Ranked Set Sampling
Ranked Set Sampling (RSS) · Also known as: RSS
Ranked Set Sampling (RSS) is a data collection method introduced by G. A. McIntyre in 1952 that improves estimation efficiency when visual ranking of units is easier or cheaper than actual measurement. By deliberately selecting and measuring units that are ranked as most likely to yield desired outcomes, RSS reduces variance compared to simple random sampling while maintaining unbiasedness.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use RSS when measurement is costly or difficult but ranking is cheap and reasonably accurate. Common in agriculture (crop yield, plant quality), environmental sampling (water quality, soil properties), forestry (tree volume, wood density), and any setting where visual or field assessment precedes lab analysis. Requires that ranking is at least moderately correlated with the measurement variable; if ranking is random, efficiency gains vanish.
Strengths & limitations
- Dramatically reduces variance compared to simple random sampling without increasing sample size
- Maintains unbiasedness of estimators
- Practical in field settings where rough ranking is fast and free
- Improves precision of confidence intervals and hypothesis tests
- Scalable to multiple characteristics simultaneously
- Quality of efficiency gains depends entirely on correlation between ranking and true measurement
- Set size m and number of sets r must be chosen before sampling begins
- Poor ranking (weak correlation) negates efficiency advantages
- Assumes ranking is consistent and reliable, even if rough
Frequently asked
How does RSS differ from stratified sampling?
Stratified sampling divides the population into known subgroups (strata) before sampling. RSS uses ranking judgment to create strata after preliminary observation. Stratification is predefined; RSS strata emerge from ranking.
What happens if my ranking is very poor?
If ranking is uncorrelated with the measurement variable, RSS degrades to simple random sampling in efficiency. The estimators remain unbiased, but you lose the variance reduction benefit. Validation studies before full deployment are wise.
Can I use RSS with multiple variables at once?
Yes. Multivariate RSS ranks units on a composite criterion (e.g., a visual score combining multiple attributes). Efficiency gains apply to the primary variable of interest, though trade-offs may arise for secondary variables.
How do I choose the set size m?
Set size is usually 3–5 units per ranked set. Larger m demands more ranking effort and can introduce ranking errors. Practical constraints and the cost of ranking relative to measurement determine the optimal m.
Is RSS unbiased?
Yes, under the standard ranking mechanism, the sample mean is an unbiased estimator of the population mean. The gain from RSS is reduced variance, not bias correction.
Sources
- McIntyre, G. A. (1952). A method for unbiased selective sampling using ranked sets. Australian Journal of Agricultural Research, 3(4), 385–390. DOI: 10.1071/ar9520385 ↗
- Takahasi, K., & Wakimoto, K. (1968). On unbiased estimates of population mean based on the sample stratified by successive groups. Annals of the Institute of Statistical Mathematics, 20(1), 1–31. DOI: 10.1007/bf02911622 ↗
- Wolfe, D. A. (1992). Illustrated concepts of ranked-set sampling. The American Statistician, 46(4), 229–232. link ↗
How to cite this page
ScholarGate. (2026, June 3). Ranked Set Sampling (RSS). ScholarGate. https://scholargate.app/en/sampling/ranked-set-sampling
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Cluster SamplingSurvey Methodology↔ compare
- Double SamplingSampling↔ compare
- Stratified SamplingSurvey Methodology↔ compare
- Systematic SamplingSurvey Methodology↔ compare