The Gamification User Types Hexad Scale

Written by Gustavo Tondello. Infographics by Marim Ganaba.

Several studies have indicated the need for personalising gamified systems to users’ personalities. However, mapping user personality onto design elements is difficult. To address this problem, Marczewski developed the Gamification User Types Hexad framework, based on research on human motivation, player types, and practical design experience. He also suggested different game design elements that may support different user types. However, until now we were still lacking a standard assessment tool for user’s preferences based on the Hexad framework. There was also no empirical validation, yet, that associated Hexad user types and game design elements. A collaborative research project by the HCI Games Group, the Austrian Institute of Technology, and Gamified UK sought to accomplish these two goals: (1) create and validate a standard survey to assess an individual’s Hexad user type and (2) verify the association between the Hexad user types and the game design elements they are supposed to appeal to.

In case you are not familiar with the Hexad user types yet, please take a moment to watch this video or read the user types’ descriptions at Gamified UK.

Our research developed and validated a 24-item survey to assess an individual’s Hexad user type. Would you like to know your user type? Why don’t you take a minute to take the online survey at Gamified UK?

Gamified UK Gamification User Type HEXAD Test


Hexad User Types and Game Design Elements

We asked participants to rate how much they enjoy several game design elements and compared the answers with their Hexad user type to find out which game design elements are preferred by each user type. The following infographic summarises the results.

Hexad User Types and Personality

We also assessed participants’ personalities using the Big Five personality traits and compared the results with the Hexad user types. The following infographic shows the user types more likely to appear for people with higher scores in each of the five personality traits:

Using the Survey

Using the Hexad user types framework and the survey is more effective than asking users about design elements directly because the survey’s goal is to understand more about user psychology in a gamified context rather than just determining game elements they prefer. Furthermore, users are not necessarily gamers and might not be aware of their game preferences or be familiar with game design vocabulary. Therefore, our survey aims to use a common vocabulary.

There are several ways to use the Hexad model to personalise gameful applications. For example, designers would be able to screen their target audience using the suggested survey and choose the adequate design elements for each user. In research, the survey can be used to better understand user engagement and enjoyment in studies regarding gameful applications.

If you wish to use the Hexad user types survey, please read the original publication to be presented at CHI PLAY 2016, which contains details about the validation study and the survey items. Use of the scale is free for academic purposes. For commercial uses, please contact us. If you would like to join the HCI Games Group as a graduate student working on gamification and games user research, please also contact us.

The Gamification User Types Hexad Scale

7 thoughts on “The Gamification User Types Hexad Scale

  • July 12, 2019 at 11:52 AM


    I’m doing my master thesis at the HES-SO (University in Switzerland) on the gamification topic and I’m using the survey that you have done and empirically validated (2016 and 2019) in my study. I read both articles and I can’t find the way you calculated the score and determine the player type.

    In your article of 2019 there is just that sentence that said “After each completed survey, the website calculated the scores for each user type and presented the user with a chart of the results.” but there is no information about that algorithm in particular.

    I guess each item have a score (as displayed on this website: from -3 to 3. We calculated the score for each subscale. For example: if a person respond “somewhat agree”, “somewhat agree”, “agree” and “strongly agree” for the 4 item for player so the score is 1+1+2+3=7. And the dominant score become the player type of the respondant. But I’m not sure about that. By additionning the score, in my case, a respondent can have a maximum of 12 points (3+3+3+3). In your examples, the mean is near 20.

    So I would like to know if it was possible to get that algorithm that compute the score.

    Thank a lot in advance for your help and have a nice week-end!

    Dylan Montandon

    • July 20, 2019 at 3:38 PM

      Hi Dylan,
      Thank you for your interest in our research!
      In our paper, we assigned the values 1 to 7 to each Likert response option. So, when we added up the four items per user type, the final scores were between 4 and 28.
      You could also use Likert scores between -3 and 3, so the final scores for each user type would be between -12 and 12.
      But if you want to have results that are directly comparable to ours, just use the values from 1 to 7 and add them for each user type.
      Best regards,
      Gustavo Tondello.

      • July 22, 2019 at 4:06 AM

        Hi Gustavo,

        Thanks for your reply! It helps me a lot.

        Have a nice day,

  • September 27, 2019 at 3:31 PM

    Hi Gustavo,
    I am doing my PhD Dissertation on using the Player Type Hexad as a design framework for delivering customized gamified learning in mixed reality environments.
    Just briefly wondering if my approach is acceptable or not. I added together the scores for each of the 4 items in each latent variable 1-7, for example here is one participants’ data and how I calculated score:
    Socializer(S), Free Spirit(F), Achiever(A), Disruptor(D), Player(R), Philanthropist(P)
    24S, 24F, 26A, 9D, 18R, 25P = Combined total 126/168 giving them the following percentage of each score
    S19%, F19%, A21%, D7%, R14%, P20%
    That would be placed in descending order to and show only scores that break higher than 16.66% (meaning the score was net positive)then become the player’s “motivational fingerprint”; in this case it is A21/P20/S19/F19
    The reason I turned them into a percentage of the total score was that some people rank 5/7 the same as others rank 7/7, and I figured to only real way to compare someone’s TOP type is to weight each person’s answer by their total score. Someone who ranks everything 1/7(-3) or 2/5(-2) except two items that get 4/7(+1) would still have a motivational fingerprint that reflected how heavily they weighted the different types, rather than comparing to the maximum possible score.
    4S, 4F, 9A, 10D, 5R, 4P is a good example. The participant scored 1/7 to every single S,F, and P item, but scored D and R quite high compared to those.
    According to the traditional hexad this person is a “Disruptor” but this ignores a lot of information. This participant has a total score of 36 so their fingerprint is actually
    S11%, F11%, A25%, D28%, R14%, P11% = Motivtional fingerprint of D28/A25

    This adds a lot more nuance to the player type score because we can see that the D28/A25 player is nearly equally motivated by seeing change in a system, as they are with achieving progress in that system. They are, in essence, a Mastery focused agent of change. They are refered to as “Destroyer” in the dodecaded, but I perfer to simply hybridize the dominant types together and called them a D28/A25 rather than try to find a unique label for each of the hundreds of possible iterations of the six latent scale items.

    Curious to hear your thoughts on this. Am I missing anything or have I made any wrong assumptions in adapting the output of the scale like this?
    Do you see the utlity of a Player type fingerprint, rather than simply stating one’s dominant type? I want to see (in my research) if secondary type interacts with primary type in a meaningful way. For example, is a D28/A25 meaningfully different than a D28/F21 in terms of game mechanic preferences during storyboard focus groups.

    • September 27, 2019 at 3:57 PM

      Hi Dov!
      Thank you for your interest in our research!

      I think that the method you are using makes sense.

      These user types represent archetypes of what a person totally focused on each motivation would be, but no one is actually 100% a single motivation. So, you need to consider all the scores together. For example, when we talk about “a disruptor”, I understand it as a model/archetype of what a person would be if they were totally motivated by change and nothing else. It’s a way so we can describe the motivation. However, I do not think it is useful in practice to say that someone is “a disruptor” just because their score in this type is higher than the other types.

      When I used the Hexad, I always consider all the six scores together, so I consider how much the person is motivated by each of the types of experiences. I do not consider the “dominant” type and not even the “primary/secondary”. I always consider all six scores. What you are doing seems similar, except that you are dropping the user types with low scores for each person and then focusing only on the types with higher scores, which seems fine as you are then focusing on what the person enjoys the most. Still, I would not say that you need to “hybridize the dominant type”, just think of the user preference as a profile that consists of six scores for different types of motivation.

      Best regards,
      Gustavo Tondello.

      • September 27, 2019 at 5:02 PM

        Thanks so much for your reply Gustavo,

        I really appreciate your perspective on a global “player type” view in order to customize motivational game mechanics.
        I wonder how you might go about using that score to suggest game mechanics to designers in a more complete way than by referencing primary type alone?
        My goal with motivational fingerprinting is to have a score that represents only motivating mechanics, which can then be used to match to a table of game elements that will combine well in order to satisfy the intrinsic and extrinsic needs of that specific player.
        Perhaps it would be good to include a game mechanic ranking test alongside the hexad? I suppose the issue with ranking game mechanics is that it is very hard to convey a mechanic through text alone to anyone that isn’t intimately familiar with many examples of games that have tried (successfully, or not) to use those mechanics, and how that player felt when engaging with those games…

        The way I did this is by doing focus groups to discuss storyboards based on 12 different mechanics and compare the quantitative responses, and qualitative feedback about what could be improved in each storyboard, to their full player type score to see how it matched up. I ended up with so much data that I am not really sure how to make sense of it!

        • September 30, 2019 at 2:06 PM

          Hi Dov,

          We did already conduct some initial studies for ranking and classifying gameful design elements. I’m not sure if you’re familiar with this paper already; if not, please check it out:
          But you’re right, this study is a good starting point, but it is imperfect because a shor textual description of each element does not capture the full complexity of said element, and is subject to participants’ past experiences and understanding.

          Regarding the approach to select gameful design elements for each user based on their user type scores, my thesis has some initial suggestions ( – check chapters 3, 5, and 7). However, the whole gamification research community is still taking the first steps in this direction. There is certainly a need to develop more detailed/accurate methods to do this selection. This is open for research; I don’t have a final answer yet about how to do it. Actually, I believe no one has one yet, everybody is still investigating the matter using different approaches.



Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.