+ Reply to Thread
Results 1 to 5 of 5

Thread: Confirmation bias.

  1. #1
    Points: 10, Level: 1
    Level completed: 19%, Points required for next Level: 40

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Confirmation bias.




    Hello.

    I am interested in confirmation bias. Specifically I am interested in when a hypotheses can be said to be no longer confirmation bias.

    eg Let's say you think an online blackjack game is rigged. According to statistics, if you are holding 18, and the dealer is showing 7, your odds of winning that hand are 87%. If you believe that your odds in a particular game are being fixed so that the dealer is winning more than 13% of the time, each occasion that you lose in this situation, it can be attributed to confirmation bias, rather than support of the rigged hypotheses. (is this right?)

    So my question is, how much data do you need to collect before you can begin to, or even completely discount, confirmation bias? Can you collect enough data to ever discount it?

    thankyou.

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,185 Times in 1,146 Posts

    Re: Confirmation bias.

    How familiar are you with statistics? Do you know what level of significance (i.e., alpha) is? If you set your alpha and power then you can do a sample size calculation. More or less you have to state what level would be acceptable to reject you null hypothesis when it is try or accept it when it is false.
    Stop cowardice, ban guns!

  3. #3
    Points: 10, Level: 1
    Level completed: 19%, Points required for next Level: 40

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Confirmation bias.

    Not very familiar at all. But now I have some new things to look up. Let me understand what you replied, and I will have some more questions I expect.

    ty.

  4. #4
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,185 Times in 1,146 Posts

    Re: Confirmation bias.

    Great. You are looking at a probability problem as you can see based on reporting 87%. However, if you want to test a hypothesis you need to set some information. The easiest way to think about this is to simplify the question to say coin flips. We know heads or tails on a fair coin is 50:50. So we say is a coin unfair if we get 4 heads in a row? We could calculate the probability of 4 heads and see it is possible. If we collected 100 tosses we would get closer to 50% heads if it was a fair coin, though we could also get 99% heads. Law of big numbers says things converge to the true probability as the number of events increases. So how many hands do you need to be sure. Well a deck of cards is more complex. I will also point out that I enjoy these questions, but probability questions are not my specialty!
    Stop cowardice, ban guns!

  5. #5
    Points: 10, Level: 1
    Level completed: 19%, Points required for next Level: 40

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Confirmation bias.


    Actually what I want to test is more complex than a deck of cards. I just supplied that as a simple example. But now I am going to try to understand with a coin toss. That is a good idea ...

    (have you seen the explainations of Significance Levels (Alpha) and P values in Statistics... omg)

    I will get there.

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats