Causality-Based Visual Analysis of Questionnaire Responses.

Area of Science:

• Data Science
• Causal Inference
• Survey Analysis

Background:

• Causal reasoning is crucial for extracting actionable insights from questionnaire data.
• Classical statistical methods and existing visual tools face challenges in scalability and expert knowledge integration for complex questionnaire data.
• Current causal discovery in questionnaires is often time-consuming and relies on trial-and-error.

Purpose of the Study:

• To develop a systematic and efficient solution for causal reasoning in questionnaire analysis.
• To enable effective exploration of questionnaire data and derivation of causal relationships.
• To address the limitations of existing methods in handling large search spaces and complex causal structures.

Main Methods:

• Utilized association mining algorithms to identify question combinations with potential causality.
• Developed an interactive approach for exploring causal sub-graphs.
• Integrated expert-derived requirements into a visualization tool.

Main Results:

• The proposed system facilitates efficient exploration of questionnaire data for causal inference.
• A comparative study demonstrated the usability and efficiency of the developed visualization tool against state-of-the-art systems.
• The association mining approach effectively uncovers potential causal links between questions.

Conclusions:

• The presented systematic solution enhances the effectiveness and efficiency of causal reasoning in questionnaire analysis.
• The visualization tool, informed by expert input, improves the interactive exploration of causal relationships.
• This work offers a scalable and practical approach for deriving causality from complex survey data.