Monday 10 March 2025
Data workers, those individuals who spend their days analyzing and manipulating complex datasets, often have a mental model of their data that is different from the way it’s actually stored. This can lead to errors and difficulties when trying to work with the data.
A recent study looked at the internal and external representations of complex data held by ten participants who had been working with the same dataset for over a year. The researchers found that the mental models of these data workers were diverse in structure and often diverged from the reified data model they themselves had developed.
One of the main challenges faced by data workers is the need to recall the data model, which can be difficult especially if it’s complex or hierarchical. In addition, the study showed that even those who designed the data model didn’t always use it when describing their data. This highlights the importance of understanding how people think about and interact with complex datasets.
The researchers identified two parallel hazards associated with mental and reified data models. The first hazard is that a person’s mental model may not accurately represent the actual data, which can lead to errors in analysis or visualization. The second hazard is that a person’s mental model may be incomplete or vague, making it difficult for them to effectively work with the data.
To address these hazards, the researchers suggest several approaches. One approach is to design interfaces that support multiple models of the data, allowing users to switch between different views and representations. Another approach is to provide tools that help users annotate and structure their mental models, making it easier to retrieve and use the relevant information.
The study also highlights the importance of embracing diversity in mental models, rather than trying to standardize or normalize them. By acknowledging and working with these differences, data workers can create more effective and user-friendly interfaces for analyzing and visualizing complex datasets.
In practical terms, this means that developers should focus on creating tools and interfaces that are flexible and adaptable, allowing users to work in the way that feels most natural to them. It also means that data workers should be encouraged to share their mental models with others, either explicitly through annotation or implicitly through design decisions.
Overall, the study provides valuable insights into the complex relationships between data, mental models, and interfaces. By understanding these relationships, developers can create more effective tools for data analysis and visualization, and data workers can work more efficiently and effectively with complex datasets.
Cite this article: “Mental Models of Data: A Study on the Relationships Between Data, Mental Models, and Interfaces”, The Science Archive, 2025.
Data, Mental Models, Interfaces, Complex Datasets, Data Workers, Errors, Analysis, Visualization, Design Decisions, Annotation
