Measurement as the Foundation of Science

All demographic analysis rests on the quality of the underlying data. Flawed measurement leads to flawed inference. The Institute of Experimental Demography houses a dedicated Survey and Measurement Lab that pioneers new methods for collecting more accurate, nuanced, and ethically sound data on sensitive population topics. We move beyond standard questionnaire design to develop innovative techniques that reduce social desirability bias, capture hard-to-measure concepts like social networks or norms, and seamlessly integrate biological and digital data with traditional surveys. Our goal is to push the boundaries of what can be reliably measured in population studies, thereby enhancing the validity and power of all downstream causal analyses.

Cutting-Edge Methodological Contributions

A major innovation is our work with indirect questioning techniques. For topics where respondents may lie due to stigma (e.g., illegal migration, discriminatory attitudes, abortion), we employ methods like the list experiment or endorsement experiment. In a list experiment, respondents are randomly assigned to see a list of non-sensitive items or the same list plus one sensitive item. They simply report how many items they endorse, not which ones. By comparing the average count between groups, we can estimate the prevalence of the sensitive behavior or attitude without anyone having to admit to it directly. We have used this to accurately measure undocumented migration experiences and prejudice.

We also excel in network-based sampling and measurement. Standard surveys often miss marginalized or hidden populations. We develop and implement Respondent-Driven Sampling (RDS) and Network Scale-Up methods to estimate the size and characteristics of populations like sex workers, people who inject drugs, or irregular migrants. Furthermore, we collect detailed social network data within surveys, mapping who people talk to about important matters, who they rely on for support, and how these networks influence demographic behaviors like contraceptive use or migration plans.

The integration of biomarkers and physiological measures into population surveys is another specialty. Our field teams are trained to collect finger-prick blood samples for HbA1c (diabetes marker), dried blood spots for pathogen exposure, saliva for cortisol (stress), and anthropometric measurements. This transforms a social survey into a rich biosocial dataset, allowing researchers to study the embodiment of social inequality. We have developed streamlined, respondent-friendly protocols that achieve very high compliance rates.

• Digital Diary Platforms: Created smartphone apps for experience sampling, where respondents report moods, activities, and interactions in real-time, reducing recall bias.
• Paradata Analysis Unit: Systematically analyzes the 'data about data collection'—interviewer observations, call records, response latencies—to diagnose and correct survey errors.
• Cross-Cultural Cognitive Interviewing: Employs intensive qualitative probing in multiple languages to ensure survey questions are understood consistently across diverse cultural contexts.
• Geospatial Linked Data Protocol: A secure method for linking survey respondents' home locations to external geographic data (pollution, service availability) without compromising privacy.

From Innovation to Global Standard

The tools developed in our lab are not kept in-house; they are disseminated globally. We publish open-source software packages for implementing list experiments and analyzing network data. We conduct training workshops for national statistical offices and survey firms around the world, building capacity for high-quality data collection. We also engage in fundamental methodological research, using experiments embedded within surveys to test how different question wordings, orderings, or modes (phone vs. web) affect responses. This work continuously improves the craft of survey demography. In an era of declining response rates and growing mistrust, innovative and ethical measurement is more critical than ever. By investing in the science of measurement, the Institute of Experimental Demography ensures that the empirical foundation of population science remains solid, trustworthy, and capable of capturing the full complexity of human lives. This commitment to methodological excellence underpins every claim we make and every policy recommendation we offer, ensuring our impact is built on rock-solid evidence.