lot of smaller categories are left out of the figure, such as eating and drinking (2%), operating vehicle controls (3%), looking at another vehicle (4%), or smoking-related distractions (1%). (Note also that one crash can have multiple related distractions, but they don’t report the overlaps so you can’t do anything about them.)

So I redid the categories. I don’t see why eating and drinking should be a separate category from grooming, or why singing/dancing should be separate from adjusting the radio. So I made a new category called “physically doing something besides driving,” which includes eating or drinking, using an electronic device (besides phone), grooming, reaching for an object, smoking-related activity, operating vehicle controls, and singing/dancing to music. Also, for some reason their figure lists “looking at something outside the vehicle” but only includes “attending to unknown outside vehicle” in that category. I added two other types of distraction to that category, “attending to another vehicle” and “attending to person outside” — bumping up the outside distraction substantially.

Here’s my new version of their figure based on the same data (from table 13 in the report). It’s more comprehensive but uses fewer categories:

TeenCrashInfographic-adjusted

Now cellphones are fourth. So that’s a lesson about using arbitrary category collapsing and then ranking the categories. (This happens all the time with occupations, for example, where people say, “The top X occupations…” but the occupations reflect different levels of granularity.)

Anyway, back to cellphones

The New York Times reporter Matt Richtel won a Pulitzer Prize for his reporting on distracted driving. And he published a book — A Deadly Wandering — that tells the tragic story of a driver who killed someone while he was texting. Unfortunately, he is prone to hyping the problem of texting, which his audience is unfortunately prone to fixating on. I previously pointed out that, on the website promoting his book, his publisher uses an extremely wrong statistic, claiming that texting “continues to claim 11 teen lives per day.” He has mentioned this statistic (or its variant, that texting kills more teens than drunk driving) on Twitter, and also in media appearances. I pointed out that this number is more than all the teens killed in motor vehicle accidents, so it’s obviously baloney. I emailed Richtel about this, and he told me he would “get it fixed.” I emailed the publisher. I emailed the Times. I emailed the Diane Rehm show. No one changed anything. Cellphone crashes are like child abuse: people will believe any statistic about how bad it is and attack anyone who’s skeptical.

Of course I don’t want to minimize the problem of distracted driving, and there’s nothing wrong with telling people it’s dangerous. And it’s not my area of expertise. So I’ve only given the issue a few hours. But playing into a public hysteria about a very narrow, behaviorally-driven problem, rather than exposing the systemic problem that it reflects, is not good.

And now that Richtel is selling a book about texting, he’s got a conflict of interest — if he hypes the problem in his NYT stories, he makes more money. So here’s the NYT headline on his story:

nyt-richtel-phones-head

And this is his lead paragraph:

Memo to parents: Distracted driving by teenagers is riskier than previously thought, particularly when it comes to multitasking with a cellphone.

Again, it is true the report finds cellphone distraction causes more accidents than police reports have shown — so this is not irrelevant — though, of course, even with the new accounting they still cause orders of magnitude less than Richtel’s own promotional site claims. But mentioning phones in the headline sets the NYT apart from most of the coverage of this report:

  • Washington Post: “AAA: 58 percent of teens involved in traffic crashes are distracted”
  • ABC News: “Distractions a Problem for Teen Drivers, AAA Study Finds”
  • Houston Chronicle: “Distraction a factor in 6 in 10 teen driver crashes”
  • Chicago Tribune: “Distracted driving a key contributor to teen crashes, study shows”

On my first page of Google News searches, only the LA Times also mentioned phones: “Teen drivers distracted by cellphones, talking in most crashes.”

Who cares?

Some people who are tired of me complaining about this think you can’t have too much hype about safe driving, so who cares? But the distraction matters. The evidence that phones are a fundamental cause — a social cause — of accidents and deaths is very weak, although they are certainly the proximate cause in many cases. But we don’t have randomized controlled trials to test the effects of phones. I suspect the people crashing while futzing with their phones are mostly the same people who would be crashing for some other reason if cellphones didn’t exist.

When I look at the video compilation the AAA put out to accompany their report — which mostly shows teens crashing while using their phones — I am struck by what terrible drivers they are. They look down for three seconds and drive straight off the road without noticing. In contrast, I routinely see people driving on the freeway completely absorbed in their phones — driving obnoxiously slowly but using their peripheral vision to keep going straight. They are at grave risk of an accident if something crosses their path or traffic stops, but they’re not veering all over the road. Their slow speed probably mitigates their risk of crashing. I AM NOT RECOMMENDING THIS, I’m just saying: bad drivers cause accidents, and if you give them a phone they’ll use it to cause an accident.

Did you know teen driving fatalities have fallen by more than half in the last decade? (During that time incidentally, teen suicides have risen 45%.) Did you know that, from 1994 to 2011, mobile phone subscriptions increased more than 1200% while the number of traffic fatalities per mile driven fell 36% (and property-damage-only accidents per mile fell 31%)? Don’t count on Matt Richtel to tell you about this.

And yet, of course, thousands of people die in car accidents every year in the U.S. — at rates higher than the vast majority of other rich countries. But as long as people drive, there will be bad drivers. If we really cared, we would replace individual cars with mass transit (or self-driving cars) — putting transport in the hands of computers and professionals. Nothing’s perfect, of course, but preventing car accidents isn’t rocket science, and blaming a systemic problem on the individual behavior of predictably error-prone drivers doesn’t seem likely to help.

6 Comments

Filed under In the news