Forum Statistics

General out-of-character discussion among players of Cantr II.

Moderators: Public Relations Department, Players Department

returner
Posts: 948
Joined: Sun Nov 01, 2009 8:11 am
Location: Melbourne, Australia

Forum Statistics

Postby returner » Mon Apr 19, 2010 3:11 pm

So, in my boredom, I've decided to try out some software which performs different types of statistical analysis on forums.

I'll continue to post interesting and relevant statistics in this thread, which will mainly assist with the administration of the game and provide new perspectives on the public operation of the game.

While the idea came from Seko's frequent posting of statistics within Cantr, my intention is very different. I also will work with public data which is somewhat harder to interpret and use effectively. In a similar attempt to Seko, I will accept ideas for statistics.

The first statistic relates to the author of posts in the 'Accepted Suggestions' thread, and the frequency of their accepted posts. The collected data begins in 2007 and ends in 2010. Any author with less than 2 accepted suggestions is considered statistically insignificant and thus not shown in the data.
In descending order of the number of accepted suggestions:

6 - SekoETC
6 - formerly known as hf
4 - Doug R.
3 - Nakranoth
3 - T-shirt
3 - Cookie
2 - fishfin
2 - Agar
2 - Ahoyhoy
2 - Pie
2 - Cogliostro
2 - Marian
2 - *Wiro
2 - SCUBA
2 - The Industriallist
2 - TatteredShoeLace

Interestingly;
When we look at the type of author (either ex/current staff member OR lay member), 43.75% members on the list are staff (this is a conservative figure based partly on my own knowledge and partly on what role the forum says they have), whereas 56.25% of members on the list are lay, normal members.
This is good, INITIALLY it shows that general members of Cantr have an equal chance of getting their ideas accepted by staff.

However, when we look at the author type AND the number of accepted suggestions they have, we see a greater chance for a staff member to have his or her suggestion accepted.
Basically, if you are a staff member, your suggestion is more likely to be accepted but the difference is hardly significant.
53.33% of accepted suggestions are by authors who are 1) ex/current staff and 2) have more than 1 accepted suggestion,
46.66% of accepted suggestions are by authors who are 1) normal members and 2) have more than 1 accepted suggestion

I'm both surprised at the result and happy with the result.
Even though the data and the means of interpreting it are shallow and not very accurate (in the sense that posts in Accepted are frequently moved to Implemented, and authors with less than 2 accepted posts aren't counted), it still gives a good indication that the staff members who are accepting suggestions are doing so in a fair and reasonable manner and are not giving more value or weight to staff member posts.
My surprise is based on the expectation that we'd see a higher number of staff-accepted posts, and my happiness is attributed to the fact that this is not the case.






More to come. :mrgreen:
This account is no longer active - please send any PMs to my new one.
User avatar
EchoMan
Posts: 7768
Joined: Fri Aug 26, 2005 1:01 pm
Location: Stockholm, Sweden

Re: Forum Statistics

Postby EchoMan » Mon Apr 19, 2010 3:18 pm

How many percent of a forum users total posts in suggestions, accepted and rejected ended upp in accepted?

A much better number for this kind of statistics. Lets say Seko have posted 1000 ideas and "only" 6 made it to accepted, while a normal user got both of his accepted. Does then staff have it easier to get a suggestion accepted?

If you are to conclude any kind of statistical facts about things, please make sure you are measuring the right values. :)

One could also assume that a staff members suggestion could be more sensible, more thought through and thus have a higher chance of getting accepted, because staff normally have a higher understanding of the game than "normal" users do in general.
User avatar
Pirate Lass
Posts: 845
Joined: Wed Aug 20, 2003 4:05 pm
Location: Portage, Indiana
Contact:

Re: Forum Statistics

Postby Pirate Lass » Mon Apr 19, 2010 3:24 pm

Of course most current staff members were normal players first and due to their desire to be active and find new ways to further the game (and submitting ideas), they either applied or were selected to be staff.

So I wouldn't look at it as staff suggestions possibly carrying more weight - but that the most active people with ideas have moved into the staff category.

Heck, I remember when most were not on staff. LOL :lol:
returner
Posts: 948
Joined: Sun Nov 01, 2009 8:11 am
Location: Melbourne, Australia

Re: Forum Statistics

Postby returner » Mon Apr 19, 2010 3:41 pm

To EchoMan, I don't really claim to state statistical fact, only a rough idea based on what I have available.. You're absolutely right, but throughout the post I do make mention to 'statistical insignificance'. This was really just a warmup and only provides a little bit of insight.

If I had all the data available, I'd incorporate so much more into it..
Perhaps that will be my next attempt. I'll try to look at:
- Date of registration (ie, how long they've been around)
- Number of posts (ie, activeness)
- Number of posts per day - this is pretty much a summary of the first two points (registration date / number of posts)
- Number of suggestions, including Suggestions, Accepted/Implemented, Rejected
- Type of author (ex/current staff, lay members)
(The reason why I believe ex/current staff have greater weight is due to their attained and inherent respect through being a staff member)

Then I will try to create a rank for the author, from 1 to 6, based on 'respectability' and activeness within the community. 1 refers to someone who is inactive through posts, a 'new'ish member and is not staff. 6 would be someone who has many posts, an older member and is staff.
Once we have that rank, we can see for each rank, what the likelihood of their suggestion being accepted is.
ie,
Rank 1: 5% chance of being accepted
Rank 2: 10% chance of being accepted
Rank 3: 20% chance of being accepted
Rank 4: 30% chance of being accepted
Rank 5: 40% chance of being accepted
Rank 6: 50% chance of being accepted

One issue with this, is that it ignores the value of the suggestion. Some new guy may suggest something which goes against the culture of Cantr players, such as a machine gun or a fighter jet or space exploration. Despite this, it still gives a better indication than what my first post did.
Actually, to refute my own criticism, if someone has a low rank, then it would be expected that they do not have a keen understanding of the game and it's culture so naturally, due to low respect and activity, they're more likely to get rejected.


The overall intention of this is to test whether there is a correlation between your 'rank' in the community and the likelihood of your ideas having merit and being accepted.
This account is no longer active - please send any PMs to my new one.
User avatar
Doug R.
Posts: 14857
Joined: Wed Mar 23, 2005 6:56 pm
Contact:

Re: Forum Statistics

Postby Doug R. » Mon Apr 19, 2010 3:51 pm

This also doesn't take into account the fact that the original author's suggestion may not be what was accepted. And I just wanted to restress what EchoMan said - Between Seko and I, we've probably posted more suggestions than the next 5 top posters combined.
Hamsters is nice. ~Kaylee, Firefly
AlchemicRaker
Posts: 311
Joined: Thu Feb 05, 2009 4:56 am

Re: Forum Statistics

Postby AlchemicRaker » Mon Apr 19, 2010 3:53 pm

...and I think we don't need any sort of ranking system to start segregating members by some sort of 'betterness'...
returner
Posts: 948
Joined: Sun Nov 01, 2009 8:11 am
Location: Melbourne, Australia

Re: Forum Statistics

Postby returner » Mon Apr 19, 2010 4:25 pm

Doug R. wrote:This also doesn't take into account the fact that the original author's suggestion may not be what was accepted.


Indeed, in many cases it is a variation of the original suggestion. If completely different, then a new thread is made. But the title of the thread is USUALLY an indication of what the accepted suggestion actually is, and usually the final posts relate to the original suggestion. It still has statistical significance, and thus I would mention this issue in the conclusion of the report. I would specifically say that the data is not based on the original suggestion itself, but rather the ultimate suggestion at the end caused/invoked/started/inspired by the original author.

Doug R. wrote:And I just wanted to restress what EchoMan said - Between Seko and I, we've probably posted more suggestions than the next 5 top posters combined.


And I'll restress what I said in response to EchoMan which I'm guessing you didn't read: the upcoming statistic would take this into effect, through the number of suggestions the author made.

Natso wrote:...and I think we don't need any sort of ranking system to start segregating members by some sort of 'betterness'...


Huh? It's not an official rank, it's a rank to deindividualise and anomise members, (the absolute opposite of what you imply it does) so that the statistic can be carried out. It's funny you interpreted it that way, though. It certainly has nothing to do with betterness; rather, it represents the level of involvement within the community. And I certainly don't want to see every member labelled as a rank. It's just to use for my statistics...

It's frustrating that you make posts that seem like 'arguments against this statistic', when in reality they are posts which are enlightening me to potential statistical hiccups. In fact, every post you make provides another calculation to perform on the statistic, making it more and more accurate. So, in hindsight, frustration is probably not the response I need despite the negative connotations inherent in your posts.. instead, I'll appreciate each post, negative or not, and take it into effect in formulating the final statistics and thus conclusions.
This account is no longer active - please send any PMs to my new one.
User avatar
SekoETC
Posts: 15526
Joined: Wed May 05, 2004 11:07 am
Location: Finland
Contact:

Re: Forum Statistics

Postby SekoETC » Mon Apr 19, 2010 5:56 pm

Might want to check how many I have in Rejected too.
Not-so-sad panda
User avatar
Piscator
Administrator Emeritus
Posts: 6843
Joined: Sun Jul 02, 2006 4:06 pm
Location: Known Space

Re: Forum Statistics

Postby Piscator » Mon Apr 19, 2010 6:09 pm

I checked and apparently I only ever started 5 suggestion threads, one of them accepted and another one implemented. That last one I implemented myself though, so I guess I'm distorting the statistics. ;)
Pretty in pink.
User avatar
Cdls
Posts: 4204
Joined: Sat May 14, 2005 7:09 pm

Re: Forum Statistics

Postby Cdls » Mon Apr 19, 2010 6:19 pm

Your stats also fail to take into account that staff members usually more articulate and descriptive in their suggestions, not to mention they post suggestions that actually have a chance of making it into the game...unlike others who would rather suggest such things as a button to 'slap a b**ch' or...suggesting to make animals evil.
User avatar
Bowser
Posts: 1201
Joined: Thu Dec 04, 2003 8:55 pm
Location: Washington, D.C.

Re: Forum Statistics

Postby Bowser » Mon Apr 19, 2010 6:27 pm

returner wrote:Any author with less than 2 accepted suggestions is considered statistically insignificant and thus not shown in the data.


So sad when a person is labeled insigificant, even if it is just statistically speaking.
Homer wrote: "Weaseling out of things is important to learn. It's what separates us from the animals ... except the weasel. "
User avatar
EchoMan
Posts: 7768
Joined: Fri Aug 26, 2005 1:01 pm
Location: Stockholm, Sweden

Re: Forum Statistics

Postby EchoMan » Mon Apr 19, 2010 9:09 pm

Cdls wrote:Your stats also fail to take into account that staff members usually more articulate and descriptive in their suggestions, not to mention they post suggestions that actually have a chance of making it into the game...unlike others who would rather suggest such things as a button to 'slap a b**ch' or...suggesting to make animals evil.


I'm assuming yugo's suggestions were edited out of the results... or? Dangit, did you let yugos flametrollfloodrantinfg contaminate your results? Time to do it all over again. :)
returner
Posts: 948
Joined: Sun Nov 01, 2009 8:11 am
Location: Melbourne, Australia

Re: Forum Statistics

Postby returner » Tue Apr 20, 2010 2:24 am

Cdls wrote:Your stats also fail to take into account that staff members usually more articulate and descriptive in their suggestions, not to mention they post suggestions that actually have a chance of making it into the game...unlike others who would rather suggest such things as a button to 'slap a b**ch' or...suggesting to make animals evil.


Hahaha!

Well, they would be rank 1, and that doesn't matter. You would expect rank 1 and rank 2 to have a very low percentage of accepted suggestions.

The real comparison would come from rank 3 and 4, versus rank 5 and 6. Meaning, we are looking at formidable members at 3 and 4 against respected and staff members at 5 and 6.
Personally I'd expect to see 5 and 6 having a greater chance of acceptance, whereas 3 and 4 have a lower chance.
If there is an increase in the chance of the suggestion being accepted based on rank, we can say two different things. One, the person who suggested it has a keen understanding of Cantr culture so OF COURSE they'll be accepted - there is nothing interesting in this result. The other conclusion is that those who have respect ARE being accepted more.
You still won't be able to differentiate between the two, as it comes down to a subjective opinion on whether the suggestion is good or bad.

From that statistic, you would have to look at the aim of the game, compared to a consensus of 1) what direction players believe the game is SUPPOSED to be going, and 2) what direction the player WANTS the game to go. Specifically, role-playing or simulation.
Then, we look at the suggestion and scale it from 1 to.... 6? (lol), where 1 is a suggestion that strongly tastes like a role-playing implementation, and 6 feels like a simulation implementation.
Once we have this scale of the suggestion, we compare it to the player consensus and we can ascertain whether the suggestion is 'good' or 'bad' and the subjectiveness re: opinion on suggestions becomes less of an issue.

Then this statistic can be combined with the previously mentioned/proposed original statistic, to gain a very accurate perception of the following: Do good suggestions have less merit if they are suggested by a person with low rank.
The key components of this question are enlarged for explanation and source and definition:
'Good': Defined through the statistic of the consensus of player opinion on the direction Cantr is/should take.
'Less Merit': Defined through being rejected/still in suggestions thread (ie not accepted/implemented)
'Low Rank': Defined through previously mentioned statistic (rank from 1 to 6).

Upon review, it's a lot of work for little output, and much of the work would be manual. The potential benefits include looking at new ways to suggest things, including through anonymity or player-voted. To expand on this, if the result did say that 'yes, the suggestion was good but it was suggested poorly and by a low-rank member', then we would need to look at alternative ways of improving Cantr through suggestions.. Ie, you post a suggestion anonymously; to be reviewed by GAB/GAC (whoever handles it), OR, all players are sent a message in Cantr in which they can vote on a new 'large' implementation, and are also provided with a paragraph in favour and against the suggestion. (this is a better method I believe).
This account is no longer active - please send any PMs to my new one.
User avatar
Dudel
Posts: 3302
Joined: Wed Oct 01, 2008 5:21 am

Re: Forum Statistics

Postby Dudel » Sun Apr 25, 2010 9:18 am

On a personal level I disregard more staff suggestions than "normal user suggestions", then again there are a lot more of them. :lol:

It sounds, to me, like you just want to validate either your own suggestions or someone else's with the common argument of "Well excuse me I'm not staff and blah, blah, blah, blah."

Someone with "less merit" is also a PISS POOR way to organize people as someone like yourself or Colo(whatever-the-hell-his-name-is) come up with good ideas but at the same time FLOOD suggestions with so much crap, people stop caring and disregard you for that reason. Such a thing has NOTHING to do with being a staff member or post count... or anything other than your own zealousness.

I think I've got one Suggestion, maybe two in total, and neither are honest suggestions. One is a duplicate and the other is my K.I.S.S. thread which is a suggestion about suggestions (as people were bogging things down to the point where the suggestions section became more of a chore than it should be... at least for non-staff.)

Not to mention that the current "state-of-Cantr" prevents a lot of ideas from being implemented simply based player mentality. So a "good idea" will be ignored REGARDLESS of who suggested it because players scoff at the thought.

Oh and your "Roleplaying vs Simulation" mission wont get far as you could tell people to their face what they want isn't what they actually want and they'll still try and steer you in other directions. You're fighting an uphill battle and the hill's angle is almost a ninety degree angle.

In the end of things, you missed your mark and are now trying to justify an action which has no point to being.
returner
Posts: 948
Joined: Sun Nov 01, 2009 8:11 am
Location: Melbourne, Australia

Re: Forum Statistics

Postby returner » Mon Apr 26, 2010 10:46 am

I've abandoned this project anyway. Dudel, your assumption of my motivation is entirely incorrect, and claims you make seem a little far-fetched (ie you said that I make some good suggestions here and there - thankyou - but you say I make a lot, so people stop caring about suggestions and disregard me..? Utterly untrue and farfetched)
Then you said it has nothing to do with post count or staff member'ness, but you forget there is another variable I included - the quality of post, which discredits your next points as well.
I do want to validate my own suggestions. So do you. So does every single person on this board, unless they realise that their suggestion won't work or they post a joke suggestion.
But that's not the intention of the poll. The intention of the poll was to see if low-rank users = low-rank suggestions. To see if there is a correlation between your (not YOU, but people in general) activeness and understanding of the community, and the likelihood of being accepted.
It has nothing to do with anything you seem to think it does. :wink:

The ultimate result of having a poll like this, is so we can reflect on our current suggestions system and perhaps look at new ones.

One result I had in mind, was to have suggestions EMAILED to staff (which dismantles what you think I am trying to do), so STAFF ONLY can discuss it and accept/decline it.

Then after it's accepted, I was thinking that a player-wide/game-wide poll is posted on the Character Menu so everyone can vote on it's implementation.

A fairer system, with less dribble and chit-chat, and more professionalism, efficiency and speed of implementation.



My original poll would help to establish whether or not this 'Suggestion Revolution' is a good idea, or at least back it up.
This account is no longer active - please send any PMs to my new one.

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest