Data Collection & Processing
AI tools can streamline many stages of data collection and processing, improving speed, efficiency, and in some cases, participant engagement. For example, AI-powered chatbots can support data collection by guiding respondents through surveys in an engaging, conversational format.
That said, AI should be seen as a tool to support — not replace — human researchers, especially when it comes to ensuring data quality and ethical oversight.
On this page
What AI can help with
- Collecting and processing both qualitative and quantitative data (e.g. using chatbots or AI transcription tools).
- Creating synthetic data or populations to study in isolation or supplement human datasets.
- Synthetic data is artificially generated to mimic the statistical properties of real data, without containing any identifiable information — helping to protect individuals' privacy.
- Delivering real-time behavioural interventions or guiding participants through them (e.g. via virtual coaches).
- Automating reminders and adapting interventions based on user responses.
Tools
Qualtrics Paid
Turns static open-text questions into a dialogue. Its Conversational Feedback feature generates follow-ups tailored to each respondent's answer, and prompts for more detail when a response is too vague to act on — richer qualitative data without a longer survey upfront.
Rasa Free & paid
An open-source framework for building custom conversational agents, tailorable for simulating interviews or collecting narrative data. Its CALM engine uses LLMs for understanding while you define the conversation logic — more control and auditability than a fully generative chatbot, but more setup than a plug-and-play tool.
Vink / Otter AI Free & paid
Transcribe interview audio to text. Vink is free and open-source, adapting OpenAI's Whisper to run locally on multilingual data — useful for sensitive recordings. Otter AI is the commercial alternative, with HIPAA compliance on Enterprise plans under a signed Business Associate Agreement. Either way, use enhanced informed consent, secure data management and human verification of the output.
Julius AI Data Transformation Free & paid
Cleans, transforms and explores datasets through natural-language queries rather than code. Produces a preliminary assessment report flagging errors and outliers before transformation. Best suited to single-file, exploratory work; verify outputs manually for research use.
YData SDK Free & paid
Generates synthetic tabular, time-series and text data under specified privacy constraints such as k-anonymity and differential privacy — useful for datasets that mimic real behavioural or survey data without exposing it. Note the older ydata-synthetic package is deprecated; ydata-sdk is the maintained one.
Synthea Free
Generates complete, realistic but entirely fictional patient histories and health records. Free of the cost, privacy and security restrictions of real patient data, which suits public health research and intervention testing.
Listen Labs Paid
Runs the full qualitative workflow — study design, recruitment, AI-moderated interviews with adaptive follow-ups, transcript analysis and reporting. Suited to rapid, large-scale or multilingual studies. It is a commercial research platform rather than an academic CAQDAS package: retain transcripts and prompts, validate AI-generated themes against the source material, and secure ethics and data-governance approval.
Example: Chatbots compared with web surveys
Researchers compared the effectiveness of chatbots and traditional web surveys in longitudinal data collection on daily news consumption. The findings indicated that web surveys generally yielded higher data quality and more favourable user evaluation. While chatbots offer the potential for increased interactivity and engagement, the study cautioned that they do not necessarily enhance data quality (Zarouali et al., 2024).
Example: Delivering interventions
Researchers reviewed 43 studies on text-based AI chatbots for health behaviour change, covering areas like physical activity, diet and smoking cessation (Fu et al., 2026). Chatbots acted either as routine coaches or on-demand assistants, most built on frameworks like CBT. Most studies reported positive effects, though only about a third showed clearly observable improvements — the strongest evidence being in physical activity, smoking cessation, stress management and diet. The authors call for more rigorous, long-term trials, and for more evidence on cost and on underexplored behaviours such as sleep and weight management.
Example: Synthetic data and populations
Researchers developed an agentic AI architecture to simulate the attitudes and behaviours of over a thousand real individuals. By applying large language models to qualitative interviews, the generative agents replicated participants' responses on the General Social Survey with 85% accuracy. This approach demonstrates the potential of AI-driven agents to emulate human behaviour in social science research, offering a scalable method for studying individual and collective behaviours (Park et al., 2024).
However, while agentic AI provides opportunities, it also has risks. Researchers of a recent paper on potential harms arising from increasingly agentic AI systems argue that as machine learning systems gain more autonomy, they can introduce new forms of harm, particularly affecting marginalised communities. They emphasise the importance of proactively anticipating these harms rather than merely responding to them after deployment. The paper concludes that recognising the agency of algorithmic systems does not absolve human responsibility but highlights the need for robust governance to mitigate systemic and long-term impacts (Chan et al., 2023).