It may not seem like a big thing, but you and I know it, and journalists know it, and so do marketers and political advisors: words matter. Consider today's Google News headlines regarding October unemployment rates.







On top of this block of stories about unemployment figures, the Wall Street Journal's headline says the jobless rate declines in 23 states, but the teaser from the article refers to the rate as "little changed in most states." Meanwhile, the article from Bloomberg claims that the rate in Nebraska "remained" 4.6 pct:

Nebraska's unemployment rate stayed steady at 4.6 percent in September

However, a few sentences later, they seem to suggest a different picture:

The state's July 2010 rate was 4.7 percent and June's was 4.9 percent.

But you might say, yeah, but that's a pretty small change, small enough to maybe qualify as "staying steady." But then, what do we make of another bit of data mentioned later in the article:

Unemployment also spiked in Lincoln, the capital city, to 4.1 percent, from August's 3.9 percent.

So a drop of .2 percent over two months is "staying steady", but an increase of .2 percent qualifies as a "spike".

One has to wonder if the Bloomberg has some vested interest in painting a picture of unrelenting gloom here.

The funniest thing about all this? These rates--the "steady" 4.6 percent for Nebraska and the "spike" to 4.1 percent for Lincoln are HALF the unemployment rate for the country as a whole. So these ten-of-a-percent changes that seem so intransigently headed up are in fact changes to a number that most cities in the US would kill for.

Again, one has to wonder....does this sound like news reporting to you?

What I believe -- by Isaac Asimov

I believe in evidence. I believe in observation, measurement, and reasoning, confirmed by independent observers. I'll believe anything, no matter how wild and ridiculous, if there is evidence for it. The wilder and more ridiculous something is, however, the firmer and more solid the evidence will have to be.
-Isaac Asimov, scientist and writer (1920-1992)

Real Life Cognitive Accuracy

The arguments over health care reform have produced a lot of wild claims, some for, lots against. How do we, as average American consumers of health care, distinguish the heat from the light? In this article from the New York Times, Paul Krugman demonstrates one method: look deeper, get more of the story than the headline, consider the numbers, etc.

Krugman addresses three myths he says are "believed by many people who consider themselves well-informed, but who have actually fallen for deceptive spin":

1. That Obama's plan is a "government takeover of one-sixth of the economy, the share of G.D.P. currently spent on health." This argument fails on several levels, mainly for a lack of specificity. Krugman responds that two-thirds of the health dollars spent in the US either already comes from the US government or comes out of the pockets of those without health care coverage. Only one-third comes from insurance companies. Did you know that? I didn't! When you have those numbers at hand, the argument about a "take-over" falls apart.

2. That the health care reform bills coming up for a vote "do nothing to cut costs." Krugman notes:

To support this claim, critics point to reports by the Medicare actuary, who predicts that total national health spending would be slightly higher in 2019 with reform than without it.

The problem with this "fact" is that it leaves out the fact that the "slightly higher" spending, about 1% higher than what is predicted if reform fails, would actually include coverage for an additional 35 million Americans! So we pay a tiny bit more and get something closer to universal coverage. What's more, in the second decade after the passage of the current proposed program, the numbers look even better. Sounds like a good deal when you include more facts!

3. That the reform program doesn't pay for itself. This flies in the face of the CBO's report that shows it does in fact pay for itself while increasing coverage for those who need it most. Krugman notes that critics of the program justify this myth by saying that when things get tough, some future Congress will loosen the controls and let the costs rise. But that says nothing about the program, only about Congress, and Krugman notes that this has not happened with cost controls on Medicare--those already enacted have "stuck" despite changes in budgets and revenues.

So there you have it: a real-life exercise in how to read between the lines, or rather, how to read the whole sentence instead of just the part that supports your personal confirmation bias.

Framing and semantics

Researchers continue to study the process known as "framing" and their results have interesting implications for language. This post on BPS RESEARCH DIGEST, titled "How framing affects our thought processes", describes a set of studies evaluating reactions to the words "tax" and "offset" by participants making decisions between two products. Left-leaning participants did not appear to respond differently to the two words, while right-leaning participants were much less likely to choose a more expensive "green" option if the price difference was labeled "tax" instead of "offset."

The study also showed that, since we tend to favor what we consider first in a one-on-one comparison task, the order in which participants considered the two options was strongly influenced by the different labels. Republican respondents tended to evaluate the lower-priced item longer when the price difference was labeled a "tax", and in some cases, failed to consider the higher-priced item first even when instructed to.

The statistically significant difference would suggest that how we say things can affect not only a direct reaction to our words, but also the underlying evaluation process of our overall message. The researchers, David Hardisty and Weber, concluded:

What might seem like a trivial semantic difference to one person can have a large impact on someone else.

I wonder if we will ever come to a day when people don't automatically dismiss "semantic differences" as "trivial".

The External Referent

The issue of using an external referent has many complexities. For example, consider standard international (SI) units. Science has long recognized the value and importance of standard measures, like ohms and grams and decibels. Really, the principle of reproducibility rests squarely on this idea--I cannot reliably reproduce your experiment if I use one type of unit and you use something else that may or may not directly translate.

On the other hand, as scientists learn more about the way things work, and develop increased precision, one familiar unit of measure after another gives way to a precise standard, usually tied to a relatively abstract and often invisible (to us) phenomenon like the width of hydrogen's line in a mass spectroscope or the number of vibrations in a finely crafted tuning fork. While these careful and deliberately defined units make science work more effectively, we often find them hard to relate to in our daily life.

The time-honored units replaced by these scientific split-hairs sprang from tasks we no longer perform or from ways of looking at the world few of us still have. The furlong, approximately 220 yards, derived from the length of a furrow--it also applied to one side of an acre and to 1000 links of chain. These units have little meaning for most people today, but when the units came into use, everyone who used them could visualize their external referent.

When Robert Crease asked his readers about their favorite standard unit, he was deluged with answers. In his follow-up post, he compares some "pre-SI" units with those in use in the lab today. Surveyors used a unit called a chain, a actual piece of metal chain 22 yards long, to measure land area--a vivid and unequivocally external referent point. Farmers and land owners could easily picture a "collop", defined as the area needed to graze "one sow or two yearling heifers or six sheep or twelve goats or six geese and a gander."

In explaining the appeal of these old, familiar, approximate, good-enough measures, author Eric Cross notes that they were:

reckoned on the things a man could see about him, so that, wherever he was, he had an almanac.

Or in other words, a different kind of external referent. It would seem to suggest that the referent we choose will serve us more effectively if we and those we intend to share our results with have some method for visualizing the basis of the referent.

Really? We Still Read About This?

In this new century, with breakthroughs in neuroscience, physics, astronomy, meteorology, and many other scientific and technological fields, do we really still get excited about a groundhog pulled out of its den and waved in the air? This year, Reuters reports, over 12,000 people attended this childish theatrical event, some of whom:

came from as far as Chile and the Netherlands

Setting aside the likelihood that anyone would come to freezing Pennsylvania from where it is currently summer SOLELY to see this event, and overlooking the forced nature of this event, what does this say about our culture?

Reuters provides a delicious if perhaps inadvertent, commentary on this foolishness by noting that other groundhog viewing occur around the country, and:

New York City's "Staten Island Chuck" did not see his shadow.

Well, duh. What are the odds that every place in the US would have the "same" weather at the "same" time? Even if every groundhog had the "same" experience, does "winter" mean the "same" thing in all 50 states? Not hardly.