And finally, let's take a look at age differences. The following graph shows the average age of the respondents who selected particular motivations as their primary motivation. Consistent with findings elsewhere, players driven by power and competition tend to be younger. The motivations with the highest average age were exploration, immersion, and socializing. It's interesting to note that the motivations with lower average ages suggest a more "hard-core" mentality while the motivations with the higher average ages suggest a more "casual" mentality.
Well, that's depressing. No wonder it's so hard to find roleplayers around. =( Is there any hope for us?
When I took this survey, I placed roleplaying far down my list not because I don't like roleplaying but because my MMORG (WOW) doesn't really do much to support RP. I suspect as games mature, this element will rise dramatically.
But as currently crafted, where any real RP comes only from player interaction and the game environment itself has little RP support, it is hard to pay much attention to that element.
Simple examples: in a RP paper-and-pencil game, if my character spent 2000 gold on repairs with a particular blacksmith over several years, that smith would remember me and at least shoot-the-breeze when I showed up. In WoW, not a peep. Ot if I walk into a shop in full battle gear, I might not get quite the same reaction as if I showed in town garb.
RP servers is a nod in the right direction but I think the games themselves could do a lot more to support it. They don't feel like RPs despite their name.
I had a really hard time as a result of your earlier because my player types model has an "explorer" category and your data found no evidence to support its existence. Now, I look at what you're presenting here, and I see exploring is the third most important motivation.
Hey Richard - See this page from the presentation of that earlier data and where I discuss the problem with the Explorer Type (from 2005). The findings showed that "Analyzing Game Mechanics" was not highly correlated with "Exploring the World". In other words, they don't cluster together and don't coalesce into a Type as you suggested.
This is why I argued that the "Explorer" didn't exist. The underlying components don't cluster as they should if those two components tend to appear together. But this is very different from saying that "Analyzing Game Mechanics" and "Exploring the World" don't exist as motivations altogether.
As always, an interesting article, but I do have a question. I have a hard time seperating immersion from role playing. Example: I can be immersed in the game, role playing with other player, when that immersion is broken by an arriving player "talking" about a RL subject in a general chat area.
Perhaps your definition of the two would be helpful.
Tal - Great question.
In my mind, players can enjoy the sense of being in a fantasy world but have no desire to role-play. I guess I've always framed "Immersion" to be more about the world, and "role-playing" to be more about a specific way of interacting with other people.
I think you're right to point out though that they may be highly similar concepts to role-players like yourself.
In the bar graphs on pages 2 and 5, the 12 subcomponent motivations do not appear grouped into the 3 supercomponents (Achievement, Social, Immersion). In particular, on page 2, Progress is vastly far ahead of the other Achievement subcomponents (as noted in the text). Does this call into question the usefulness of grouping the subcomponents into the 3 supercomponents? Or is it just an effect of which data was gathered (since this survey was about primary motivation, rather than rating all the motivations, as in previous surveys)?
Will - I think it's the latter. Since players could only pick one motivation, correlations between motivations aren't part of the data. And it's those correlations that lead to the clustering.
So for example, people who like Warcraft 3 probably also like Warcraft 2, but when it comes down to it, people who like both probably still prefer Warcraft 3.
I originally role-played with pen-and-pencil RPGs, and I have to agree with the post by Marc. I have yet to see any MMORPG that makes any effort to foster real role-playing. The NPCs don't remember your character, or react to him or her any differently than they do to any other. Moreover, the other players in the game frequently choose oddball names ("1337roxxorz" or something equally inane), or behave in ways that would draw terrible repercussions in a real-life situation, but which the designers make no effort to control.
While this undoubtedly gives a game broader appeal, it does tend to undermine any effort at real roleplaying. That said, the game I play (Eve Online) is fun, anyway, and I didn't really expect it to be any different than it is.
Basically, if you want a role-playing game in which you can truly immerse yourself, find a good GM and play a table-top, pen-and-paper RPG. If you want a chance to kick back and have an enjoyable time in a computer game, play an MMORPG.
Nick>In other words, they don't cluster together and don't coalesce into a Type as you suggested.
Ah, OK. Having (re-)read your criticism, I note that it seems to revolve around the fact that you identified two sub-types of explorer. This is reassuring, actually, as they map onto the two types that emerged from my 8-type model; I'm suitably impressed by the fact that you picked these out, because although the two sub-types of killer were fairly obvious, the two sub-types of explorer weren't, yet your analysis appears to have found them.
I still think immersion is the end result of the process of moving between types, though, rather than a type in and of itself, so fortunately we do still have something to argue about!
Richard, I don't think Nick's surveys do a good job of identifying motivations. Many of his questions are of the "how much do you do this" type rather than the "what do you enjoy doing" type; even the ones that ask about enjoyment are "how much" rather than "what". The quantitative nature of the questions means that the answers are highly constrained by the games the survey taker is playing.
In Warcraft, for example, the game is almost entirely geared towards achievement - more levels and more gear. Is it any wonder that players spend most of their time doing that? True exploration is limited to breaking into unreleased content areas - for which one can be banned - and finding bugs on the test realm. It's highly discouraged, so even players who prefer that kind of play won't be able to engage in it much.
To actually get at peoples' preferences, they have to have choices. For example, you could ask them which of several games they preferred, and why. Unfortunately, that requires substantial numbers of survey takers who have played the same set of multiple games.
Warren - For this particular question, respondents were asked to choose the most enjoyable aspect of the game from a list. So there wasn't an issue with the "how much", "how often" question stems.
And in general, the more constrained multiple-choice questions were generated from earlier qualitative responses to open-ended questions of the form: "Why do you play MMOs? What are the most appealing aspects of the game to you?" Here are some links to those earlier articles (1, 2, 3, 4).
As much as possible, I try to move back and forth between open-ended exploration and quantitative clustering. So even in the recent survey phases, I've asked respondents in open-ended questions to describe motivations not covered in the multiple-choice questions.
And even then, most of the motivation questions in the survey are of the "how important is it to you to ..." and "how much do you enjoy ..." types rather than the "how often do you ..." types, so it does try to hit a level of generality beyond the current game.
It is true though that games shape player's preferences in complicated ways. On the other hand, if players don't know their own preferences until they have played multiple MMOs, then that puts us in the weird position of arguing that first-time MMO players don't actually have preferences (or that we could never know what they are) ... which I disagree with. I think preferences can change over time and thus a first-time player's motivations are perfectly valid. I may have hated sushi when I first tried it, but it grew on me over time. My initial preferences of sushi were not any less valid than my current preferences.
Hi, Nick! Thanks for the response. I'm actually somewhat surprised anyone was still checking this page; I didn't see the dates til after I posted. I wonder if Richard will be back.
I have to plead guilty to not being clear on which survey was which. I had just taken the "part 29" survey, which presumably postdates the previous comments in this thread. I understand better now that that's different from both the article we're commenting on here and the "Motivations Assessment", and I'll try to be clearer myself.
That said, I think the issue of game dependent constraints is still valid.
Marc's comment near the top is apropos. Roleplaying was low on his list not because he wouldn't like to be roleplaying, but because it's virtually impossible to roleplay in Warcraft. My contention is that asking about the most enjoyable aspect of a game is more of a judgement on the game than on the player. Progression comes out on top because that's what the current game designers are good at, not necessarily because that's what the players would really like to be doing. The same issue applies when you ask what the most appealing aspects of "the" game are, as opposed to asking, for example, what people would really like to see that the game doesn't support.
This could perhaps be tested by looking at correlations between the available games and the motivations of the people who play them. For that matter, it might be worthwhile to correlate to what part of the game people are playing; in Warcraft, for example, physical discovery is important for newer players, but becomes unimportant in the end game.
I would point out that the Bartle explorer type should be expected to be missing from your data, because, according to Bartle, it's a rare type. Rare types won't have a big statistical impact, and are likely to be missed in statistical analyses.
That doesn't mean they won't have a big impact on the game, though. For example, I think your analysis is likewise missing Bartle's killer type, as they are also extremely rare. However, they can have an outsize impact. I once followed a killer type in Diablo for an evening; over that time, he managed to monster kill scores of people, causing them to drop their items, which he then picked up and destroyed. Stuff like that caused people to think the game was "dominated" by griefer types, even though they were in reality a tiny percentage of the population.
Warren - I actually have the data on "what would you like to see more of in an MMO" (open-ended) from Phase 28. Just that it would involve hand-coding 2000 responses and other analysis got prioritized over it. It's both the beauty and tragedy of dealing with numbers. I wish I had more time.