Not signed in (Sign In)

Vanilla 1.1.2 is a product of Lussumo. More Information: Documentation, Community Support.

    • CommentAuthorLSK
    • CommentTimeMar 5th 2006
     

    I read somewhere that a certain percentage of statistics are made up. I do not remember this amount. It would be helpful if somebody could tell me this rate. Thanks in advance!

    •  
      CommentAuthorZack
    • CommentTimeMar 6th 2006
     

    It is believed that 63% of all statistics are somehow false. Of that number, 32% are simple mistakes in calculation, 29% are presented as representing an unrelated variable, and 2% are completely fabricated.

  1.  

    Those are the numbers I saw in a Southern Michigan University study.

    • CommentAuthorCrushthor
    • CommentTimeMar 16th 2006
     
    I have been banned

    If 2% are completely fabricated, then there is a chance that the Southern Michigan University study is completely fabricated!

  2.  

    Wow. What an astute observation.

    • CommentAuthormcdan333
    • CommentTimeApr 3rd 2006
     

    Southern Michigan University has one of the most highly respected pseudo-statistics departments in the world. Those study results were peer-reviewed in several leading academic publications, including The Journal of Ersatz Mathematics. If they say 2% of statistics are completely fabricated, you can trust that they are correct.

    •  
      CommentAuthorhitchhiker
    • CommentTimeJun 15th 2006
     

    actually the percentage of made-up statistics is always rising and falling because of all the new statistics occuring and the new ones being made up, the rate of which can very up to 60% difference.

    •  
      CommentAuthorMrFingers
    • CommentTimeJun 15th 2006 edited
     

    The problem isn't so much that the statistics are completely made up, than the fact that often the data is collected and presented by an entity with a biased agenda. So, often critical counter-data will be ignored and the data which is displayed will be done so in a misleading or exaggerating manner.

    For example, I want to show you that the US dollar is depreciating as much as I possibly can. I chose a currency to compare it against that has been appreciating alot to distort/exaggerate the changes in the value of the dollar. Then I display the data on a graph where the origin is not at zero, but at where the dollar price is recently lowest on the 'y axis' and the most inappropriate and misleading time frame on the 'x axis' - this way I make the dollar appear to be totally imploding all the way to the bottom of the graph by magnifying both the data and the graphics.

    Statistics is a dirty and psychological business. Along with a shit-load of other professions lol. Here is an interesting and true statistic though;
    There has been more data collected on the human race in the last decade than the entire rest of human history put together.