Unlocking Potential: The Power of Gamification in Employee Data Science Learning

Data science is a critical competency for employees in the digital era, offering unparalleled insights and competitive advantages to those who master it. However, navigating the journey to data literacy presents its own set of challenges:

• First, Resource Navigation. Employees often find themselves overwhelmed by the significant volume of resources available.
• Second, Lack of Motivation. There is a widespread inability to recognize the utility value of data science in employees’ daily tasks.
• Lastly, Lack of Learning Culture. The absence of a supportive learning culture within companies further hinders progress.

Concerning all these difficulties in employees’ data science learning, a team at the Center for Transformational Play at Carnegie Mellon University works with 99P Labs to revolutionize data science training. This initiative brings together the latest in educational research and gamification strategies to create a training program that is as engaging as it is informative.

Let’s first get to know the team!

Team Photo

Abhishek Bora: Hi! I’m Abhishek Bora, I’m a graduate student pursuing a Masters in Information Systems Management at CMU. With a background in engineering and management, I’m interested in exploring experiential ways to make humans learn and grasp complex concepts.

Fern Zhang: I’m a graduate student in the Masters of Educational Technology and Applied Learning Sciences program at CMU. With a background in psychology and economics, I’m interested in leveraging data, technology, and research-based evidence to improve users’ learning experiences.

Jackson Chen: I’m an undergraduate student studying Statistics and Machine Learning at CMU. I believe in hands-on learning through experience and hope to improve data science intuition through educational technology.

Ruby Wu: I’m a graduate student in the Masters of Art in Interaction Design program at CMU. With a background in sociology and a focus on AI/XR/game design, I’m passionate about bridging business goals and user needs, to improve the overall experience.

Steve Lu: I am an undergraduate student at Carnegie Mellon University studying Electrical and Computer Engineering. I have a passion for game design and have been working as a game designer for 3 years now. I am releasing my first game soon and hope to work on more in the future.

Why Gamification

The team first proposed the use of gamification — a dynamic blend of psychology and technology that promises to revolutionize adult learning and data literacy. Empirical research underscores a fundamental truth: intrinsic motivation, rooted in human needs for competence, autonomy, and relatedness, drives engagement and deep learning. Gamification leverages this by incorporating game elements into educational content, making the learning process both enjoyable and effective.

Moreover, the evidence is clear. Gamification can enhance employee training and data literacy education by up to 40%, according to Zichermann. By following structured design models like the ADDIE model and incorporating gamification early in the training design process, gamified learning can significantly improve learning outcomes.

Team Design Workshop for Gamification Idea

Understand Learners

The team adopted a multifaceted approach to understand the diverse backgrounds and prior knowledge of their target learners to design effective instruction.

1. The literature review concerning data literacy training for employees reveals that learners often feel intimidated by and resistant to the training. Therefore, it is essential for the training to adopt a friendly tone and incorporate data that is pertinent to the employees’ everyday tasks, making the learning experience more accessible and relevant.
2. As for primary research, the team used data scraped from LinkedIn to analyze the departmental and educational backgrounds of current employees. This was complemented by conducting in-depth interviews with a few employees, leading to the creation of two distinct personas: the practitioner and the manager.

Data Analysis of Learners Background

User Interview Insights

The manager persona highlights individuals who possess data but lack the knowledge to utilize it effectively. Conversely, the practitioner persona represents those who, despite not having direct access to data, seek to understand how to leverage data in conversations with business stakeholders. This understanding aids in crafting targeted learning experiences that address the specific needs and gaps of each group.

Learning Objectives

Meanwhile, the team engaged with data science experts from both 99P Labs and Carnegie Mellon University to pinpoint the essential skills and knowledge necessary for proficient data-related performance in the workplace. Following this, they implemented a “think-aloud” technique with current employees, understanding their daily practices to uncover skill gaps and define the primary learning objectives tailored to the needs of the two personas.

Manager Learning Objectives

1. Frame the Problem with Data

• Understand the utility value of data science and possess a data-driven mindset.
• Ability to systematically identify and articulate a business problem, determine relevant data requirements, and formulate a clear, data-driven problem statement that guides subsequent analysis.

2. Understand and Use Basic Data Science Concepts

• Understand key data science concepts, such as statistical significance, recognizing their applications and limitations to inform decision-making.

Managers often come from a mechanical engineering background and may not naturally adopt a data-driven mindset. Therefore, the crucial initial step is to help them appreciate the value of a data-driven approach. While these managers aren’t expected to conduct data analysis themselves, they need to understand the fundamental concepts behind using data in decision-making processes.

Practitioner Learning Objective

1. Analyze and Convey Data to Business

• Ability to understand business requirements; collect and use relevant data to achieve those requirements.

Practitioners frequently encounter difficulties in effectively communicating data findings to other departments and their managers. Therefore, they need to develop a stronger business understanding and improve their ability to present their insights clearly.

Manager Training: Strategy Quest

The primary gamification strategy for manager training involves the use of simulations, which allow learners to observe the consequences of their decisions in real-time.

In Module 1 simulation, the learner works as a sales manager, facing a decline in sales of autonomous solar-electric fusion vehicles in 2050. The training is structured to require learners to make a series of decisions at various decision points, opting either for data-driven solutions or alternative methods to address challenges. This setup leads to diverse outcomes, enabling learners to directly experience the effectiveness of a data-driven approach by comparing the results of their choices.

Manager Training Simulation Example

After learners gain motivation in learning data science, Module 2 is paired with scaffolded data science instruction designed to help managers grasp both the potential and limitations of data science, particularly in visualization and predictive modeling. This instructional content draws from established resources like Harvard Business School’s “Data Science for Managers,” meticulously broken down to various levels of Bloom’s Taxonomy.

To enhance learning outcomes, the instruction employs diverse interactive methods tailored to different cognitive processes, such as drag-and-drop for memorization, hot-spot questions for application, or simulation for evaluation, ensuring an effective learning experience of data science.

Immediate Feedback Example

Practitioner Training: Data Quest

For practitioner training, the approach to simulation takes a more immersive turn, crafting an experience similar to a game where learners are bombarded with vast amounts of data. The challenge lies in selecting a limited subset of this data to inform business decisions. This setup is designed to simulate real-world decision-making scenarios that involve navigating through multiple variables, thereby fostering an internal motivation to utilize data science tools effectively.

Participants face the constraint of restricted data access, compelling them to prioritize and strategize which data to collect and analyze. Nevertheless, the simulation still consists of scaffolding designs, such as hints and feedback, to ensure the effectiveness of learning.

Data Quest Interface

User Testing Feedback

We conducted 3 user testing sessions with current employees from 99P Labs to evaluate the effectiveness of learning content and iterate our design.

For Manager Training Strategy Quest, we received positive feedback for its effectiveness and relevance. As a busy manager, the learner appreciated the short time commitment of taking the course: this concise 30-minute training, from his perspective, was “easy to understand”, “actual interaction was fun”, and “concepts were relevant to work”. Moreover, the learner considered the design of incorporating simulation and case studies innovative and effective in employee training.

Learners were also impressed by the Practitioner Training Data Quest because of its intuitive design, supportive feedback, and immersive data-driven environment. The training effectively encouraged learners to “to identify what data is important” and explore data visualization tools.

Recommendations & Takeaways

The experiment approach of using gamification provided insightful feedback to our team, and we concluded the following suggestions and takeaways for future employee training in data science.

1. Gamification is an effective method in employee training. Our user testing proved that gamified elements can increase learners motivation, leading to long-term learning habits.
2. Trainings should be short (20–30 minutes for each module). Time is essential for working employees. Thus, employees are more likely to participate in concise yet effective trainings.
3. Content should be easy to understand. When introducing technical data science concepts to novice learners, the design should introduce a lexion to the learners with affordances in order to enhance learning and prevent learners from feeling intimidated.

Next Steps

Based on the user testing experiences, we also designed an iteration plan for both trainings to improve the learning outcomes:

1. Adding more simulation and data science modules to the Manager Training Strategy Quest
2. Adding video-based scenarios into the Practitioner Training Data Quest
3. Monitoring the Game progress and stats of each employee
4. Adding more storylines and real-life scenarios to all simulations

Conclusion

In an era where data literacy is crucial for competitive advantage, data science training is a valuable method for employees’ reskilling and upskilling. Using the innovative approach of gamification, the CTP team aims to transform the daunting journey into data literacy into an engaging and impactful learning experience.

Through carefully crafted simulations and scaffolded instruction tailored for both managerial and practitioner personas, the program not only aims to bridge the skills gap but also cultivates a culture of curiosity and continuous learning. This initiative represents a pivotal shift towards making data science accessible and meaningful, ensuring that employees are equipped to navigate and lead in the digital landscape.