Time-Location Sampling
Also known as: Venue-Based Sampling, Time-Space Sampling, Venue-Day-Time Sampling, VDTS
Time-location sampling, also called venue-based or venue-day-time sampling, is a probability-sampling method for reaching populations that lack any list frame but reliably congregate at identifiable places and times. Developed and codified by Muhib, Stueve, and colleagues in a 2001 Public Health Reports article for a CDC youth study, it replaces the impossible task of enumerating a hidden population with the tractable task of enumerating the venues, days, and time slots where that population gathers. The analyst first builds an ethnographic frame of venue-day-time (VDT) units, then draws units at random, intercepts and enrolls eligible attendees on site, and finally weights respondents by how often they attend so that frequent venue-goers do not dominate the estimate. Because selection probabilities are known at each stage, the design yields defensible, variance-estimable population quantities rather than a convenience sample. For migration research it is especially valuable: day laborers at hiring corners, migrants at remittance shops, consulates, places of worship, markets, and transit hubs are mobile and unlisted, but they are observable in space and time. The method thus converts the visibility of a mobile migrant population into a genuine sampling frame.
Key highlights
- Creates a true probability-sampling frame for populations that have no list frame, by sampling venue-day-time units with known selection probabilities.
- Yields population-level estimates with honest, design-based variance under a clustered multistage structure, unlike convenience or snowball samples.
- Naturally suited to mobile migrant groups, who are unlisted but observable in space and time at hiring sites, remittance shops, consulates, and gathering places.
- The attendance-frequency weighting explicitly corrects the over-representation of the most venue-active members, improving representativeness of the final estimate.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use time-location sampling when the target population has no usable list frame yet predictably congregates at identifiable venues during identifiable time periods, and you need probability-based estimates rather than a convenience sample. It is well suited to mobile and hard-to-reach migrant groups — day laborers, recent arrivals, seasonal workers, migrants using remittance shops, consulates, markets, or places of worship — provided formative fieldwork can enumerate the venues and approach attendees safely and ethically. It is also appropriate when you want estimable variance and known selection probabilities, which respondent-driven and snowball approaches cannot fully guarantee. The method is less suitable when the population is genuinely dispersed and never gathers, when a large share is reachable only at private homes or fully online, or when venue access is denied or unsafe; in those cases network-based estimation or multiplicity sampling may be preferable. It also presumes you can measure attendance frequency reliably, since the attendance weighting depends on it.
Strengths & limitations
- Creates a true probability-sampling frame for populations that have no list frame, by sampling venue-day-time units with known selection probabilities.
- Yields population-level estimates with honest, design-based variance under a clustered multistage structure, unlike convenience or snowball samples.
- Naturally suited to mobile migrant groups, who are unlisted but observable in space and time at hiring sites, remittance shops, consulates, and gathering places.
- The attendance-frequency weighting explicitly corrects the over-representation of the most venue-active members, improving representativeness of the final estimate.
- Coverage is bounded by the venue frame: anyone who never visits an enumerated venue has zero chance of selection and is silently excluded.
- Building and maintaining the venue-day-time frame demands intensive, costly formative ethnography that must be repeated as venues open, close, and shift.
- Estimates are biased if attendance frequency is mismeasured, since the weights and therefore the population quantities hinge directly on self-reported attendance.
- Results generalize only to the venue-attending segment of the population, which may differ systematically from members who avoid public venues.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How does time-location sampling differ from respondent-driven sampling?
Both target hidden populations without a list frame, but they exploit different structures. Respondent-driven sampling recruits through the population's own social network, with members referring peers, and corrects for network size. Time-location sampling instead exploits the population's presence in physical space, building a frame of venues, days, and time slots, drawing those units at random, and intercepting attendees on site. Time-location sampling tends to give cleaner, design-based selection probabilities and variance, but only for people who visit enumerated venues; respondent-driven sampling can reach more isolated members through their contacts but relies on stronger modeling assumptions about referral and network size.
Why must respondents be weighted by how often they attend venues?
Because attendance is the only way into the sample, the chance of being selected rises with how frequently and how widely a person attends the venues on the frame. Someone present at venues every day is far more likely to be intercepted than someone who appears once a month, so an unweighted sample over-represents the most venue-active members. The method asks each respondent about their attendance frequency and assigns a weight inversely proportional to their resulting inclusion probability, so that high-frequency attenders are down-weighted. Without this correction the estimates describe the busiest venue-goers rather than the population as a whole.
What is the biggest threat to validity in a venue-based design?
Coverage of the venue-day-time frame. The method can only sample people who visit an enumerated venue during an enumerated time slot, so any member who never appears at a listed venue has zero probability of selection and is invisible to the estimate. If formative fieldwork misses important venues, or if a subgroup deliberately avoids public gathering places, the sample will be systematically unrepresentative no matter how rigorously the random selection and weighting are executed. This is why thorough, continually updated ethnographic frame construction is treated as the foundational and most consequential step of the entire pipeline.
Sources
- 1.Muhib, F. B., Lin, L. S., Stueve, A., Miller, R. L., Ford, W. L., Johnson, W. D., & Smith, P. J. (2001). A Venue-Based Method for Sampling Hard-to-Reach Populations. Public Health Reports, 116(Suppl 1), 216-222.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Time-Location Sampling. ScholarGate. https://scholargate.app/migration-studies/time-location-sampling