Boards
user ages:
age: number of users
14: 471
15: 772
16*: 1094
17: 1031
18*: 1191
19: 887
20: 796
21: 792
22: 660
23: 648
24: 677
*= ages at which names beginning with "-" are particulalr common
no disrespect is meant to anyone outside of the 14-24 age bracket, but you can imagine numbers dwindle in both directions.
before you ask, this took about 6 minutes, and i found it interesting.
14: 471
15: 772
16*: 1094
17: 1031
18*: 1191
19: 887
20: 796
21: 792
22: 660
23: 648
24: 677
*= ages at which names beginning with "-" are particulalr common
no disrespect is meant to anyone outside of the 14-24 age bracket, but you can imagine numbers dwindle in both directions.
before you ask, this took about 6 minutes, and i found it interesting.
Re: user ages:
Re: user ages:
Re: user ages:
Why did you specifically look at names beginning with "-", out of interest?
Re: user ages:
i used your data to plot a histogram. it shows a non-normal distribution of ages, supporting my theory that most people over 24 don't give their ages, and that the real median age may be higher than suggested using only those ages that are reported.
if anyone cares to discuss further statistical tests that might be carried out on the data then please go and do something more worthwhile with your day, like i'm going to do now that i've realised what a pathetic square i am becoming.
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
then it is interesting
otherwise yo jus saying there's a thousand 16 year olds here
percentages tell the real tales innit
Re: user ages:
Re: user ages:
*sets self on fire and jumps out of the window*
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
+------+-------------+
| age | histo |
+------+-------------+
| 12 | |
| 13 | ** |
| 14 | **** |
| 15 | ******* |
| 16 | ********** |
| 17 | ********** |
| 18 | *********** |
| 19 | ******** |
| 20 | ******* |
| 21 | ******* |
| 22 | ****** |
| 23 | ****** |
| 24 | ****** |
| 25 | ***** |
| 26 | **** |
| 27 | *** |
| 28 | ** |
| 29 | * |
| 30 | ** |
| 31 | * |
| 32 | * |
| 33 | |
| 34 | |
+------+-------------+
There's a lot that won't admit their age too.
Re: user ages:
Re: user ages:
The statistician in me is inclined to apply some kind of smooth transformation to the age variable to reflect 'stage of musical development' rather than chronological age, leading to something that'd probably look a lot more normal and pass various goodness of fit tests for Normal-ness.
Of course that's just subjective handwaving but most statistics is, unless you have a really convincing mathematical model for the distribution of observations you're making, which most of the time you don't. Not my favourite bit of maths by a long way.
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
That said, I spent less time on it than the original guy did (mainly cos I have direct access to the database but yeah).
Statistics is one of the most dull, inelegant, messy and utilitarian areas of maths, so yeah it is a bit sad unless you have interesting uses for it (like, say, our recommendations system that's all based on statistics). Maths in general is not 'sad' though.
Re: user ages:
cant remember a fucking bit of it now.
i enjoy(ed) it though.
Re: user ages:
Some kids at another school had to do an 'experiment' or something for coursework for their stats A-level, so they studied the deterioration in their dart-throwing accuracy with increasing alcohol intake. :) So just a big excuse for going to old man pubs and playing darts, they wrote it up properly and all.
Re: user ages:
Re: user ages:
Re: user ages:
i may have to cut and paste that into my user profile.
that is so cool.
Re: user ages:
;-)
Re: user ages:
*feels sorry to disappoint tom*
Re: user ages:
;-) :-) :-D
Re: user ages:
Re: user ages:
Re: user ages:
I think if you make a highly dubious simplifying assumption that people's paired ratings follow bivariate normal distribution, you could use this to find a confidence interval: eek but frankly the sight of that page makes me sick. Coming up with a really convincing statistical model for the relationships between pairs of artist ratings would be pretty hard I'd imagine. Clustering and factor-based analysis are better approaches but also a lot more complex - I went for something simple and elegant :)
I found statistics bored the pants off me at uni, and covered it from a very mathematical point of view so you probably know more than me about the practise of it.
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
Re: user ages:
i would do the rock-on sign if it didn't look so undeniably shite in ascii form.
Re: user ages:
Re: user ages:
Re: user ages:
But to get back to me....
I, being self-absorbed am, only concerned with how many 44 yr olds or older are on the website....i.e. is I the grandaddy? (I believe Grim Jim may be a rival).
You have missed my demographic from your figures.
(seems to be the story of my life)
Re: user ages:
USERS
then go to SEARCH USER PROFILES
then type 44 into AGE, and see how many results come up.
and there are 21 who lisrt themselves as 44.