# Lying with statistics **URL:** \ **Category:** Miscellaneous and Personal Stuff I Must Share\ **Created:** [July 6, 2012, 6:49pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453 "2012-07-06T18:49:09Z")\ **Posts on this page:** 20\ **Page:** 2
**Author:** ![Maastricht](https://avatars.discourse-cdn.com/v4/letter/m/a88e57/32.png) [@Maastricht](https://boards.straightdope.com/u/Maastricht)\ **Post date:** [July 9, 2012, 8:07am UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/21 "2012-07-09T08:07:25Z")
["_How to lie with statistics_’](http://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728/ref=sr_1_1?ie=UTF8&qid=1341821152&sr=8-1&keywords=how+to+lie+with+statistics)is an actual book. A classic from 1993. It was required reading when I went to college. ---
**Author:** ![Mangetout](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mangetout/32/19_2.png) [@Mangetout](https://boards.straightdope.com/u/Mangetout)\ **Post date:** [July 9, 2012, 9:05am UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/22 "2012-07-09T09:05:57Z")
> [@denquixote](#): > > It appears to me as if Shark was comparing students at one school only against each other while you are using some standardized performance chart or evaluation. Indeed - against a national average (which is more meaningful than a local one) ---
**Author:** ![grude](https://avatars.discourse-cdn.com/v4/letter/g/e47774/32.png) [@grude](https://boards.straightdope.com/u/grude)\ **Post date:** [July 9, 2012, 9:14am UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/23 "2012-07-09T09:14:23Z")
This seems to be more of an issue of people misinterpreting things but I hate people who use average life expectancy as some kind of indication of when an ADULT will die. Someone will say hey Nigeria has an average life expectancy of 35(may not be actual#) holy shit people there don’t live to see 40?! HEADSMACK! ---
**Author:** ![Mangetout](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mangetout/32/19_2.png) [@Mangetout](https://boards.straightdope.com/u/Mangetout)\ **Post date:** [July 9, 2012, 9:27am UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/24 "2012-07-09T09:27:35Z")
> [@purple\_cow](#): > > One of my favorites from the past was “Al Gore’s house in TN uses 254 times more energy in one year than the average person’s house does per month.” People of course focus on the 254 and completely miss the juxtaposition of a year and a month. So, the statement is true, but it’s really a lie. (there were lots of other fudges that brought the actual value down to like 3X average but the year vs month was the really egregious part) Agreed. Deceit seems the only clear motive for such a juxtaposition. The other way around wouldn’t necessarily raise concerns (X uses thrice as much in a month as Y uses in a year). ---
**Author:** ![CalMeacham](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/calmeacham/32/35_2.png) [@CalMeacham](https://boards.straightdope.com/u/CalMeacham)\ **Post date:** [July 9, 2012, 12:56pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/25 "2012-07-09T12:56:59Z")
> [@Maastricht](#): > > ["_How to lie with statistics_’](http://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728/ref=sr_1_1?ie=UTF8&qid=1341821152&sr=8-1&keywords=how+to+lie+with+statistics)is an actual book. A classic from 1993. It was required reading when I went to college. It’s far older than that. I bought my (used) copy in the 1980s. The book is actually from 1954: > **[How to Lie with Statistics](https://en.wikipedia.org/wiki/How_to_Lie_with_Statistics)** > > How to Lie with Statistics is a book written by Darrell Huff in 1954, presenting an introduction to statistics for the general reader. Not a statistician, Huff was a journalist who wrote many how-to articles as a freelancer. > The book is a brief, breezy illustrated volume outlining the misuse of statistics and errors in the interpretation of statistics, and how errors create incorrect conclusions. > In the 1960s and 1970s, it became a standard textbook introduction to the subject of statistics for... ---
**Author:** ![Nava](https://avatars.discourse-cdn.com/v4/letter/n/da6949/32.png) [@Nava](https://boards.straightdope.com/u/Nava)\ **Post date:** [July 9, 2012, 12:58pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/26 "2012-07-09T12:58:28Z")
> [@Dave\_Hartwick](#): > > One that I used to see in charts for managed funds or brand vs brand performance is fooling with the Y axis. The variance might only be 2%, so change the unit to dollars, say, or days, whatever, and then start the axis at a few points below the worst performer. That way the 2% variance might take up 70% of the axis. > > Not sure if I explained it very well and don’t feel like looking for an example, sorry. Moving zero and blowing up the scale, if I understood it correctly. Makes a minor variation in a very narrow margin look like it’s “all over the chart!” (which it is, but not all over the percentile range). An example from yesterday’s Spanish news: Election figures for Libya were being reported, depending on the channel, as “over 60%” or “62%”. Similar participation rates for European or American elections would have been reported as “almost 40% abstention rate”: the statistic is the same, but the focus is completely flipped. ---
**Author:** ![Small\_Clanger](https://avatars.discourse-cdn.com/v4/letter/s/9fc348/32.png) [@Small\_Clanger](https://boards.straightdope.com/u/Small_Clanger)\ **Post date:** [July 9, 2012, 2:04pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/27 "2012-07-09T14:04:37Z")
There’s a BBC Radio 4 program all about statistics [More or Less](http://www.bbc.co.uk/programmes/b006qshd) It seems to me that _every single time_ they analyse a headline or politician-quoted statistic they demolish it. ---
**Author:** ![Kevbo](https://avatars.discourse-cdn.com/v4/letter/k/e47774/32.png) [@Kevbo](https://boards.straightdope.com/u/Kevbo)\ **Post date:** [July 9, 2012, 3:32pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/28 "2012-07-09T15:32:17Z")
One of my favorites (I think I mentioned it in a recent thread) : Alcohol is involved in around 31% of traffic fatalities. This means you could eliminate over 2/3 of traffic deaths by keeping the damned sober people off the road. Of course that is a joke, but I do think that so many law enforcement resources have been appropriated by MADD initiatives, that general traffic enforcement is suffering, to the point that drunks are becoming harder to spot. ---
**Author:** ![dngnb8](https://avatars.discourse-cdn.com/v4/letter/d/977dab/32.png) [@dngnb8](https://boards.straightdope.com/u/dngnb8)\ **Post date:** [July 9, 2012, 3:41pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/29 "2012-07-09T15:41:37Z")
Ive always said Liars Figure and Figures Lie ---
**Author:** ![mhendo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mhendo/32/3159_2.png) [@mhendo](https://boards.straightdope.com/u/mhendo)\ **Post date:** [July 9, 2012, 4:05pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/30 "2012-07-09T16:05:24Z")
> [@billfish678](#): > > I seem to recall a new story years ago (and I doubt it is unique) along the lines of “20 percent of Americans live in poverty!” > > Which sounds bad until you realize that the definition of poverty was litterally the bottom 20 percent (economics wise) of the population. > > Or whatever the actual percentage was… As far as i know, the official federal poverty line in America is, and has been for quite some time, a dollar value. Or, more accurately, a set of values, depending on how many people are in your family. The 2012 figures are listed [here](http://aspe.hhs.gov/poverty/12poverty.shtml/). As you can see, there are actually two different versions of the poverty measure: a set of thresholds used by the census bureau for statistical purposes, and a set of guidelines used by government departments for administrative purposes. The two sets of numbers, while reasonably close, are not identical. The 2011 Census thresholds can be found in [this file](http://www.census.gov/hhes/www/poverty/data/threshld/thresh11.xls) (Excel spreadsheet). > [@Mangetout](#): > > Indeed - against a national average (which is more meaningful than a local one) Actually, it’s also possible (although rather unlikely) for 90% of the students in a class to be above that class’s average. Say you have 30 students. 8 have a score of 90, 10 have a score of 88, 9 have a score of 85, 2 have a score of 10 and 1 has a score of 14. The class average for this class is 79.96, meaning that only 3 students scored below the average. Any uneven, bimodal distribution allows a large majority of a sample to fall above or below the average. ---
**Author:** ![toyo](https://avatars.discourse-cdn.com/v4/letter/t/ea5d25/32.png) [@toyo](https://boards.straightdope.com/u/toyo)\ **Post date:** [July 9, 2012, 4:08pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/31 "2012-07-09T16:08:11Z")
> [@purple\_cow](#): > > One of my favorites from the past was “Al Gore’s house in TN uses 254 times more energy in one year than the average person’s house does per month.” People of course focus on the 254 and completely miss the juxtaposition of a year and a month. Are you sure you are remembering correctly? An internet search doesn’t bring up anything like what you quoted. The two Snopes articles have it the other way, comparing Gore’s energy use in one month to the average homeowners use in a year. ---
**Author:** ![Rala](https://avatars.discourse-cdn.com/v4/letter/r/8797f3/32.png) [@Rala](https://boards.straightdope.com/u/Rala)\ **Post date:** [July 9, 2012, 4:09pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/32 "2012-07-09T16:09:24Z")
I’ve always liked [this chart](http://www.seanbonner.com/blog/archives/piratesarecool.jpg) - Global Average Temperature vs Number of Pirates. ---
**Author:** ![Snickers](https://avatars.discourse-cdn.com/v4/letter/s/5daacb/32.png) [@Snickers](https://boards.straightdope.com/u/Snickers)\ **Post date:** [July 9, 2012, 4:40pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/33 "2012-07-09T16:40:35Z")
At orientation before starting my first year in college, the powers that be had all us students take a health questionnaire. According to its results, 100% of incoming Marquette freshmen did not smoke. Simply walking around on campus the first week gave the lie to that statistic. While filling out the questionnaire, I looked around the room and noted that I was one of very few students that didn’t have a parent accompanying them. I’m guessing that my fellow students felt it more prudent to lie to preserve family harmony than be honest on a questionnaire that didn’t really accomplish anything anyway. ---
**Author:** ![Iggy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/iggy/32/3364_2.png) [@Iggy](https://boards.straightdope.com/u/Iggy)\ **Post date:** [July 9, 2012, 4:59pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/34 "2012-07-09T16:59:14Z")
I posted in [this](http://boards.straightdope.com/sdmb/showpost.php?p=15248003&postcount=16)over in the Elections category in the [How is Obamacare the “largest tax increase in US history”?](http://boards.straightdope.com/sdmb/showthread.php?p=15248003#post15248003) thread. Without spinning this thread too far off, statistics about comparative terms (the biggest, largest, smallest, etc… ) are easy to lie with if a frame of reference is not provided. In one frame of reference, an assertion may be true. In a different (and equally valid) frame of reference the assertion may be false. The good news: your raise was twice as big as your bosses’ raise! The reality: You were making $25,000 and got a 20% raise. You now make $30,000. Your boss was making $100,000 and only got a 10% raise. He now makes $110,000. Of course in another valid way of looking at it your boss’ $10,000 raise was twice as big as your $5000 raise. ---
**Author:** ![FatBaldGuy](https://avatars.discourse-cdn.com/v4/letter/f/ecd19e/32.png) [@FatBaldGuy](https://boards.straightdope.com/u/FatBaldGuy)\ **Post date:** [July 9, 2012, 5:09pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/35 "2012-07-09T17:09:30Z")
One of my favorites is the insurance ads that claim, “Customers who switched from Company A to Company B saved an average of $XXX.” What this statistic completely ignores is the people with Company A insurance who compared rates and didn’t switch because their premiums were already lower. ---
**Author:** ![mhendo](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/mhendo/32/3159_2.png) [@mhendo](https://boards.straightdope.com/u/mhendo)\ **Post date:** [July 9, 2012, 5:16pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/36 "2012-07-09T17:16:18Z")
> [@FatBaldGuy](#): > > One of my favorites is the insurance ads that claim, “Customers who switched from Company A to Company B saved an average of $XXX.” > > What this statistic completely ignores is the people with Company A insurance who compared rates and didn’t switch because their premiums were already lower. Yeah, my wife used to roll her eyes because i would yell at the TV: “Of course they saved money, because if they didn’t then they wouldn’t have switched!” ---
**Author:** ![Zyada](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/zyada/32/3028_2.png) [@Zyada](https://boards.straightdope.com/u/Zyada)\ **Post date:** [July 9, 2012, 5:42pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/37 "2012-07-09T17:42:43Z")
Sometimes it’s about cherry-picking your source. There was a local vote to add a tax to car rentals (and, IIRC, hotels, cabs, things that visitors were more likely to use). A local paper quoted an Enterprise spokesperson as saying that 20% of their business was with locals renting while their car was being repaired. If I hadn’t had a car accident prior to reading the article, I would not have known that Enterprise cuts deals with the insurance companies to provide cheaper rates in trade for getting preferred supplier status. ---
**Author:** ![Enderw24](https://avatars.discourse-cdn.com/v4/letter/e/ba9def/32.png) [@Enderw24](https://boards.straightdope.com/u/Enderw24)\ **Post date:** [July 9, 2012, 5:48pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/38 "2012-07-09T17:48:42Z")
My current favorite can be found in my office’s break room. There’s a liquid dishwashing detergent Ajax that proudly proclaims on the 34 ounce bottle: 40% MORE! Than competitors’ 24 oz bottle Why yes. Yes it is. ---
**Author:** ![Zyada](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/zyada/32/3028_2.png) [@Zyada](https://boards.straightdope.com/u/Zyada)\ **Post date:** [July 9, 2012, 6:09pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/39 "2012-07-09T18:09:07Z")
> [@don\_t\_ask](#): > > My favorite has always been the Pepsi taste challenge from many years ago on TV. > > What I admired was the clever wording of the stated facts at the end. Having shown a few avowed Coke lovers choosing Pepsi in a blind test, the voice over said, “In recent blind taste tests the majority of Coke lovers preferred the taste of Pepsi.” > > So what does that mean? Let’s say I break up all my Coke lovers into groups of 3 and call each group a “test.” Whenever 2 or 3 people prefer Pepsi that is one test where “the majority of Coke lovers preferred the taste of Pepsi.” So how many do I need to make my tagline true? Only two, even if in 2000 cases it’s not true. > > Beautiful. An apparently compelling stat that tells you nothing at all. It’s even better. I got this from a stats textbook: the Pepsi people labelled all of the cups containing Pepsi with “M”, and all of the cups containing Coke with “Q”. An independent test showed that people preferred the drink in cups labelled M over those labelled Q when _both cups had the same drink in them_. At a rate greater than what Pepsi was getting. ---
**Author:** ![billfish678](https://avatars.discourse-cdn.com/v4/letter/b/7bcc69/32.png) [@billfish678](https://boards.straightdope.com/u/billfish678)\ **Post date:** [July 9, 2012, 6:21pm UTC](https://boards.straightdope.com/t/lying-with-statistics/627453/40 "2012-07-09T18:21:06Z")
> [@mhendo](#): > > As far as i know, the official federal poverty line in America is, and has been for quite some time, a dollar value. Or, more accurately, a set of values, depending on how many people are in your family. The 2012 figures are listed [here](http://aspe.hhs.gov/poverty/12poverty.shtml/). > > As you can see, there are actually two different versions of the poverty measure: a set of thresholds used by the census bureau for statistical purposes, and a set of guidelines used by government departments for administrative purposes. The two sets of numbers, while reasonably close, are not identical. The 2011 Census thresholds can be found in [this file](http://www.census.gov/hhes/www/poverty/data/threshld/thresh11.xls) (Excel spreadsheet). > . Thanks for the info. But you are missing my point. The article made a big deal about X amount of people (or people earning less than X) being in poverty. But it was the definition they made big deal about when it could not by definition be any other way. Now, if the article had made a big deal about something along the lines of “how does a family of Y making less than X make ends meet?!” that would be something else. But thats not what the article did. Did you know that a family of 4 earning less than 30K lives in “poverty”?! Btw did you know the definition of poverty is 4 living on less than 30k? Well, no shit Sherlock. [Previous page](https://boards.straightdope.com/t/lying-with-statistics/627453.md?page=1) [Next page](https://boards.straightdope.com/t/lying-with-statistics/627453.md?page=3)