An article about the Stimsims game data reports

Game Data Reports

Game Data Reports


June 05, 2020

Author: Victoria

What Are the Game Data Reports?

The game data reports are visualizations of the anonymous analytics data collected during game play.

Generally, the data from one question is used to segment players. These segments are then compared on metrics like their final score or their time spent in sections of the game, to see if any interesting variations emerge.

Why Do I Make Them?

I believe games can provide more information about the problem solving process than traditional problems can. While the data is anonymous, there is still a rich amount of information to be gained through good problem design. I hope that with a large enough player base, this data might provide new insight into how players think and how educational content can adapt to different thought strategies to increase the chances of eureka moments occurring.

How Are the Graphs Made?

The data is collected while players make choices in the games. This data is then accessed in Google Data Studio to create graphs on any potentially interesting insights. The graphs have a live connection to the data, so they change to reflect any updates.

What Problems Do the Graphs Have?

There are several problems to be aware of when viewing the data.

The Data is Anonymous

Anonymity is important for the protection of players, but it does create a few caveats in the reliability of the data. Since the data is anonymous, it's impossible to know if the data comes from a human or a program. There are security mechanisms and tricks that can be used to keep the data clean, but none are perfect.

The Initial Data Dimensions Change During Game Development

During the course of development, choices are added or removed, depending on player feedback. This generates data that can distort the official games data, as the early dimensions will have more data than the dimensions added later in the project.

Additionally, early data in development is generated by my play testers and I. This design process is important to ensure that potentially relevant insights can be visualized in the graphs.

The dirty development data is handled through a range of techniques that I'll cover in future articles. Briefly, they include using different development and production properties, data filters and date ranges, to keep the data relevant to the question being asked.

This Is An Area of Active Research and Development

Exactly what data can be collected, how reliable it is, and how it can be synthesized into visualizations, is an area of active research. There will be lessons learned from each game as they come out, and as user interest in the games increase. I will likely change my methods as more information becomes available. I'll write about any useful updates in future articles.

Conclusion

I will be writing more on this topic in future, as any details or useful techniques become clear to me. For now, analytics and data studio allow me to share any interesting information I learn about how people solve problems in my games.