My name is danah boyd and I'm a Principal Researcher at Microsoft Research and the founder/president of Data & Society. Buzzwords in my world include: privacy, context, youth culture, social media, big data. I use this blog to express random thoughts about whatever I'm thinking.

Relevant links:

Archive

number games and social software

Over the last month, i’ve been driving Mimi’s Hybrid on and off. One of my favorite things about the Hybrid is that it tells you how many MPG you’re averaging over time. I find myself driving around town trying to maximize that number, getting uber excited when it goes up and super sad when it goes down. It reminds me of when i used to try to maximize my miles per hour when going from Boston to New York only this is more environmental. Yet, it’s not the environment that i’m concerning myself with – it’s all about number games in the same way that people obsess over every pound on the scale or the calories in every bite.

Then i was thinking about Tantek and Jason raving about Consumating. I love the fact that it’s a lot of cool geeky people but i can never get over the lameness that i feel when i log in and look at my score. And yet, i can’t be bothered to answer the questions that make me feel all uncomfortable in the hopes that someone will like my answers and rate me higher. It’s a catch-22 for me. Yet, i totally understand why Tantek and Jason and others absolutely love it and why they go back for more.

And then i was thinking about the people on Yahoo! Answers who spend hours every day answering questions to get high ranks. It’s very similar to Consumating only it’s not all embarassing because it’s not really about you – it’s about the answers. There’s no real gain from getting points but still, it’s like a mouse in a cage determined to do well just cuz they can.

This all reminds me of a scene in some movie. I can’t recall what movie it was but it was about how you just want to be the best at *something*, anything… to have something to point at and say look, i’m #1! The validation, the proof of greatness! Even if that something is problematic attention getting like being the #1 serial killer. (Was it Bowling for Columbine?)

I started wondering about these number games… They’re all over social software – Neopets, friends on social network sites, blog visitors, etc. Who is motivated by what number games? Who is demotivated? Does it make a difference if the number game is about the group vs. the individual, about one’s self directly vs. about some abstract capability?

Are there some number games that work better than others in attracting a broader audience? I’m thinking about Orkut here… if the game is to get as many Brazillians on the site as possible, you only need a few obsessives to be the rallying forces; everyone else is part of the number game simply by signing up. So there are tons competing in the number games but only a few invested.

Does anyone know anything about how these number games work as incentives?

Print Friendly, PDF & Email

15 comments to number games and social software

  • I’ve noticed with a few people that private profile views are beneficial. But when the profile views are public and everyone can see them, it is clear who is more “popular”, and who is more of a “looser”. So, maybe that could put some people off that aren’t doing some intense networking.

    Great post by the way!

  • It strikes me that the attraction of these number games bears a certain similarity to the attraction of “leveling up” in certain RPG-style games. In games like WoW, there’s definitely a huge social benefit to leveling. However, I remember older games, like Diablo, where the entire game mechanic was driven by clicking your mouse on different areas of the screen. Yes, this is a huge over-simplification, but the point is that people would do these mind-numbingly simple tasks just to earn experience points or get the next skill or what have you. I never understood the draw, but I always suspected that it appealed to an innate desire for self-improvement of sorts. I don’t know if it’s the same as the social number games you describe here, but it would be interesting to know if there is a difference when the number game is socially motivated vs personally motivated.

    (there is certainly an argument that number games like Diablo are socially motivated, because players generally discuss the game with their friends, but I think that’s different, slightly more indirect social motivation than that which you describe here.)

  • Gaming is an excellent example of how earning numbers matters. Microsoft monetarized this in Xbox Live by allowing players to earn Achievement Points from games which can later be used to purchase media from their online marketplace (e.g. game add-ons, videos, demos).

    In the case of Social Media, we’re already seeing this in popular web services like Flickr and Digg. Netscape has “bought” over the top users on these services as determined under various metrics. Digg makes it very apparent since they now have a Top Digg Users page aggregating popular users based on quality submissions. It’s almost essential for social media participants to have this as a side-mission while they read and post new material. This gives users the opportunity to level up and have their voices recognized by the masses.

    On the flipside, I’ve recently read that Del.icio.us is on the decline and while it is a popular site, I think their downturn is due to the lack of attention to popularity as a driving force. The do show what’s popular, but having a new space to track popular users might be useful too.

    In conclusion, thinking of social media as a game might help sustain a that particular user community better than if there were no measure at all.

  • We put together a talk for GDC and eTech that addresses exactly this question – the slides are online at http://shufflebrain.com/GDC2006.htm. Feel free to email me if you have questions.

  • There is a theory in social psychology that attempts to distinguish between common-bond and common-identity attachment to groups. This was something used by Lea and Spears in their SIDE theory and has been used to try and understand how and why people contribute content or participate in online communities – what gives them motivation and what kind of motivation it is.

    One of the older social psychology articles that deals with this can be found here: http://psp.sagepub.com/cgi/content/abstract/20/5/484
    Sassenberg has a paper in Group Dynamics from 2002 that considers this when applied to the Internet: Sassenberg, K. (2002). Common bond and common identity groups on the Internet: Attachment and normative behavior in on-topic and off-topic chats. Group Dynamics, 6 (1), 27-37. and then there is relatively recent work by Kraut et al. here http://jcmc.indiana.edu/vol10/issue4/ling.html and there is a journal article in submission right now I believe. In essence, this work attempts to understand how individuals behave when they have either individual or group goals. I know this is “psychy” but this work attempts to address at lest parts of the questions you ask fairly directly so you might find it interesting.

  • heh. i was about to write “i don’t have the link right in front of me, but i know that AJ Kim has done a ton of work in this area, you might want to check out her slides from supernova.”

    nevermind!

  • Doug — http://tygar-blog.com — linked to this nice NYTimes article (bottom) about points and pistachio ice cream. Even though people like vanilla more than pistachio ice cream, folks often chose pistachio when the point value was higher. You can bias choices by changing the value of the rewards. Think about when you were a kid playing skeeball and how badly you wanted to collect 2,000 tickets so you could get that mini TV. It would have been cheaper to buy it somewhere else than play the hundreds of games of skeeball to collect enough tickets to get the TV, but that doesn’t enter into your mind when the TV was right in front of you.

    You can also learn about reward systems from casinos and psychology. If rewards are regular — say, a pellet of food if you push a button twice — then animals and people will perform that regular action to get the reward and move along. If you randomize the reward like a slot machine, people tend to keep doing that activity even if they’re rewarded — waiting for the next reward to come along. There’s reason to use regular rewards to hook a person initially (say, quickly leveling a character when starting an RPG) and random rewards to keep them coming back (say, collecting a set of magic armor that appears piece by piece, randomly).

    http://www.nytimes.com/2006/04/25/business/26leonextra.html?ex=1155096000&en=e92f7ee913dc9243&ei=5070

  • Hi there – this is Yumio, from Yahoo! Answers. I am also amazed at how much the simple “existence” of points motivate behavior on Answers. Then I remembered how on Solitaire, the computer game, turning on the “Vegas option” where you lose/win fictitious money made me play the game 10X more times. I think there is something innate about us humans which likes being measured by numbers. Maybe I should put $$ signs on the points…. 🙂

  • Human beings are always trying to compare/define. Numbers are easy to use to gauge comparison because the hierarchy is built in. Complicated algorithms try add some level of “fairness” based on prioritized criteria, but it’s still a numbers game. It’s much more challenging (and I would say valuable) to work developing your own personal value (i.e. positioning/differentiating) than get stuck in the number hierarchy loop. 🙂

  • As part of my dissertation, I looked at the learning curve for new commenters. They would start out at 1s and many would gradually get to the point where there average score was actually above 4. That is, they had learned to “game” the social system, and had managed to find a way of getting highly ranked.

    But then a funny thing happened. Inevitably, they settled back to an average of 2 or 3, and that remained their average for the remainder of their time on Slashdot. Having learned how to win, they stepped back and started playing the game for fun.

    I suspect this may be the case in lots of social software venues, but it may be that the ceiling on Slashdot created the effect, just as folks who beat a particular video game may go back to play it in a more exploratory way. Maybe in venues that are ranked rather than scored that competitiveness never dies off.

  • (copied from less active Many-to-Many comments — the problem with cross-posting 😉

    The online game industry knows a lot about this — they call people with a significant amount of this behavior an “Achiever”.

    The theory, originated by Richard Bartle 16+ years ago, is that all players of online games fall into four types: Achievers, Explorers, Socializers, and Killers.

    There has been a lot of good academic work on this topic in recent years, though like a lot of good ideas has been overly popularlized and thus simplified too much.

    How it applies to your thoughts is that a large percentage of the population (some say 50%+) have as their primary motivation to sustain a feeling of achievement. It can be lifting weights, doing the New York Sunday crossword puzzle, and of course, gaining experience and levels in an online game.

    Not the entire population has this as their primary motivation, but when you add it to people who have it as a secondary motivation it is such a large population that the online games primarily cater to this Bartle type. In recent years as game designers we’ve been perfecting the our appeal to this type of player — knowing how long to wait before giving feedback that an achievement is in progress, how often someone needs a break so they don’t burn out, etc. A lot of it feels like Pavlovian research.

    Thus more recently designed games like Worlds of Warcraft have learned from previous generations of game design to make an almost perfect game for the achiever dominant Bartle type. It isn’t just marketing or better graphics — they’ve really learned how to addict people to the feeling achievers get when they achieve.

    However, not all the population is achievers, and there does appears to be some sexual dimorphism — there are still more women who are achiever dominant then other types, but a lower percentage then there is for men.

    As I said, there is quite a bit of academic research on this topic in the last few years — people have been investigating if the Bartle types are real, if there are more then the 4 types, how much crossover is there between the types, and trying to figure out the observer biases caused by limiting the sample of people to those that are comfortable with computers or games.

    The original paper is “Players who Suit MUDs” by Richard Bartle, if you search for others papers that refer to that paper you’ll find quite a few interesting ones.

    It is disapointing to me that a lot of social science people automatically discount this type of research since it originated in the game industry, but I think its lessons apply to many different disciplines, ranging from industrial design (how to make a car that people care about keeping the MPG high), software engineering (how do keep people interested in the information we are delivering without burning out) to marketing and adversing.

  • You have great ideas but i am sticking with my first thought. Just yesterday my husband and i discovered that you can buy an over ride kit for a Prius that lets you (or hopefully passenger) use the navigation while driving – and another one that turn the car into “stealth mode” meaning 100% battery as long as it is going under 34 then it switches to normal. I does not hurt the car because it can do this anyway in Europe – they just dont care to activate that nice energy saving little feature for us americans. Am not schilling for anything – dont know the site name – just google prius navivation override kits. but spell it right 🙂

  • Don’t ya think it’s just raw competitiveness? We’re genetically wired to compete, and numbers are an easy vocabulary for measuring ourselves against our past performances and each other.

    I think this competitive drive is one of the reasons why people constantly strive to develop new games, sports, tests, etc. – to create new opportunities to become the best or first at something.

    You even see it in blogging with people who get off on being the first to post a comment! 🙂

  • Check “Permissive Marketing”. The same behaviour is seen with any marketing points program, particularly frequent flyer miles. People going to the CEO level change hotels and traveling plans to optimize their miles.

    For me it just seems an example of you get what you watch. In case studies on business PKI (Performance Key Indicators) you usually noticed that people will optimize what’s measured. The same is seen with all kinds of reward systems (“you get what you pay for” – doesn’t really matter if it’s monetary or not).

    As someone who uses “points” to generate behaviour from kids (in the scouts), I’ve always been surprised on how well this works.