# Stats question **URL:** \ **Category:** Factual Questions\ **Created:** [October 27, 2015, 10:25am UTC](https://boards.straightdope.com/t/stats-question/735859 "2015-10-27T10:25:22Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![blueslipper](https://avatars.discourse-cdn.com/v4/letter/b/b5a626/32.png) [@blueslipper](https://boards.straightdope.com/u/blueslipper)\ **Post date:** [October 27, 2015, 10:25am UTC](https://boards.straightdope.com/t/stats-question/735859/1 "2015-10-27T10:25:22Z")
> **[oOh!media | Out of Home Advertising](https://oohmedia.com.au)** > > oOh! is Australia and New Zealand’s leading Out of Home advertising company. With a network of over 35,000+ locations including billboards, bus shelters, airports, shopping centres and train stations, oOh! can make your brand Unmissable. Inlink Boosts Intent to visit Websites by 26%. Without Inlink Office 50% With Inlink Office 63% This is not a 26% increase. That is a fundamental error in how stats are done. Am I right? If I am correct, how do I explain this to the Inlink people? ---
**Author:** ![kk\_fusion](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/kk_fusion/32/15276_2.png) [@kk\_fusion](https://boards.straightdope.com/u/kk_fusion)\ **Post date:** [October 27, 2015, 10:38am UTC](https://boards.straightdope.com/t/stats-question/735859/2 "2015-10-27T10:38:41Z")
The way I understand it, 50 % of people surveyed preferred getting their infomation from the website, say 500 out of 1000. After the introduction of the advertised product, 630 people chose the website. 630 is 26 percent more than 500. The percentage _share_ only increased by 13 though. The stats seem correct but poorly presented. ---
**Author:** ![Clark\_Cello](https://avatars.discourse-cdn.com/v4/letter/c/58f4c7/32.png) [@Clark\_Cello](https://boards.straightdope.com/u/Clark_Cello)\ **Post date:** [October 27, 2015, 11:40am UTC](https://boards.straightdope.com/t/stats-question/735859/3 "2015-10-27T11:40:38Z")
> [@blueslipper](#): > > [Home](http://www.inlink.com.au/powerof3/#next) > Inlink Boosts Intent to visit Websites by 26%. > > Without Inlink Office 50% > With Inlink Office 63% > > This is not a 26% increase. That is a fundamental error in how stats are done. > Am I right? If I am correct, how do I explain this to the Inlink people? That is indeed an increase of 26%. It is also an increase of 13 percentage points (or various other terminologies). ---
**Author:** ![RedSwinglineOne](https://avatars.discourse-cdn.com/v4/letter/r/a9adbd/32.png) [@RedSwinglineOne](https://boards.straightdope.com/u/RedSwinglineOne)\ **Post date:** [October 27, 2015, 10:01pm UTC](https://boards.straightdope.com/t/stats-question/735859/4 "2015-10-27T22:01:27Z")
An increase from 50 to 63 of anything is an increase of 26%. It just confuses matters that the units in question happen to also be %. ---
**Author:** ![ddunn801](https://avatars.discourse-cdn.com/v4/letter/d/71c47a/32.png) [@ddunn801](https://boards.straightdope.com/u/ddunn801)\ **Post date:** [October 28, 2015, 7:56pm UTC](https://boards.straightdope.com/t/stats-question/735859/5 "2015-10-28T19:56:45Z")
It is correct, but could be confusing. Many people use basis points (bps) when describing absolute changes in percentages. For example, this is a 13 bps increase. ---
**Author:** ![Clark\_Cello](https://avatars.discourse-cdn.com/v4/letter/c/58f4c7/32.png) [@Clark\_Cello](https://boards.straightdope.com/u/Clark_Cello)\ **Post date:** [October 28, 2015, 8:25pm UTC](https://boards.straightdope.com/t/stats-question/735859/6 "2015-10-28T20:25:50Z")
> [@ddunn801](#): > > It is correct, but could be confusing. Many people use basis points (bps) when describing absolute changes in percentages. For example, this is a 13 bps increase. Actually, bps are hundredths of percents. So a change from 10.20% to 10.41% is a change of 21 bps. The change here (50% to 63%) is 1300 bps. ---
**Author:** ![Senegoid](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/senegoid/32/6606_2.png) [@Senegoid](https://boards.straightdope.com/u/Senegoid)\ **Post date:** [October 28, 2015, 11:10pm UTC](https://boards.straightdope.com/t/stats-question/735859/7 "2015-10-28T23:10:31Z")
There is no error here. This is marketing people who know how to lie with statistics, and do so regularly for a living. The statistic is correct, but they knowingly word it in ways that make it look, to the non-statistician, bigger and better than it is. Same as all those commercials that say things like “Bayer Aspirin reduced headaches _significantly_ more than Brand X”, using the word “significantly” in its specific technical meaning, knowing fully well that it’s not the same as the common everyday usage lay meaning.