This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Adaptive Cluster Sampling Author(s): Steven K. Thompson Reviewed work(s): Source: Journal of the American Statistical Association, Vol. 85, No. 412 (Dec., 1990), pp. 1050- 1059 Published by: American Statistical Association Stable URL: http://www.jstor.org/stable/2289601 . Accessed: 07/12/2011 14:35 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact firstname.lastname@example.org. American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Statistical Association. http://www.jstor.org AdaptiveClusterSampling STEVENK. THOMPSON* In manyreal-world sampling situations, researchers wouldliketobe abletoadaptively increasesampling effort inthevicinity ofobservedvaluesthatarehighorotherwise interesting. Thisarticledescribes sampling designs inwhich,whenever anobserved valueofa selectedunitsatisfies a condition ofinterest, additionalunitsareaddedto thesamplefromtheneighborhood of thatunit.Ifanyoftheseadditional unitsatisfies thecondition, stillmoreunitsmaybe added.Sampling designsuchas these, inwhich theselection procedure isallowedtodependonobservedvaluesofthevariableofinterest, areincontrast toconventional designs, inwhichtheentire selection ofunitstobeincluded inthesamplemaybedetermined priortomaking anyobservations. Becausetheadaptiveselectionprocedure introduces biasesintoconventional estimators, severalestimators aregiventhatare designunbiasedforthepopulationmeanwiththeadaptiveclusterdesignsofthisarticle;thatis, theunbiasedness doesnot dependonanyassumptions aboutthepopulation. TheRao-Blackwellmethod isusedtoobtainimproved unbiasedestimators; becauseoftheincompleteness oftheminimal sufficient statistic,morethanoneoftheseimproved estimators areobtained. Simplecriteria aregivendetermining whenadaptivecluster sampling strategies aremoreefficient thansimplerandom sampling ofequivalent samplesize.Motivation forthedesigns inthisarticle isprovidedbya widevariety ofsampling situations infields suchas ecology,geology,andepidemiology. Forexample, ina surveyofa rarebirdspecies,onceindividuals ofthespecies aredetected, additionalobservations at nearbysitesoftenrevealmoreindividuals. In a studyofa contagiousdisease,the additiontothesampleofclosecontactsofinfected individuals revealsa higherthanaverageincidencerate.Theresults and examples in thisarticleshowthatadaptiveclustersampling strategies givelowervariancethanconventional strategies for certaintypesofpopulations and,inparticular, provideanextremely effective wayofsampling rare,clustered populations....
View Full Document
This note was uploaded on 01/28/2012 for the course STATISTICS 3010 taught by Professor Ooz during the Spring '11 term at Cornell University (Engineering School).
- Spring '11