So here it is, the RPG Design Feature Survey. There’s some really interesting results from this, I only wish the sample size was a bit larger. 32 people responded to the survey, I hope GMs and game designers find it useful when thinking about their games.
Each question is pared with a dysfunctional form of the question. For example “The PCs are heroes and unlikely to die” being the functional form of the question while “The PCs are expendable” being the dysfunctional form. Each form of the question gets a different viewpoint on the subject. You may not mind the PCs being expendable but you’d really like them to be the heroes.
I include some analysis in with the data but it will take some explaining to understand it. Lets take an example.
Q7. System has rules for social conflict |
|
|
|
|
|
Answer Options |
Response Percent |
Response Count |
I like it that way. |
37.50% |
12 |
I expect it to be that way. |
18.80% |
6 |
I am neutral. |
28.10% |
9 |
I can live with it that way. |
9.40% |
3 |
I dislike it that way. |
6.30% |
2 |
|
answered question |
32 |
|
skipped question |
0 |
Q108. No system for social conflict |
|
|
|
|
|
Answer Options |
Response Percent |
Response Count |
I like it that way. |
12.50% |
4 |
I expect it to be that way. |
0.00% |
0 |
I am neutral. |
34.40% |
11 |
I can live with it that way. |
28.10% |
9 |
I dislike it that way. |
25.00% |
8 |
|
answered question |
32 |
|
skipped question |
0 |
Here’s the analysis for the example questions. It starts out with the two questions side by side. The A question is the first question listed and the B question is the second question listed. The second line is the count result that got the most votes for each question. The third line is what the result text was. These results are used in the Kano model of analysis.
A |
B |
12 |
11 |
I like it that way. |
I am neutral. |
Positive |
Negative |
23 |
-17 |
|
|
Exciter |
FD |
Trends Linear |
|
Weighted Positive |
|
That can be useful but I realized that this wasn’t the whole story so I looked at two more ways of comparing the data. There are a lot of “I am neutral” results but they’re often less than half of the total number of votes. I needed to look at the votes that showed a preference either way. In the fourth line, the votes that are positive are counted, the votes that are negative are counted, if the positive votes are higher than the negative votes it will say “Positive” otherwise it’s “Negative”.
The fifth line is a weighted result. I was thinking that a person that responds “I like it that way” or “I dislike it that way” will want to or not want to play a game based on the question, while everyone else may or may not based on other factors. So I made those votes count for two and ignored the neutral votes (cause they don’t really care). If the number is positive then people like the idea in the question, if it’s negative they don’t. Bigger numbers mean stronger feelings either way.
A – First Question |
B – Second Question |
12 – Highest result count for first question |
11 – Highest result count for second question |
I like it that way. – What the most chosen result was. |
I am neutral. – same here for the b question. |
Positive – Comparing the number of positive vs negative votes |
Negative – same for the B question |
23 – A strong weighted positive result |
-17 A weighted negative result |
|
|
Exciter – This is a Kano Exciter result. |
FD – this is just my note as to which form of the question comes first. |
Trends Linear – positive result on one side, negative on the other, it trends toward a Kano Linear result |
|
Weighted Positive – Comparing the weighted results. |
|
Here’s the analysis key
Recap of Kano Result types
Mandatory Features
Mandatory features for a game are things that are required for the game to be enjoyed. These are things like a ruleset or consistency in applying rules. They’re things that, if left out will make the players totally dissatisfied with the game.
The interesting thing is, with a mandatory feature once the need is satisfied, no more satisfaction results. If you pile on rules that players don’t need to play, they aren’t going to be any more satisfied with the game. If important rules are missing or poorly made, the player’s satisfaction will be reduced.
Mandatory results included, sticking to the game rules, PDF costing $10, The GM having the final say and having combat rules.
Linear Features
Linear features are things that increase satisfaction for the players the more it is done. This may be in game rewards like money or experience (dependent on the game) or time for their character in the spotlight. The more you give them the more satisfaction they will derive from the game.
Linear features are the most intuitive features because their relationship is direct. More is better less is worse.
Linear results were, a unified mechanic, good artwork, simple vehicle rules, combat not being determined by equipment, not using miniatures and maps for combat, basing a game on alternate history, not diceless and not generating a character randomly.
Exciter Features
Exciter features are ones that the players like when they see it but don’t require.
The nice thing about exciters is that since the players don’t know they need them, leaving them out does not negatively impact the game but adding them in enhances their enjoyment.
Exciters were, PDFs for $1-3, social conflict rules, using some kind of points to influence a story, using the player’s description of a character as their background, using attributes, consistent rules application, a detailed setting for the game, a free game, A Sci-fi setting,
Reverse
The reverse result means the player would like the opposite of the feature in the survey. I included the reverse results in the linear results but just reversed them.
Indifferent
Indifferent means the player is not interested in the feature either way. Especially with these results I look at the other methods of analysis to get a little more detail into how the respondents felt about them.
Vague
Vague means they have given contradictory responses and further more detailed questions on this subject may be required to resolve the contradiction. The vague results in this survey were because I messed up asking the question. Still there is useful information in these results, they just can’t be used against the Kano model.
How’d it go?
I think the results are intriguing. I’d like to do more on some of the subjects that people have brought up because now I want to analyze those ideas. However, the number of respondents were much lower than I was hoping for and that’s a bit of a downer. Maybe someone a bit more prominent in the community could do this and get a more enthusiastic result. So will I do more? If you comment that you find this data useful or even just interesting, I will. If no one comments, I’ll just leave well enough alone.