How Politicians Misuse Statistics
Let’s debunk an old canard, one you’ve probably heard – there are three kinds of lies: lies, damned lies and statistics. It’s not true – statistics don’t lie, but the people who use them often do. And since few people have the patience, the access or the knowledge to read and interpret all of the data behind the statistic, the manipulators – be they snake-oil salesmen or politicians – can use actual, non-lying statistics to support a lie and make you believe it.
Take, for example, a well-known statistic from the past couple of years: U.S. auto companies pay their workers $76 an hour. When many news sources and politicians uncritically reported that statement a couple of years ago, people on all sides of the political spectrum were stunned – and they could be forgiven if they imagined that statistic adding up to $3,000 a week paychecks. Auto workers could also be forgiven for being outraged at the reports as they looked at their weekly paystubs of about $1,100 a week – excellent pay, make no mistake, but far less than the widely spread statistic.
And yet, the $76 an hour statistic was not a lie. It was taken directly from financial reports made by the Big Three auto companies. How could there be such a discrepancy between two facts that appear to represent the same thing? The answer is simple – statistics rely on the data used to derive them.
In this case, the $76 figure represents the total hourly labor cost for the auto company. They got the figure by adding up all of the money they spend in the category of labor – wages, benefits, supplemental pay to laid-off employees, retirement benefits to retired employees and disability pay to injured workers – then divided that figure by the number of hours that workers actually put in as labor. In short, $76 is an accurate reflection of the company’s hourly labor costs, but no worker pocketed $76 an hour, even when you add in the benefits.
The fact is that legacy costs – retirement benefits and health insurance costs paid to retired workers and fanatically protected by the unions – artificially inflated the hourly labor figure. By using the distortion, many politicians allowed the opposition to point out the misused statistic and dismiss the reasonable argument that union auto workers make about 50 percent more than those working for foreign auto makers.
Distorting statistics is certainly not confined to one side of the political spectrum. When President Obama gave his State of the Union speech in January, the instant polls following the speech were overwhelmingly positive – unbelievably so, in fact. The reason for that was another way that politicians and the media distort or misreport statistics – a distorted survey sample. Most networks surveyed people who had watched the speech – and it’s a foregone conclusion that those who watched the speech were far more likely to already be disposed to liking the president. That didn’t stop left-wing pundits and politicians from citing the positive statistics as proof that the president’s popularity was surging in the wake of his speech.
Nearly every statistic that you hear from a politician is distorted in some way to appear more favorable to his position. He’s not lying – he’s selectively reporting the statistics, or relying on data that was skewed to begin with. Without access to the underlying data and information about how the data was collected and analyzed, you really don’t know how reliable those truth-y sounding numbers are.
The truth is that politicians – left, right, Conservative, Progressive or Moderate – use statistics to manipulate and sway the public. Even worse, they often do so inadvertently because they, too, rely on statistics handed to them by lobbyists and special interest groups to sway the politicians to vote for their agenda. In fact, the county would probably benefit greatly if it passed a law requiring anyone who takes public office to take a basic college-level statistics course so they could recognize spin when they see it.
In the meantime, the best thing you can do is educate yourself, not only about statistics but about how to better understand them. Don’t just swallow the statistical lies whole. Instead:
• Consider who is quoting the statistic and what they have to gain. This is true whether the statistic is about how many children prefer coco-coated rice puffs for dinner or how much a bill will add to your taxes.
• Educate yourself about the basics of statistics so that you understand the differences between mean, median and mode, and how the data for the statistic was collected.
• Go directly to the source. With even a basic understanding of how to calculate statistics, you can read summaries of data and understand exactly how the statistic you heard fits the facts.
• Check what the other side is saying. Anytime a politician or media source reports a statistic, there’s a good chance that people who oppose the policy or politician will try to debunk it.
There’s a good chance that the truth lies somewhere between the two, and understanding both sides of the issue will help you find the truth.