What’s in a Number?
What are the President’s approval ratings? How are the unemployment figures these days? Are you for or against calling french fries “freedom fries”? What’s the employee satisfaction rating in your company? How many Americans are tired of polls?
Enough with the numbers already.
You can’t click through the news channels these days without running into a couple of bickering pundits hurling statistics at each other as if they were tossing Frisbees. You’ll see the same statistics-filled exchanges going on in Congress, high-school debates, campaign speeches, and the front page of every newspaper in the country. When it comes to proving a point or knowing where groups of people stand on an issue, the world has gone nuts for numbers. Yet often we don’t know where they came from, how they were interpreted, or what they really mean.
There’s no doubt statistics are powerful tools for proving points and selling products. But if you’re a customer, should you trust them? If you’re a business using statistics to sway opinion, how do you honestly prove a point and earn trust? And if you’re a researcher sorting through columns and columns of statistics, how do you delve deeper and make sense out of why people answer the way they do?
Andrew Lang, a 19th century poet and novelist, once said: “He uses statistics as a drunken man uses lamp posts—for support rather than illumination.” It seems statistics are still being used much the same way. Rather than using the numbers to help prove a point, people are making the statistics themselves the point—often without careful interpretation of the findings. And unless we know why the numbers are what they are, we’re cheating ourselves of some valuable information that helps us predict, prevent, and prevail.
When you get right down to it, numbers don’t always speak for themselves. Who knows, they might be the result of faulty or subjective techniques. And even if the numbers are accurate and the “what” questions have been answered, that’s just half of the equation. The other half is the “why.” We need to probe in depth and find out why respondents answered the way they did. We need to be careful we don’t deceive by making an untrue true, by creating a sample or circumstance that tells customers what they want to hear.
Even when we do our very best to make research fair and accurate, innocent mistakes can skew results. From how a question is phrased, to how the results are interpreted, to a simple error in tallying the results, all research has at least some bias and flaws. Just ask the Gallup organization, which admits to having an error rate of plus or minus 5 percent.
Sure, statistics can deceive people, but they also can be fun. I mean, who doesn’t enjoy knowing that only 13 percent of people brush their teeth from side to side? Or the “fact” that 56 percent of women pay the bills in a marriage. (Now there’s a statistic that can be interpreted at least a couple different ways.)
The bottom line is, statistics are part of the story, never the whole story. Unless we know for certain the correct steps and measures are taken to gather information and interpret the findings—including the who’s, what’s, and why’s behind them—statistics will never be more than, well, numbers. IBI
Enough with the numbers already.
You can’t click through the news channels these days without running into a couple of bickering pundits hurling statistics at each other as if they were tossing Frisbees. You’ll see the same statistics-filled exchanges going on in Congress, high-school debates, campaign speeches, and the front page of every newspaper in the country. When it comes to proving a point or knowing where groups of people stand on an issue, the world has gone nuts for numbers. Yet often we don’t know where they came from, how they were interpreted, or what they really mean.
There’s no doubt statistics are powerful tools for proving points and selling products. But if you’re a customer, should you trust them? If you’re a business using statistics to sway opinion, how do you honestly prove a point and earn trust? And if you’re a researcher sorting through columns and columns of statistics, how do you delve deeper and make sense out of why people answer the way they do?
Andrew Lang, a 19th century poet and novelist, once said: “He uses statistics as a drunken man uses lamp posts—for support rather than illumination.” It seems statistics are still being used much the same way. Rather than using the numbers to help prove a point, people are making the statistics themselves the point—often without careful interpretation of the findings. And unless we know why the numbers are what they are, we’re cheating ourselves of some valuable information that helps us predict, prevent, and prevail.
When you get right down to it, numbers don’t always speak for themselves. Who knows, they might be the result of faulty or subjective techniques. And even if the numbers are accurate and the “what” questions have been answered, that’s just half of the equation. The other half is the “why.” We need to probe in depth and find out why respondents answered the way they did. We need to be careful we don’t deceive by making an untrue true, by creating a sample or circumstance that tells customers what they want to hear.
Even when we do our very best to make research fair and accurate, innocent mistakes can skew results. From how a question is phrased, to how the results are interpreted, to a simple error in tallying the results, all research has at least some bias and flaws. Just ask the Gallup organization, which admits to having an error rate of plus or minus 5 percent.
Sure, statistics can deceive people, but they also can be fun. I mean, who doesn’t enjoy knowing that only 13 percent of people brush their teeth from side to side? Or the “fact” that 56 percent of women pay the bills in a marriage. (Now there’s a statistic that can be interpreted at least a couple different ways.)
The bottom line is, statistics are part of the story, never the whole story. Unless we know for certain the correct steps and measures are taken to gather information and interpret the findings—including the who’s, what’s, and why’s behind them—statistics will never be more than, well, numbers. IBI