Welcome to WebmasterWorld Guest from

Forum Moderators: rogerd

Message Too Old, No Replies

Member Ranking System

Which factors would you use?

4:23 am on Dec 6, 2005 (gmt 0)

5+ Year Member

In the webmaster group on a community we've decided to start a Member Ranking System - (a bit PHPBB - however not at ALL).
we don't want contributors to stand out...
So anybody got a good formular on how to define/rate/rank what a good contributor is?

So far our cardinal focus has been:
(Comments given) / (Comments Received)

But which other factors would you include?

6:36 pm on Dec 6, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I have seen some forums that let the members rank you. Like a star system where if you really agree or approve a post, you can give it "5 stars." Nominal post would be 1 star. Could get out of hand though...
9:37 pm on Dec 6, 2005 (gmt 0)

5+ Year Member

I dont think that gets out of hand. I used to be a member of a forum like that. When you made a post that benifited the community, or looked like it took a lot of time/research people would rate you up. You started at an average rating of 3.0 and people could just vote on any post/reply you made. I think thats a great idea.

The only time people complained about it was when they made a statement about somthing controvertial... but thats just comon scence if you want your "karma" rating to be high.

9:58 pm on Dec 6, 2005 (gmt 0)

WebmasterWorld Administrator rogerd is a WebmasterWorld Top Contributor of All Time 10+ Year Member

Similarly, some software allows members to award another member "reputation points".

A few other imperfect but possible criteria:
- average length of posts
- average number of posts in threads started by member
- moderator-awarded points (i.e., mod sees someone being helpful or insightful and gives bonus points; perhaps not visible to members)

12:42 am on Dec 7, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I do post rating on my forum, and that contributes to rank. Controversies over unfair rating are inevitable, so expect that if you go that route. I've decided the pros outweight the cons, though.

Other things also contribute to rank... login regularity, age of account, use of certain functionality, etc. all feed into an internal point system.

Some experts argue hard against rank, calling it a distraction and making a "game" out of your forum. I happen to like it and think it builds community, personally. It also helps keep behavior under control, since if your account is deleted you lose your rank and start over if you try to sign up again. It's also a measure of trust, e.g. you might reward higher-rank members with additional functionality or greater ability to rate others, moderate, etc.

5:57 pm on Dec 7, 2005 (gmt 0)

10+ Year Member

Best implementation I've seen was a "thanks" button. The number of "thanks" was listed with each post, by where the member was identified by handle, postcount and avatar.

Posts: 2500
Thanks: 6

Posts: 25
Thanks: 6

Which member is more helpful?

8:51 pm on Dec 7, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I like the Thanks button scenario. I may actually try that on one of my threads. Good Idea!
10:43 pm on Dec 7, 2005 (gmt 0)

5+ Year Member

Regarding the member to member rating - I am personally not to keen on it - it has a lot of good points... Clearly it'll give the right reward..
But then again it is not something depending on the scenario - the key issue for the members should not be to rate eachother... but to give feedback to eachother!
1:18 am on Dec 8, 2005 (gmt 0)

WebmasterWorld Administrator rogerd is a WebmasterWorld Top Contributor of All Time 10+ Year Member

I like the "Thanks" button idea too. Very simple in concept, which is good.

Unless you have an exceptionally well-mannered and mature membership, though, I can imagine "click rings" of a few members who click each others buttons to inflate ratings.

You'd probably need some kind of tracking system anyway (to prevent self-clicks and repeated clicks by the same member), so a few thoughts for limiting the fraud potential come to mind:

- limit on "thanks clicks" per time period, e.g., no more than two per day
- limit on "thanks clicks" one individual can bestow on another individual, e.g., up to three. (Thus, if I have 30 points, one can be assured that 29 of them didn't come from a single buddy.)

1:54 am on Dec 8, 2005 (gmt 0)

WebmasterWorld Administrator webwork is a WebmasterWorld Top Contributor of All Time 10+ Year Member

Sticky from rogerd:

"This to to advise that your posting is being put on hold for being too thankful."

And I thought I was a meanie.

1:54 pm on Dec 8, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Combining this with something from the thread-splitting discussion, the busy board where I moderate (as opposed to my own tiny site board) uses reputation points, the kicker being that after you give someone a point, you have to give points to 20 other people before you can give one to that original person again. It doesn't force anyone to go outside their "comfort zone" and participate in other parts of the forum, but it does give the broader participants some recognition. And it definitely prevents a few friends from giving each other points repeatedly. (The site's vbulletin, and this system seems to be provided in the program.)
3:05 pm on Dec 8, 2005 (gmt 0)

WebmasterWorld Administrator webwork is a WebmasterWorld Top Contributor of All Time 10+ Year Member

Some really neat and interesting ideas posted so far.

In the democracy of point assignment what are the statistics for those who actively participate, those who sometimes or rarely participate and those who never participate?

I would think that IF participation stats are skewed then it would be problematic to limit the awarding of points or voting without consideration of the statistical spread regarding participation.

IF 50% don't bother to vote or do so rarely and another 20% only vote occasionally, then should the considerations about "over voting" for the remaining 30% rule the decision making process OR, if 30% are actually taking the time to take the extra step to show approval - WELL, then what?

IF the ratio of active voters to non-active voters is 1 active to 5 inactive/rarely then maybe instead of saying "you can only vote 1 in 20 times for the same person" perhaps it should be 1 in 5 times.

Sometimes it's only fair that if 20% of the people are actively engaged in whatever process that they should then be granted or assigned 80% of the "whatever". Don't limit the 20% by virtue of skewed stats that include the indifference of the other 80T%.

Of course, depending on how you model the voting rights system, you may find 80% not voting because they feel the system is broken from the start, such as unlimited voting allowing certain members to skew the voting.

"Oh, the troubles and labors of building a working democracy!" (From: "Memoir of an Ex Small Town Mayor", by Webwork.)

3:42 pm on Dec 8, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I would limit the clicks to one per post per member and not let a member click their own posts. I think that would resolve the problem of "click rings." Or another scenario would be to limit the total amount of clicks a member could use during a span of time. Members might start being more selective as to who they click. If you gave all members the ability to click 5 posts per month, they may really think as to who they are giving these clicks to.

Interesting discussion, though.


Featured Threads

Hot Threads This Week

Hot Threads This Month