|
Acassolopez
Nov 27, 2012, 5:23 PM
Post #1 of 11
(3115 views)
Shortcut
Registered: Jun 22, 2012
Posts: 4
|
Greetings fellow climbers, I am doing a project for my statistics class on whether gender affects climbing ability. Here's the thing, it's not very scientific I just have to show I learned something. All I need is the average grade you're climbing at on top rope (YDS grading please) and your gender. For those who care I'm going to do a chi squared test for dependency. (If I were taking more variables into consideration I would probably do multiple regression) Thanks, Ashley UPDATE: after talking to my teacher and bringing up everyone's input, we decided to use the 8a.nu database to do a multiple regression on the top ten professional climbers in both genders. Thank you amarius for the suggestion. This has been an interesting learning process. Thank you all for your input.
(This post was edited by Acassolopez on Nov 27, 2012, 11:57 PM)
|
|
|
|
|
NJSlacker
Nov 27, 2012, 5:31 PM
Post #2 of 11
(3106 views)
Shortcut
Registered: Feb 17, 2008
Posts: 212
|
so, you've biased your sample by having non-random subjects volunteer their own data. The fact that it's semi-public will also skew results (most likely to have people grade themselves higher than if it was anonymous). I climb 10d toprope, male.
|
|
|
|
|
amarius
Nov 27, 2012, 5:36 PM
Post #3 of 11
(3100 views)
Shortcut
Registered: Feb 23, 2012
Posts: 122
|
I recommend contacting the good folks at http://www.8a.nu - they have a searchable database of climbers who track their progress.
|
|
|
|
|
Acassolopez
Nov 27, 2012, 5:38 PM
Post #4 of 11
(3099 views)
Shortcut
Registered: Jun 22, 2012
Posts: 4
|
You're totally correct. The data I already have is skewed. However, I am not being graded on the outcome of the project just the performing of it. I plan on discussing the many flaws of my project in my presentation. Thanks for your input.
|
|
|
|
|
LostinMaine
Nov 27, 2012, 5:44 PM
Post #5 of 11
(3090 views)
Shortcut
Registered: May 8, 2007
Posts: 539
|
Are you sure you want average grade of top rope? That won't be a fair indication of ability, simply just a central tendency measure of all climbs one has done on a top rope. Based on the limited sample size you will get, the range will mask any mean differences of the sample population (i.e. gender won't show up as a factor unless you have a huge sample size, I would guess). Why not just do a 2-sample t-test? Oh.. male, 5.10a might be my average...
|
|
|
|
|
Acassolopez
Nov 27, 2012, 5:53 PM
Post #6 of 11
(3078 views)
Shortcut
Registered: Jun 22, 2012
Posts: 4
|
Because of time constraints yes. When I get off for winter break I plan on doing more research and gathering more variables. I'm not doing a 2 sample t test because we went over that the semester before and I have to use something we learned this semester. I know nothing significant will show up in this test because of my limited sample size.
|
|
|
|
|
rsd212
Nov 27, 2012, 5:58 PM
Post #7 of 11
(3070 views)
Shortcut
Registered: May 12, 2008
Posts: 95
|
What is even meant by average grade on top rope? Average of all top-rope climbs I've ever done? Average grade of the climbs I sent on top-rope in the gym on recent visits? Do I include the 20 laps of easy climbing I did for endurance? Are you just looking for the grade that if someone asked me "can you send this on top rope?" I'd say "yeah, probably, lets give it a go"?
|
|
|
|
|
Acassolopez
Nov 27, 2012, 6:07 PM
Post #8 of 11
(3063 views)
Shortcut
Registered: Jun 22, 2012
Posts: 4
|
What I mean by that is what kind of routes are you working on? It's really just a personal estimate on what you're comfortable with at this present time.
|
|
|
|
|
petsfed
Nov 27, 2012, 7:30 PM
Post #9 of 11
(3003 views)
Shortcut
Registered: Sep 25, 2002
Posts: 8599
|
male 5.10c Since gender is (mostly) binary, I expect that, at best, you'll see a small difference between population means, like probably smaller than sum over quadrature of their standard deviations. In my world (physics), that means that there is no difference. Moreover, since you're not also measuring (for instance) how long the person has been climbing, real differences between genders will be impossible to show. Were I you, I'd also ask how long people have been climbing, and figure out the experience-grade relation and show how that varies with gender, if you can.
|
|
|
|
|
satch
Nov 28, 2012, 2:26 AM
Post #10 of 11
(2926 views)
Shortcut
Registered: Mar 28, 2005
Posts: 94
|
male - 11b my wife: female - 11c Guess that may make it harder to reject the null hypothesis.
|
|
|
|
|
jeepnphreak
Nov 28, 2012, 5:50 PM
Post #11 of 11
(2863 views)
Shortcut
Registered: Jul 29, 2008
Posts: 1259
|
Me- after blown rotator cuff, 10b/c and getting better. wife 5.11 b/c
|
|
|
|
|
|