What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 131 Stack Exchange communities.

Hang around SO for a length of time and you’ll eventually notice that some quality answers receive an unusually large number of upvotes, while others receive just a few votes or no votes at all. A user may gain 80 rep for an answer to a very simple question, while another user might gain just 10 rep for an answer to a very difficult question.

Since the difficulty of the question isn’t being factored into the reputation score, the voting system can result in an inaccurate representation of the answer’s true value based on the question’s difficulty. This kind of lop-sided valuation might give an inaccurate representation of a user’s contribution to SO. According to this, “Reputation is a rough measurement of how much the community trusts you; it is earned by convincing your peers that you know what you’re talking about.” So in theory, someone might gain a large amount of reputation for answering very simple questions, be upvoted by the community, and then be thought of as a trustworthy expert, while someone else who has answered tough questions be thought of as a novice.

To help balance this, I’d recommend that a weighting system be introduced for questions. Now I know a few questions have been asked about weighting questions before, however the other suggestions seem to rely on the person posting the question to assign the difficulty level to their own issue. My suggestion would be to let the community decide how difficult a question is, and then factor that into the calculation for reputation. For example, a scale of ‘basic’, ‘intermediate’, and ‘difficult’ could be added to a question, and just like the up/down vote, each user could cast their rating of how difficult the question is. Reputation would then factor in the number of votes for each difficulty rating received to the overall reputation calculation to anyone receiving upvotes.

Potential issues

While a change like this might make understanding why you received X points of reputation less easy to understand at a quick glance, it would be truer to the definition of what reputation is, and why it’s valuable. Three difficulty options may not be ideal, and I wouldn’t be the person to suggest how to best factor in the difficulty to the final rep calculation, but this is what a meta discussion is for. I’m also not sure of how older questions would be dealt with.

If I’ve just rehashed something that’s been covered before I apologize. Searching brought up the suggestions of self-determined difficulty rating but I found nothing on the community making the call.

tl;dr – Questions should have a community voted on difficulty ranking that gets factored into the reputation calculation. Users should earn more rep for answering harder questions.

share|improve this question
7  
If it affects the reputation I can potentially earn, then what would stop me from ranking every question as difficult? (I am a rep-whoring soab of course. ;) ) –  Bart Aug 17 '12 at 15:44
    
Likewise, if someone else is getting upvotes on a question I answer and it makes me mad I can mark the question as easy so his upvotes are worth less. Additionally you need to worry about people going after all posts someone answers and marking the questions as easy/hard when checking for voting fraud. –  Servy Aug 17 '12 at 15:47
    
@Servy - the serial downvote check that exists could be used to monitor such activity I would imagine. Plus, the difficulty of a question would be a relatively small adjustment to the overall rep, i.e. not have a multiplicative effect. –  j08691 Aug 17 '12 at 15:50
    
@j08691 I'm not saying it couldn't possibly be handled, I'm just saying you'd have to handle it, and that could involve a lot of additional work. –  Servy Aug 17 '12 at 16:10
    
Some answers to really difficult questions get zero upvotes, when not enough people understand them. How do we solve this? –  Bo Persson Aug 17 '12 at 16:15
    
This would definitely stop Jon Skeet's ridiculous reputation. Do we really want that? –  Cole Johnson Aug 18 '12 at 4:06

2 Answers 2

Give "experts" a better upvote, eg. a "double upvote". Experts are generally those who would be more active, answering various questions. They will often see other recently/already answered questions.

If they think that the current answer is exceptionally good, then they can give the question a "double upvote".

Since experts are probably more active, and will see more unusual questions, then this could partially improve the reputation given for great answers to difficult questions.

An expert could be defined as a user which has earned a bronze/silver/gold badge in one of the question's tags.

share|improve this answer
    
I like the idea of only users with a certain amount of rep being able to cast difficulty votes and maybe only X number of votes can be accepted per question. Experts earn rep for voting on question difficulty and maybe can earn a badge after a certain number of votes. –  j08691 Aug 18 '12 at 19:48

I do not think this will work. A great answer to a difficult question having few upvotes is an indication that few people have visited that question. To make your system reliable you need at least a couple of difficulty votes for each question, but the questions that needs this the most are visited by few people.

Also, try to keep things simple. I think the current voting system is enough as it is already. The reputation will still, over time, show which persons are active and which ones are participating less.

share|improve this answer
    
I like the idea of simple, however the current system doesn't stay true to what it has defined reputation as being useful for. I also don't think an option to weigh a question's difficulty would complicate things for users. The developers who would need to create the system maybe, but I'm not them so it's no skin off my nose. –  j08691 Aug 17 '12 at 15:52
    
I think it will complicate things. The problem you are trying to solve is essentially "not enough people engage in upvoting answers to difficult questions, making great answers get unfairly little reputation". How do you think your proposed solution would help in this? Why would people vote on the difficulty of a question rather than on good answers to the same? –  Emil Vikström Aug 17 '12 at 15:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .