Leveraging Artificial Intelligence to Enhance the Survey Process
Author
Ting Yan
Vice President, Methodology & Quantitative Social Sciences
September 2025
NORC is harnessing AI to elevate survey design, sampling, and data integrity across the research lifecycle.
Survey organizations face persistent challenges: declining response rates, rising data collection costs, and increasing demands for timely, high-quality insights. As artificial intelligence (AI) reshapes industries across the globe, its integration into survey research offers a powerful opportunity to enhance operational efficiency and data quality.
“As artificial intelligence (AI) reshapes industries across the globe, its integration into survey research offers a powerful opportunity to enhance operational efficiency and data quality.”
NORC applies a survey process framework (Groves et al., 2004) to guide the strategic use of AI across every stage of the survey lifecycle. From study design to data dissemination, AI is not a standalone solution but a set of tools that, when thoughtfully implemented, can significantly improve the rigor and reach of survey research. Below, we highlight applications of AI at key stages of the process.
Define Research Objectives
AI accelerates the early stages of research by synthesizing existing literature, identifying emerging data needs, and informing the development of innovative survey content. NORC uses generative AI to support rapid evidence reviews and surface novel topics that align with evolving policy and societal priorities.
Sampling
During survey sampling, we can use machine learning to enhance sampling efficiency. For example, NORC’s Big Data Classifier (Dutwin et al., 2024)—a machine learning tool designed to predict characteristics of sampled addresses—enables more effective stratification and oversampling. It also supports adaptive survey designs to mitigate nonresponse bias and improve representativeness.
Questionnaire Development & Evaluation
NORC is exploring the use of large language models (LLMs) to support structured expert reviews of survey instruments. Preliminary findings from our Methodology & Quantitative Social Sciences (MQSS) team show strong alignment between LLM-generated feedback and that of human reviewers, suggesting promising applications for early-stage instrument testing when led by a trained methodologist.
Design & Implement Data Collection
NORC has piloted the use of generative AI to transform how web surveys are conducted (Barari et al., 2025a). In one experiment, a textbot engaged respondents in real-time elaboration and quality probing on open-ended questions. Even with minimal fine-tuning, the AI improved response specificity and depth.
Post-Survey Data Processing
AI streamlines coding, editing, and validation of data. LLMs are used to automate the classification of open-ended responses (Lerner, 2024). NORC has also developed tools to detect AI-generated responses, helping to safeguard data integrity by flagging potential bot or fraudulent entries (Huang, Sepulvado, & Lerner, 2025; Lerner et al., 2025)
Data Analysis & Dissemination
AI supports the monitoring of data use and citation patterns (Lafia, 2024) and powers interactive tools such as chatbots to improve user engagement with survey findings.
These examples illustrate the breadth of AI’s potential across the survey lifecycle. While not exhaustive, they underscore the importance of a structured framework to guide AI integration. Realizing the full benefits of AI requires rigorous design, continuous evaluation, and a commitment to understanding its impact on survey error and data quality (Lerner, 2024).
NORC’s MQSS team leads a dedicated research program on AI in survey methodology. Our expertise positions us at the forefront of innovation, ensuring that AI enhances—not replaces—the scientific foundations of high-quality survey research.
References
Barari, S., Angbazo, J., Wang, N., Christian, L., Dean, E., Slowinski, Z., and Sepulvado, B. (2025a). AI-Assisted Conversational Interviewing: Effects on Data Quality and User Experience. https://doi.org/10.48550/arXiv.2504.13908
Barari, S., Lerner, J., Yan, T., and Christian, L. (2025b). Generative AI in Survey Research: Principles and Use Cases. Paper presented at the annual conference of the American Association for Public Opinion Research.
Dutwin, D., Coyle, P., Lerner, J., Bilgen, I., and English, N. (2024). Leveraging Predictive Modelling from Multiple Sources of Big Data to Improve Sample Efficiency and Reduce Survey Nonresponse Error. Journal of Survey Statistics and Methodology, 12, 435-457.
Groves, R.M., Fowler Jr., F. J., Couper, M.P., Lepkowski, J. M., Singer, E., and Tourangeau, R. (2004). Survey Methodology. Wiley.
Huang, L., Sepulvado, B., and Lerner, J. (2025). Detecting AI-Generated Survey Responses: Tool Development and Bias Mitigation. Paper presented at the annual conference of the American Association for Public Opinion Research.
Lafia, S. (2024, December 12). Monitoring Data Use to Demonstrate Impact. [Web blog post]. NORC at the University of Chicago. Retrieved from https://www.norc.org.
Learner, J. (2024, October 9). The Promise and Pitfalls of AI-Augmented Survey Research. [Web blog post]. NORC at the University of Chicago. Retrieved from www.norc.org.
Lerner, J., Sepulvado, B., Huang, L., Arnold, S., Johns, M.M., Vogler, S., and Fordyce, E. (2025). From Social Media to Survey Data: Employing AI-Usage Detectors to identify AI-Generated Response in the Hirise+ Survey. Paper presented at the annual conference of the American Association for Public Opinion Research.
Suggested Citation
Yan, T. (2025, September 26). Leveraging Artificial Intelligence to Enhance the Survey Process. Retrieved from www.norc.org.
