Results 1 to 2 of 2

Thread: statistical significance in birthdate bias theory

  1. #1
    Newbie
    Joined
    May 2019
    From
    Britain
    Posts
    1

    statistical significance in birthdate bias theory

    in its simplest form, birthdate bias theory says that when an athlete was born matters. that's date of birth, not time of birth.

    to test whether this applies to my sport I collected data on around 4000 athletes from a dozen countries, and...

    1. assigned them to birth categories based on the start of the scholastic year in their country
    2. calculated number and percentage of athletes in each category who achieved, "success" in the sport (note: "there are several definitions of "success" which we do not need to go into)
    3. this results in a table, an example of which I include here:

    code / event / sum(s) / win(s) / sum(m) / win(m) / solo / multi
    USA 400m 122 (44.4%) 153 (55.6%) 243 (41.9%) 337 (58.1%) 275 580
    USA 800m 133 (48.2%) 143 (51.8%) 293 (48.8%) 307 (51.2%) 276 600
    USA 1500m 139 (48.3%) 149 (51.7%) 343 (52.4%) 312 (47.6%) 288 655
    USA 5000m 149 (51.2%) 142 (48.8%) 300 (46.4%) 346 (53.6%) 291 646
    USA 10,000m 169 (52.0%) 156 (48.0%) 362 (53.0%) 321 (47.0%) 325 683
    USA total 712 (48.9%) 743 (51.1%) 1541 (48.7%) 1623 (51.3%) 1455 3164

    this table is of athletes reaching the final of the united states national championship
    the top row shows, from left to right, in the 400m, 122 finalists were born in summer, 153 finalists were born in winter. the summer-born athletes made 243 appearances in finals, the winter-born athletes made 337 appearances in the finals. a total of 275 athletes appeared in the finals of the 400m, and they made 580 appearances in those finals.

    my question is: how do I calculate whether any of these results are statistically significant ?

    I have tried to study this topic online and the process seems to be:

    1. guess what statistical significance would mean
    2. collect your data and run the numbers
    3. compare the data with your guess from step 1

    this does not strike me as being particularly rigorous, but I am prepared to admit that some part of, "guess what statistical significance would mean," may have eluded me.

    if anyone is able to offer advice on how to do this I would be very grateful.


    thank you.
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Junior Member
    Joined
    Aug 2011
    Posts
    31

    Re: statistical significance in birthdate bias theory

    Just estimate a maximum likelihood model, then use a Likelihood-ratio test, eg. Wilks' D test for significance testing.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Replies: 0
    Last Post: Apr 14th 2013, 04:10 PM
  2. Statistical Significance
    Posted in the Statistics Forum
    Replies: 1
    Last Post: Nov 14th 2009, 03:50 PM
  3. Statistical Significance
    Posted in the Advanced Statistics Forum
    Replies: 1
    Last Post: Aug 4th 2008, 02:07 AM
  4. Calculating Statistical Significance?
    Posted in the Advanced Statistics Forum
    Replies: 12
    Last Post: Jun 29th 2006, 10:59 PM

Search tags for this page

Click on a term to search for related topics.

Search Tags


/mathhelpforum @mathhelpforum