If you have kids or are frequently around politicians you’ve no doubt experienced firsthand the use of selective facts in explaining one’s position on a certain topic or, why someone’s little brother has an arrow sticking out of his head. It seems people are imperfect and will sometimes not provide a complete set of facts when say arguing a point, applying for a mortgage, introducing themselves to waitress at Hooters, or sponsoring federal legislation. When I was a younger, more naïve shade tree economist I used to give the benefit of the doubt to people who could produce lots of numbers when making a point. Massive spreadsheets with impressive titles across the columns or down the rows convinced me that someone had their ‘S’ together. And if the purveyor of such a mountain of data could provide a statistical analysis complete with pie charts, bar graphs, or a USA Today like pictograph where two guys are arm wrestling and their foreheads come together in a Venn diagram and their bulging biceps and forearms form a Gaussian distribution with glistening beads of sweat compiling in a…. um, sorry, I got a little carried away. To sum up, I like many people, tend to believe statistics. Although I have become more skeptical as I’ve gotten older, I can still be had by a killer PowerPoint presentation.
Webster’s dictionary defines statistics as “a branch of mathematics dealing with the collection, analysis, interpretation, and presentation of masses of numerical data”. This ain’t addition we’re talking about. Statistics can be compromised by how the data is collected (and by whom), by how the analysis is performed (and by whom), and by the manner of presentation.
Let’s take my two favorite examples of statistical skullduggery for a ride.
Salsa is more popular than Ketchup(1)
Supporting statistic – A 1992 New York Times article titled ‘New Mainstream: Hot Dogs, Apple Pie and Salsa’ reported that “Last year, salsa -- a retailing category that includes picante, enchilada, taco and similar chili-based sauces -- took the condiment crown, outselling ketchup by $40 million in retail stores.”
Importance – The third paragraph of the Times article laid down the real point, “Epicures and food historians view the toppling of ketchup as the manifest destiny of good taste. Ketchup, that sugar-sweetened complement to fried food and meat, symbolizes "the bland old British-based American diet," said Elisabeth Rozin, a specialist in ethnic foods whose book "Ethnic Cuisine" will be reissued by Viking Penguin next month. The Mexican-inspired salsa is an uncooked relish fired by chili peppers that appeals, she said, "to cosmopolitan tastes."
The full story (well, at least my version) – To many Americans the thought of salsa, no matter the health benefits (there aren’t any unless you make it from scratch) or more flavorful taste compared to ketchup, besting the nation’s favorite condiment was unthinkable and wrong. And they are correct. The Times story used one statistic, retail sales, to make a point. According to a 2007 Wall Street Journal article covering the salsa/ketchup debate, the more extensive set of data favors ketchup:
“For example, take Information Resources, which tracks purchases at roughly 35,000 stores. The research firm found that salsa outsold ketchup, $462.3 million to $298.9 million, this year through August 12. But ketchup edged out salsa by units sold, 176 million to 174.9 million. And because ketchup bottles tend to be bigger, ketchup trounced salsa in pounds sold, 329.8 million to 184.6 million. Research from ACNielsen, which monitored point-of-sale data, shows similar trends for sales and units sold.
It’s also important to note that the market-research data only count purchases in stores, meaning those ketchup packets handed out with the billions of fast-food burgers consumed annually don’t count (nor does salsa served with chips and other Mexican restaurant fare).
Meanwhile, another market researcher has found that ketchup has a much greater presence in homes. According to the NPD Group’s National Eating Trends report — which is based on diaries kept by 5,000 people in 2,000 households — 48% of households used ketchup in a typical two-week period last week, three times the percentage that used salsa. That proportion has held steady for over a decade: The percentages were 48 and 15, respectively, in 1996.”
So has salsa really taken the condiment crown? With ketchup being consumed in larger quantities by more people, I’d have to say no.
There are more than 40 million Americans without health insurance
Supporting statistic – In the U.S. Census Bureau’s Health Insurance Coverage: 2006 report, the Highlights section comments that “Both the percentage and the number of people without health insurance increased in 2006. The percentage without health insurance increased from 15.3 percent in 2005 to 15.8 percent in 2006 and the number of uninsured increased from 44.8 million to 47.0 million.”
Importance – The Census Bureau’s report has been cited by politicians and pundits as hard evidence that a massive number of Americans are not insured and at considerable risk. This statistic is often cited as exhibit one as to why universal nationalized healthcare is needed.
More government, less freedom, too much money, we’ll end up having the U.K.’s shity healthcare system, blah blah blah. From a purely economic standpoint the problem with the increasing cost of healthcare isn’t the number of people that could be added to taxpayer supported roles, it’s the number of sticky hands your money goes through before it gets to your doctor. One idea that’s been floated by the Obama administration as well as the Heritage Foundation (a political Mork & Mindy pairing if there ever was one) is for the use of public insurance exchanges to lower the cost and widen the options for health insurance at a pre-tax cost. That would would solve a real problem impacting most Americans.
The full story (well, at least my version) – The Census Bureau’s highlight is just that. Cited several times within the body of the report is the problem of accurately collecting data on individual healthcare coverage:
“Compared with other national surveys, the CPS ASEC’s estimate of the number of people without health insurance more closely approximates the number of people who were uninsured at a specific point in time during the year than the number of people uninsured for the entire year.”
In addition, the Census Bureau notes that its census takers are not trained to manage healthcare questions and that many people answer incorrectly, that is, they don’t realize they’re covered under state or federal government programs:
“The CPS ASEC data underreport Medicare and Medicaid coverage compared with enrollment and participation data from the Centers for Medicare and Medicaid Services (CMS). Because the CPS is largely a labor force survey, interviewers receive less training on health insurance concepts than labor concepts. Additionally, many people may not be aware that a health insurance program covers them or their children if they have not used covered services recently.”
The Census Bureau offers many caveats to the accuracy of its healthcare survey but those who argue for universal government sponsored or supplied healthcare never mention it. A lie of omission at best, a true distortion of reality at worst. So goes statistics.
- Later
(1) For complete transparency I prefer salsa to ketchup on a hotdog.
Webster’s dictionary defines statistics as “a branch of mathematics dealing with the collection, analysis, interpretation, and presentation of masses of numerical data”. This ain’t addition we’re talking about. Statistics can be compromised by how the data is collected (and by whom), by how the analysis is performed (and by whom), and by the manner of presentation.
Let’s take my two favorite examples of statistical skullduggery for a ride.
Salsa is more popular than Ketchup(1)
Supporting statistic – A 1992 New York Times article titled ‘New Mainstream: Hot Dogs, Apple Pie and Salsa’ reported that “Last year, salsa -- a retailing category that includes picante, enchilada, taco and similar chili-based sauces -- took the condiment crown, outselling ketchup by $40 million in retail stores.”
Importance – The third paragraph of the Times article laid down the real point, “Epicures and food historians view the toppling of ketchup as the manifest destiny of good taste. Ketchup, that sugar-sweetened complement to fried food and meat, symbolizes "the bland old British-based American diet," said Elisabeth Rozin, a specialist in ethnic foods whose book "Ethnic Cuisine" will be reissued by Viking Penguin next month. The Mexican-inspired salsa is an uncooked relish fired by chili peppers that appeals, she said, "to cosmopolitan tastes."
"Bland old British-based American diet" versus "Ethnic cuisine" suited "to cosmopolitan tastes"? Based on retail sales comparing the salsa category to ketchup, the Anglo-sphere takes another one in the nuts.
The full story (well, at least my version) – To many Americans the thought of salsa, no matter the health benefits (there aren’t any unless you make it from scratch) or more flavorful taste compared to ketchup, besting the nation’s favorite condiment was unthinkable and wrong. And they are correct. The Times story used one statistic, retail sales, to make a point. According to a 2007 Wall Street Journal article covering the salsa/ketchup debate, the more extensive set of data favors ketchup:
“For example, take Information Resources, which tracks purchases at roughly 35,000 stores. The research firm found that salsa outsold ketchup, $462.3 million to $298.9 million, this year through August 12. But ketchup edged out salsa by units sold, 176 million to 174.9 million. And because ketchup bottles tend to be bigger, ketchup trounced salsa in pounds sold, 329.8 million to 184.6 million. Research from ACNielsen, which monitored point-of-sale data, shows similar trends for sales and units sold.
It’s also important to note that the market-research data only count purchases in stores, meaning those ketchup packets handed out with the billions of fast-food burgers consumed annually don’t count (nor does salsa served with chips and other Mexican restaurant fare).
Meanwhile, another market researcher has found that ketchup has a much greater presence in homes. According to the NPD Group’s National Eating Trends report — which is based on diaries kept by 5,000 people in 2,000 households — 48% of households used ketchup in a typical two-week period last week, three times the percentage that used salsa. That proportion has held steady for over a decade: The percentages were 48 and 15, respectively, in 1996.”
So has salsa really taken the condiment crown? With ketchup being consumed in larger quantities by more people, I’d have to say no.
There are more than 40 million Americans without health insurance
Supporting statistic – In the U.S. Census Bureau’s Health Insurance Coverage: 2006 report, the Highlights section comments that “Both the percentage and the number of people without health insurance increased in 2006. The percentage without health insurance increased from 15.3 percent in 2005 to 15.8 percent in 2006 and the number of uninsured increased from 44.8 million to 47.0 million.”
Importance – The Census Bureau’s report has been cited by politicians and pundits as hard evidence that a massive number of Americans are not insured and at considerable risk. This statistic is often cited as exhibit one as to why universal nationalized healthcare is needed.
More government, less freedom, too much money, we’ll end up having the U.K.’s shity healthcare system, blah blah blah. From a purely economic standpoint the problem with the increasing cost of healthcare isn’t the number of people that could be added to taxpayer supported roles, it’s the number of sticky hands your money goes through before it gets to your doctor. One idea that’s been floated by the Obama administration as well as the Heritage Foundation (a political Mork & Mindy pairing if there ever was one) is for the use of public insurance exchanges to lower the cost and widen the options for health insurance at a pre-tax cost. That would would solve a real problem impacting most Americans.
The full story (well, at least my version) – The Census Bureau’s highlight is just that. Cited several times within the body of the report is the problem of accurately collecting data on individual healthcare coverage:
“Compared with other national surveys, the CPS ASEC’s estimate of the number of people without health insurance more closely approximates the number of people who were uninsured at a specific point in time during the year than the number of people uninsured for the entire year.”
In addition, the Census Bureau notes that its census takers are not trained to manage healthcare questions and that many people answer incorrectly, that is, they don’t realize they’re covered under state or federal government programs:
“The CPS ASEC data underreport Medicare and Medicaid coverage compared with enrollment and participation data from the Centers for Medicare and Medicaid Services (CMS). Because the CPS is largely a labor force survey, interviewers receive less training on health insurance concepts than labor concepts. Additionally, many people may not be aware that a health insurance program covers them or their children if they have not used covered services recently.”
The Census Bureau offers many caveats to the accuracy of its healthcare survey but those who argue for universal government sponsored or supplied healthcare never mention it. A lie of omission at best, a true distortion of reality at worst. So goes statistics.
- Later
(1) For complete transparency I prefer salsa to ketchup on a hotdog.
