What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

On the Internet, Wikipedia says that:

1% of people create content, 9% edit or modify that content, and 90% view the content without contributing.

Is this true for the Stack Exchange network?

share|improve this question
    
No. We have far more people creating content than editing. Of course, if you don't count asking questions as creating content, the picture is different. –  Daniel Fischer Apr 29 '12 at 0:38
1  
1  
@Daniel Fischer Data is available. This question can be answered. –  bobobobo Apr 29 '12 at 0:43
    
@bobobobo But I don't have the data, so I don't answer but just throw out a comment according to my unscientific observations. –  Daniel Fischer Apr 29 '12 at 0:47
    
It depends how you define "people" -- are we talking about "internet users" or "registered users of the site?" The latter can be answered (approximately) using Data Explorer. –  Jon Seigel Apr 29 '12 at 0:59
    
31%, 24% and 50% aprrox. –  phwd Apr 29 '12 at 1:12
1  
@Jon - People as in people. :) I presume the vast majority of people who just view content (i.e. the 90%) are not registered users, regardless of the site. –  Nick Chammas Apr 29 '12 at 3:21
1  
OCCUPY.... somethingorother. Can you occupy a website? Sniz, I need to think about this. –  Won't Apr 30 '12 at 14:17

1 Answer 1

up vote 18 down vote accepted

That conjecture is not true for Stack Overflow, the site with more data than all other Stack Exchange sites, combined.

Stack Overflow tracks over a million users, Here is post-edit versus creation data from the current SEDE snapshot:

                            Number          Percent
                            ---------       -------
Users                       1,045,590           100
Made Edits                    244,960            23
Created Posts of any type     686,089            66
Created Questions             509,085            49
Created Answers               364,366            35
Created Wikis, etc.               616             0
Created ONLY Questions        321,722            31

Right away, we see that about triple the number of users create posts as do edit them. Even if we declare questions as "not (worthy) content", we still have 1.5 times as many "creators" as editors.
Conjecture busted. busted


Of course, many more people view the content without either contributing or being tracked as a user, and identifying the number of people who "View" content is difficult.

As of the last SEDE data, Stack Overflow had 1.8 billion "unique views", but it is not clear how that is calculated. Many repeat-views are filtered but many aren't.

Of those 1.8 billion views, only 1 out of x views is a unique person. If we set x to 651, that gives us the magic "9% of people edit content". In that case, the numbers become:

                            Number          Percent
                            ---------       -------
Users                       2,722,627           100
Tracked Users               1,045,590            38
Made Edits                    244,960             9
Created Posts of any type     686,089            25
Created Questions             509,085            19
Created Answers               364,366            13
Created Wikis, etc.               616             0
Created ONLY Questions        321,722            12

(Note that "1.8 billion" was 1,772,430,030 and 1,772,430,030 / 651 = 2,722,627)

So, if that ratio (651) held on average, then we would have 9% of people editing, 13 to 25% (depending on if you counted questions as "created content") creating content, and as few as 66% to 78% not contributing -- Assuming no overlap between those who "create" and those who edit.

Since that is not a plausible assumption, the "not contributing" percentage would more realistically be 75% to 87%.

share|improve this answer
1  
My, what a very long tail we have indeed :) –  Tim Post Apr 29 '12 at 5:47
    
According to your numbers for tracked users, it looks like SE inspires a higher level of participation than is predicted by the 1% rule. This likely holds even if we account for untracked users, as long as x is not much larger than 651, correct? –  Nick Chammas May 1 '12 at 16:39
    
I'd say that SE does inspire more participation, by design. Also the design encourages a user to be a "creator" before he can edit too easily. ... Note that "participation rate" goes up as x gets bigger and goes down as x gets smaller. At x = 1, the participation rate would be rounded to 0 percent. x = 651 is probably not too far off and still yields as many as 87% "freeloaders" -- close enough to the predicted 90%. What x actually means is that the given IP address or user viewed 651 pages on average. This seems plausible. I've viewed 1000's but many have only viewed only 1 or 2 –  Awesome Poodles May 1 '12 at 17:00
1  
rule of thumb for our view count, it is at least 2x the number of actual views versus what is counted, because we are very strict. Last time I calculated it was closer to 3x if you look at Google Analytics and perhaps Quantcast: quantcast.com/stackoverflow.com –  Jeff Atwood May 4 '12 at 20:28

You must log in to answer this question.

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