+ Reply to Thread
Results 1 to 2 of 2

Thread: Null hypothesis for Roman dice

  1. #1
    Points: 11, Level: 1
    Level completed: 21%, Points required for next Level: 39

    Thanked 0 Times in 0 Posts

    Null hypothesis for Roman dice

    Iím museum curator working on a project thatís looking for evidence of bias in Roman dice. This involves me rolling each dice a few hundred times and carrying out a chi-square test on the results.

    So far Iíve been working with this null hypothesis: ďThe die is equally likely to land on any of its facesĒ. Recently Iíve started to worry that this isnít the right approach, and Iím wondering if thereís anyone here who can help.

    Iím not a statistician, but my understanding is that the null hypothesis should assume that the data is random (as Iíve assumed above). But, given that my Roman dice are handmade and obviously imperfect, itís actually more likely that they are unbalanced than not Ė ie, I ought to be surprised if the dice DO produce random results!

    Does this mean that my null hypothesis should be: ďThe die is unlikely to land equally on each faceĒ and that I'm actually trying to prove a lack of bias?

    Any thoughts much appreciated.

  2. #2
    TS Contributor
    Points: 12,287, Level: 72
    Level completed: 60%, Points required for next Level: 163
    rogojel's Avatar
    I work in Europe, live in Hungary
    Thanked 332 Times in 312 Posts

    Re: Null hypothesis for Roman dice

    I do not think you should give up the null hypothesis - just consider how many different alternative hypothesises you have. Large bias for side 1, smaller bias for side 1 ....etc.
    Also, your research question is fully compatible with this choice of the null hypothesis.


+ 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