In Edward Tufte’s books, he stresses that one should never present a number in isolation. The question that one should answer, as a writer/presenter, is “compared to what?” For the number of jobs in a country, you’d think that the relevant comparison would be to population (ideally working age population) and population growth. If Chile were to add 1 million jobs, that would be a very different experience for the population than if China were to do the same.
I just did a Google News search to see how many reporters compared the latest jobs report from the U.S. Labor Department to the U.S. population growth rate. The answer was “none”. This New York times story was typical. It reported that 20,000 jobs were lost in January, but did not say anything about how the number of Americans had changed. In less than one minute the reporter could have discovered that the U.S. population is growing at 0.98 percent (source). The population clock says that there are 308.6 million people living in the U.S. Without leaving Google, one can calculate that 252,000 people were added to the U.S. population in January. So we have 20,000 fewer workers and 252,000 more people. To me that is a much more interesting story than simply “we have 20,000 fewer workers.”
Given that Tufte’s books are perennial bestsellers, why the reluctance of journalists to present any kind of context or comparison?
Reporters are supposed to be objective. I’m not saying that they necessarily are.
Providing context or comparision risks losing this facade. If I compare the number of jobs lost to the number of businesses founded this year in the US, I come up with a different picture than your population comparision. Which of those comparisions represents the “objective” truth? Perhaps it’s better to just report the number and let readers make their own conclusions.
— CHS
I so agree about reporters’ lack of sensible analysis regarding numbers,
and readers’ apparent disinterest (or ignorance).
However, in this case, is population growth really the right comparison?
The new people are usually newborns, or immigrants (who have only
slightly better luck at finding jobs than newborns). I’d be more interested in comparing to the population of jobs holders, or to the ‘typical’ or recent changes in the jobs number.
The reporter should answer, is the employment picture getting
better or worse? By a lot or a little or too small to tell?
Hard to see how population changes over a month could possibly
affect that conclusion.
Lack of scientific training? Somehow i think journalism needs to have a course in statistics, information presentation, etc. Tufte+remedial stats, for example.
One might argue that this is the right comparison, as only some of this could be from immigration, and mostly from new births, so the comparison is “lost ability to feed new mouths”. On the other hand one might ask, whats the comparison to new people of working age added, or even more simply, to new people of working age.
I assume this is comfort with numbers. An alternate form of the problem appears in stories like, “The US government proposes X change to Q’s budget and some groups are outraged by its inadequacy. The Europeans made the larger Y delta last year.”
A little digging shows that X change is quite significant to the whole and further we already spend 10x what the EU spends. Finally, to many, it is murky what value marginal Q dollars have at its present size.
Now that sounds like juicy detail with which to interrogate the parties. But it never arises if one doesn’t know how to work the numbers, even at this basic level.
Many people misunderstand the role of reporters. The real role of reporters in society is to get eyeballs for advertisers. In order to accomplish this, the first priority is to be entertaining, and to repackage dry facts in an entertaining way. The 2nd priority is to do the first without breaking the illusion of being objective and breaking the viewer or reader’s trust, and losing credibility. In other words, be entertaining, but in a way that doesn’t make it too obvious that all you’re trying to do is be entertaining and get people’s attention so you can sell advertising. The news consumer, after all, believes they are being told “what is really going on in the world”, not just being entertained. This is a delicate balancing act, but a good professional journalist can do it.
I had this same thought when I heard an NPR report this morning explaining that the 8.something million jobs lost in this recession were the most since WWII. The US population has more than doubled since WWII!
Most of the folk going into reporting seem to be the same folk that tried very hard to avoid math or science in school. Strong analytical ability seems unlikely in that context.
Perhaps there is hope for the Fourth Estate: http://economix.blogs.nytimes.com/2010/02/05/finding-hope-in-the-jobless-numbers/ is by David Leonhardt, one of the NY Times’s best business reporters. He notes that “job growth needs to keep up with population growth, which translates to something like 150,000 new jobs a month” (consistent with my number of 252,000 additional residents every month; we would not expect 100 percent of the population to be employed).
Philip, that’s a good point. “Compared to what” is one of the most important tools in any sort of data analysis or understanding. The other two I rely on (because good tools come in threes!) are “How do you know?” (i.e., what’s your source?) and “Define your terms” (i.e., are we both talking about the same thing when we use a particular word).
@Wayne – I’d offer a more nuanced view of the journalist’s job… it might be more accurate, for example, to say that reporters in the Style, Living, Home, Sports, etc. sections are paid to be entertaining, or write about entertainment – no pretense there, and no 2nd priority – in order to get the eyeballs that then pay for hard-news reporting that serves the public good. I doubt, for example, that any journalist covering Haiti right now would think or act as though their job is to entertain. But plenty of entertaining articles elsewhere in the paper helped to pay the expenses for the important journalistic work going on in Haiti.
Maybe David read your post. 🙂
A few folks who are trying to revamp the local journalism scene had a roundtable hereabouts recently. A few notes from that:
First, the publisher of one of our local weeklies made no bones about the fact that she considers her advertisers partners. This is, of course, true for any advertiser supported media, but it was nice to have a publisher come right out and say that. Just a little for the “Reporters are supposed to be objective” crowd. Nope. Objectivity in reporting is hoohey and a total myth. It’d be cool if it were possible, but it’s like the Berkeley polisci student saying “Well, *true* communism hasn’t really been tried yet”: perhaps strictly true, but dangerously naive if adhered to as fact.
The other side of that is that there’s a startup that’s doing volunteer reporting. The articles these folks are turning out on some local issues are amazing, well cited, with links to source documents, lots of background, reporting like I’d always hoped it could be done.
They’re currently dealing with lawyers and subpoenas and such because the big money players are trying to come after them.
So if an editor says “here’s a press release from the U.S. Labor Department, turn this into an article”, it’s going to be the rare reporter who’s going to stay late to do research, to put up an article that’s going to alienate your political sources and possibly have the lawyers knocking on the publisher’s door. Kudos to David Leonhardt for taking at least that one small step.
Presenting numbers comparatively is just an aspect of effective communication, just like presenting charts with a proper scale, providing context, and concise writing.
Whether it’s journalism or any other profession, it seems that people interested in, and capable of, effective communication are becoming increasingly rare. (I’d guess there’s very few journalist who are even aware who Edward Tufte is.)
In my opinion, effective communications is one of the most important skills a person today can have. It’s at the top of the list when I’m reviewing candidates for hiring.
@Matt Henderson
I think ideas precede communication. The reporters here completely miss the MAIN IDEA (“what does this jobs report mean”), and show no indication of even trying to think what it means.
I suspect that if these reporters were numerically literate enough to understand that job growth should be compared to growth of the working-age population, then they could crank out the right words.
As a counterexample, reporters reporting on a freeze in Florida usually have the common sense to wonder about the impact of the freeze on the citrus crops. When they do, they don’t have trouble expressing it in writing.
I disagree that putting numbers into context is contrary to objectivity. There’s a difference between “reporting” and “stenography.” A reporter can easily put job statistics into meaningful context without getting into silly partisanship and or attributing results to particular policies.
What we usually get is kind of the worst of both worlds: “The economy unexpectedly gained X jobs in December, which the Administration attributed to the success of the Stimulus Package.” We get both a number without context (what is X compared to the population? How many of those X are temporary low-paying retail jobs?) AND an unfiltered political propaganda.
@Hubbert
Looking at the first few paragraphs of that NYT article again, I’m not even sure the reporting is attempting to make a point at all. He reports that the unemployment rate dropped 0.3% in a period of a month (from 10% to 9.7%), and follows that with what seems to be a contradiction: “yet with the pace of decline slowing.”
Anyway, that’s why I like to get my news from this blog. Phil does a good job of putting raw data into interesting perspective. 🙂
The LA Times front page today has an enormous headline – “Unemployment Rate Falls to 9.7%” directly beneath the headline is a graph of job gains/losses. A graph directly below the headline shows continuous, massive job losses for the past year (with the exception of the couple thousand gain in Nov 09).
Nowhere in the article does the reporter question how the Unemployment rate could go down from 10% to 9.7% amid continuous job losses.
Much praise is given to the Obama administration for their wise policies that led to this “reduction” in unemployment. To be fair, Obama seems to be doing his part: “The federal government also added 33,000 jobs last month, including the first 9,000 of hundreds of thousands of temporary workers to be hired for the 2010 census.”
http://www.latimes.com/business/la-fi-jobs6-2010feb06,0,4287218.story