ntxiv tswv yim

Seem no yog tsim los siv raws li ib tug siv, es tsis yuav tsum tau nyeem raws li ib tug narrative.

• Introduction (Section 3.1)

Muaj ntau ntawm cov ntxhais nyob rau hauv tshooj no kuj tau echoed nyob rau hauv tsis ntev los no cov Medal chaw nyob rau hauv American Association of Public tswv yim kev tshawb fawb (AAPOR), xws li Dillman (2002) , Newport (2011) , Santos (2014) , thiab Link (2015) .

Yog xav paub ntxiv keeb kwm tom qab txog txoj kev loj hlob ntawm daim ntawv ntsuam xyuas kev tshawb fawb, saib Smith (1976) thiab Converse (1987) . Yog xav paub ntxiv nyob rau hauv lub tswv yim ntawm peb eras ntawm daim ntawv ntsuam xyuas kev tshawb fawb, saib Groves (2011) thiab Dillman, Smyth, and Christian (2008) (uas so li peb cov eras me ntsis txawv).

Ib tug ncov hauv txoj kev hloov los ntawm tus thawj rau lub thib ob era nyob rau hauv daim ntawv ntsuam xyuas kev tshawb fawb yog Groves and Kahn (1979) , uas yog ib ncauj lus kom ntxaws taub hau-rau-lub taub hau sib piv nyob nruab nrab ntawm ib tug lub ntsej muag-rau-lub ntsej muag thiab xov tooj daim ntawv ntsuam xyuas. Brick and Tucker (2007) zoo rov qab rau lub keeb kwm kev loj hlob ntawm random tug dialing zauv txoj kev.

Yog xav paub ntxiv yuav ua li cas daim ntawv ntsuam xyuas kev tshawb fawb tau hloov nyob rau hauv lub yav dhau los nyob rau hauv teb rau cov kev hloov nyob rau hauv zej zog, saib Tourangeau (2004) , Mitofsky (1989) , thiab Couper (2011) .

• Nug vs. tsom kwm (Section 3.2)

Kawm txog internal lub xeev los ntawm cov lus nug yuav ua tau problematic vim hais tias tej zaum cov neeg lawv tus kheej tsis paub txog lawv nrog lub xeev. Piv txwv li, Nisbett and Wilson (1977) muaj ib tug zoo kawg nkaus ntawv uas muaj cov evocative title: "qhia ntau tshaj peb yuav paub: Lus ceeb toom rau kev puas hlwb muaj dab." Nyob rau hauv daim ntawv tus sau phau ntawv xaus: "kawm yog tej zaum (a) unaware ntawm lub hav zoov ntawm ib tug stimulus uas ceeb los ib lub teb, (b) unaware ntawm lub hav zoov ntawm cov lus teb, thiab (c) unaware tias cov stimulus tau cuam tshuam rau cov lus teb. "

Rau nqe lus uas soj ntsuam yuav tsum xav cai coj tus cwj pwm mus qhia tus cwj pwm los yog xeeb ceem, saib Baumeister, Vohs, and Funder (2007) (psychology) thiab Jerolmack and Khan (2014) thiab cov lus teb (Maynard 2014; Cerulo 2014; Vaisey 2014; Jerolmack and Khan 2014) (Sociology). Qhov txawv ntawm nug thiab tsom kwm kuj tshwm sim nyob rau hauv kev lag luam, qhov twg soj ntsuam tham txog hais thiab qhia nyiam. Piv txwv li, ib tug soj ntsuam yuav nug cov neeg seb lawv xav noj ice cream los yog mus gym (hais nyiam) los yog cov kev tshawb fawb yuav soj ntsuam li cas feem ntau cov neeg noj ice cream thiab mus rau lub gym (qhia nyiam). Muaj yog sib sib zog nqus tsis ntseeg tau pib ntawm tej hom ntawm hais nyiam cov ntaub ntawv nyob rau hauv kev lag luam (Hausman 2012) .

A ntsiab tseem ceeb ntawm cov sib cav tswv yim yog hais tias qhia tus cwj pwm yog tsis yog. Tab sis, txiav kaw tus cwj pwm tej zaum yuav tsis muaj tseeb, tej zaum yuav tsis raug sau rau ib cov qauv ntawm kev txaus siab, thiab tej zaum yuav tsis yoojyim rau kev soj ntsuam. Yog li, nyob rau hauv tej lub sijhawm, kuv xav hais tias qhia tus cwj pwm yuav pab tau. Ntxiv mus, ib tug thib ob lub ntsiab ntsiab los ntawm cov sib cav tswv yim yog hais tias txog tus cwj pwm txawv, kev txawj ntse, kev ntshaw, thiab kev xav yog tsis yog. Tab sis, yog tias cov lus qhia txog cov internal lub xeev yuav tsum tau los ntawm kev soj ntsuam-tog twg los pab piav txog tej cwj pwm los yog raws li tus tshaj plaws yuav tsum tau piav qhia-ces nug tej zaum yuav tsim nyog.

• Tag nrho cov daim ntawv ntsuam xyuas kev ua yuam kev (Section 3.3)

Rau hauv phau ntawv ntev kev kho mob rau tag nrho cov daim ntawv ntsuam xyuas kev ua yuam kev, saib Groves et al. (2009) los yog Weisberg (2005) . Rau ib tug keeb kwm ntawm kev loj hlob ntawm tag nrho cov daim ntawv ntsuam xyuas kev ua yuam kev, saib Groves and Lyberg (2010) .

Nyob rau hauv cov nqe lus ntawm kev sawv cev, ib tug zoo kawg taw qhia rau cov teeb meem ntawm uas tsis yog-teb thiab uas tsis yog-teb kev tsis ncaj ncees yog lub National Kev tshawb fawb Council daim ntawv qhia txog Nonresponse nyob rau hauv Social Science kuaj: A tshawb fawb Agenda (2013) . Lwm pab txheej txheem cej luam yog muab los ntawm (Groves 2006) . Tsis tas li ntawd, tag nrho tshwj xeeb cov teeb meem ntawm cov phau ntawv Journal of Official Statistics, Public tswv yim txhuab peb lub hlis, thiab annals ntawm lub American Academy of Political thiab Social Science tau luam tawm nyob rau hauv lub npe uas tsis yog-teb. Thaum kawg, muaj tau ntau ntau txoj kev xam lub teb nqi; le caag yog piav nyob rau hauv kom meej nyob rau hauv ib daim ntawv qhia los ntawm lub American Association of Public tswv yim tshawb fawb (AAPOR) (Public Opinion Researchers} 2015) .

Lub 1936 Zajlus zom poll tau raug kawm nyob rau hauv kom meej (Bryson 1976; Squire 1988; Cahalan 1989; Lusinchi 2012) . Nws tau kuj tau siv raws li ib tug zaj lus piv txwv rau ceeb toom tiv thaiv haphazard cov ntaub ntawv sau (Gayo-Avello 2011) . Nyob rau hauv 1936, George Gallup siv ib tug ntau sophisticated daim ntawv ntawm cov zauv, thiab muaj peev xwm yuav tsim ntau yog kwv yees nrog ib tug ntau me me qauv. Gallup tau zoo tshaj lub Zajlus zom yog ib tug milestone txoj kev loj hlob ntawm daim ntawv ntsuam xyuas kev tshawb fawb (Converse 1987, Ch 3; Ohmer 2006, Ch 4; Igo 2008, Ch 3) .

Nyob rau hauv cov nqe lus ntawm kev ntsuas, ib tug zoo kawg thawj pab vim questionnaires yog Bradburn, Sudman, and Wansink (2004) . Rau ib tug ntau tshaj kev kho mob teem hais rau tus cwj pwm cov lus nug, saib Schuman and Presser (1996) . Ntxiv rau pre-xeem cov lus nug yog muaj nyob rau hauv Presser and Blair (1994) , Presser et al. (2004) , thiab Tshooj 8 ntawm Groves et al. (2009) .

Lub classic, phau ntawv-ntev kev kho mob ntawm cov lag luam tawm-off ntawm daim ntawv ntsuam xyuas cov nqi thiab daim ntawv ntsuam xyuas uas tsis yog Groves (2004) .

• Leej twg nug (Section 3.4)

Classic phau ntawv-ntev kev kho mob ntawm tus qauv yuav qhov muaj tseeb piv txwv thiab kwv yees yog Lohr (2009) (ntau introductory) thiab Särndal, Swensson, and Wretman (2003) (ntau tshaj). Ib tug classic phau ntawv-ntev kev kho mob ntawm post-stratification thiab lwm yam kev yog Särndal and Lundström (2005) . Nyob rau hauv ib co cov muaj hnub nyoog chaw, soj ntsuam paub ib tug heev me ntsis txog cov-teb, uas yog tsis yog feem ntau yeej muaj tseeb nyob rau hauv lub yav dhau los. Ntau yam ntaub ntawv uas tsis yog-teb hloov tau thaum kev soj ntsuam ntawm muaj ntaub ntawv hais txog tsis yog-neeg (Kalton and Flores-Cervantes 2003; Smith 2011) .

Lub Xbox txoj kev tshawb no ntawm Wang et al. (2015) siv ib tug txheej txheem uas hu multilevel regression thiab post-stratification (MRP, tej zaum hu ua "Aib P") uas tso cai rau kev soj ntsuam los laij cov nqi cell txhais tau tias txawm thaum muaj ntau ntau yam, muaj ntau yam hlwb. Txawm hais tias muaj yog ib co sib cav tswv yim hais txog qhov zoo ntawm lub kwv yees los ntawm cov txheej txheem no, no mas, zoo li ib tug pheej cheeb tsam tshawb. Cov txheej txheem twb xub siv nyob rau hauv Park, Gelman, and Bafumi (2004) , thiab muaj lawm tom ntej siv thiab sib cav tswv yim (Gelman 2007; Lax and Phillips 2009; Pacheco 2011; Buttice and Highton 2013; Toshkov 2015) . Yog xav paub ntxiv nyob rau hauv kev twb kev txuas ntawm ib tug neeg tes taw hnyav li thiab cell-raws li tes taw hnyav li saib Gelman (2007) .

Rau lwm tus le caag los weighting web kuaj, saib Schonlau et al. (2009) , Valliant and Dever (2011) , thiab Bethlehem (2010) .

Qauv piv tau npaj los ntawm Rivers (2007) . Bethlehem (2015) cav hais tias cov kev ua tau zoo ntawm cov qauv piv yuav ua tau zoo ib yam li lwm yam piv txwv le caag (xws li, stratified zauv) thiab lwm yam kev hloov le caag (xws li, post-stratification). Yog xav paub ntxiv nyob rau hauv online panels, saib Callegaro et al. (2014) .

Tej zaum kev soj ntsuam tau pom hais tias muaj tseeb kuaj thiab uas tsis yog-yuav qhov muaj tseeb kuaj paib kev kwv yees ntawm zoo xws li cov zoo (Ansolabehere and Schaffner 2014) , tab sis lwm yam los sib piv tau pom tias uas tsis yog-yuav qhov muaj tseeb kuaj ua zuj zus (Malhotra and Krosnick 2007; Yeager et al. 2011) . Ib tug tau yog vim li cas rau cov kev txawv no yog tias tsis yog-yuav qhov muaj tseeb kuaj tau zoo tshaj lub sij hawm. Rau ib tug ntau phem heev saib ntawm uas tsis yog-yuav qhov muaj tseeb piv txwv txoj kev saib rau lub AAPOR Task Force on Tsis-yuav qhov muaj tseeb zauv (Baker et al. 2013) , thiab kuv kuj pom zoo kom nyeem ntawv lub tswv yim uas raws li cov lus xaus tsab ntawv ceeb toom.

Rau ib tug meta-tsom xam rau cov nyhuv ntawm weighting los txo kev tsis ncaj ncees nyob rau hauv uas tsis yog-yuav qhov muaj tseeb kuaj, saib Rooj 2.4 nyob rau hauv Tourangeau, Conrad, and Couper (2013) , uas ua rau yus tus sau phau ntawv los xaus "kev kho me ntsis zoo li yuav pab tau tab sis fallible corrections. . . "

• Yuav ua li cas mus nug (Section 3.5)

Conrad and Schober (2008) muab ib edited volume hu Envisioning lub Survey sib tham ntawm lub neej yav tom ntej, thiab nws chaw nyob muaj ntau tus ntxhais nyob rau hauv seem no. Couper (2011) chaw nyob zoo li cov ntxhais, thiab Schober et al. (2015) muaj ib tug zoo piv txwv ntawm yuav ua li cas cov ntaub ntawv sau txoj kev uas yog txiav mus rau ib tug tshiab qhov chaw yuav ua nyob rau hauv ntau dua zoo cov ntaub ntawv.

Rau lwm nthuav piv txwv ntawm kev siv Facebook apps rau social science tshawb fawb, saib Bail (2015) .

Yog xav paub ntxiv tswv yim rau kev ua cov ntawv tshawb fawb ib tug zoo thiab tseem ceeb rau cov neeg, saib kev ua hauj lwm nyob rau hauv lub sawb tsim txoj kev (Dillman, Smyth, and Christian 2014) .

Stone et al. (2007) muaj ib phau ntawv ntev kev kho mob ntawm ecological momentary kev ntsuam xyuas thiab cov kev txoj kev.

• Kuaj uas txuas mus rau lwm cov ntaub ntawv (Section 3.6)

Judson (2007) piav txog tus txheej txheem ntawm combining cov ntawv tshawb fawb thiab kev cov ntaub ntawv raws li "ntaub ntawv kev koom ua ke," tham txog tej zoo ntawm no mus kom ze, thiab muaj ib co piv txwv.

Lwm txoj kev uas soj ntsuam yuav siv cov khoom siv thiab kev cov ntaub ntawv yog ib tug piv txwv ncej rau cov neeg uas tej yam ntxwv. Txawm li cas los, nkag cov ntaub ntawv yuav tsum tau siv ib tug zauv ncej kuj yuav tsim cov lus nug txog kev ceev ntiag tug (Beskow, Sandler, and Weinberger 2006) .

Hais txog amplified thov, no mus kom ze tsis yog raws li tshiab raws li nws tej zaum yuav tshwm sim li ntawm yuav ua li cas kuv twb tau piav nws. Qhov no mus kom ze muaj sib sib zog nqus sib txuas rau peb loj loj nyob rau hauv statistics qauv-raws li post-stratification (Little 1993) , imputation (Rubin 2004) , thiab cheeb tsam kwv yees (Rao and Molina 2015) . Nws no kuj hais txog kev siv cov surrogate variables nyob rau hauv kev kho mob kev tshawb fawb (Pepe 1992) .

Nyob rau hauv tas li ntawd mus rau lub koom teeb meem hais txog tus txheejtxheem lub digital txoj lw cov ntaub ntawv, amplified thov kuj yuav siv tau los xav hais tias sensitive sijhawm uas cov neeg yuav tsis xaiv mus qhia nyob rau hauv ib daim ntawv ntsuam xyuas (Kosinski, Stillwell, and Graepel 2013) .

Tus nqi thiab lub sij hawm kwv yees nyob rau hauv Blumenstock, Cadamuro, and On (2015) xa ntxiv rau nce mus nce los nqi-tus nqi ntawm ib tug ntxiv daim ntawv ntsuam xyuas thiab tsis muaj xws li tsau cov nqi xws li cov nqi los ntxuav thiab ua tus hu cov ntaub ntawv. Nyob rau hauv kev, amplified thov yuav tej zaum muaj siab tas cov nqi thiab cov uas tsis muaj nce mus nce los nqi zoo xws li cov digital thwmsim (saib Tshooj 4). Yog xav paub ntxiv nyob rau hauv cov ntaub ntawv siv nyob rau hauv Blumenstock, Cadamuro, and On (2015) daim ntawv yog nyob rau hauv Blumenstock and Eagle (2010) thiab Blumenstock and Eagle (2012) . Le caag los ntawm ntau yam imputuation (Rubin 2004) tej zaum yuav pab tau capture uncertainty nyob rau hauv kev kwv yees los ntawm amplified thov. Yog hais tias soj ntsuam ua amplified nug tsuas pab txog sau ua ib pawg suav, es ib tug neeg-theem sijhawm, ces tus le caag nyob rau hauv King and Lu (2008) thiab Hopkins and King (2010) tej zaum yuav pab tau. Yog xav paub ntxiv txog lub tshuab kev kawm le caag nyob rau hauv Blumenstock, Cadamuro, and On (2015) , saib James et al. (2013) (ntau introductory) los yog Hastie, Tibshirani, and Friedman (2009) (ntau tshaj). Lwm nrov tshuab kev kawm phau ntawv yog Murphy (2012) .

Hais txog enriched thov, cov kev tshwm sim nyob rau hauv Ansolabehere thiab Hersh (2012) pob khawm rau ob tug tseem ceeb kauj ruam: 1) muaj peev xwm ntawm Catalist muab ntau disparate cov ntaub ntawv qhov chaw tsim ib tug muaj tseeb tswv datafile thiab 2) lub peev xwm ntawm Catalist mus txuas lub daim ntawv ntsuam xyuas cov ntaub ntawv rau nws tus tswv datafile. Yog li ntawd, Ansolabehere thiab Hersh xyuas txhua yam ntawm cov kauj ruam kom zoo zoo.

Yuav kom ua tau ib lub tswv datafile, Catalist combines thiab harmonizes ntaub ntawv los ntawm ntau qhov chaw sib txawv xws li: ntau yam pov ntawv xaiv tsa cov ntaub ntawv snapshots los ntawm txhua lub xeev, cov ntaub ntawv los ntawm lub Post Office lub National Hloov Chaw Nyob Registry, thiab cov ntaub ntawv los ntawm lwm lub unspecified coj mus muag chaw muab kev pab. Lub gory paub meej txog yuav ua li cas tag nrho cov no tu thiab cov phiajcim tshwm sim yog tshaj ntawd tus uas duav txhua yam ntawm phau ntawv no, tab sis tus txheej txheem no, tsis muaj teeb meem yuav ua li cas ceev faj, yuav propagate uas tsis nyob rau hauv cov ntaub ntawv qhov chaw thiab yuav ua kom paub tsis. Txawm tias Catalist yog kam tham nws cov ntaub ntawv ua thiab muab ib co ntawm nws cov nqaij nyoos cov ntaub ntawv, nws yog tsuas tsis yooj yim sua rau kev soj ntsuam los saib xyuas tag nrho cov ntaub ntawv Catalist pipeline. Tiam sis, tus soj ntsuam ntawm nyob rau hauv ib qhov teeb meem nyob qhov twg lub Catalist cov ntaub ntawv cov ntaub ntawv muaj ib co tsis paub hais tias, thiab tej zaum unknowable, nqi ntawm kev ua yuam kev. Qhov no yog ib tug loj kev txhawj xeeb vim hais tias ib tug critic tej zaum yuav xav tias tus loj sib txawv ntawm cov ntawv ntsuam xyuas cov ntawv qhia rau cov CCES thiab tus cwj pwm nyob rau hauv lub Catalist tswv cov ntaub ntawv cov ntaub ntawv tau tshwm sim los ntawm uas tsis nyob rau hauv lub tswv cov ntaub ntawv cov ntaub ntawv, tsis yog misreporting los ntawm cov neeg.

Ansolabehere thiab Hersh coj ob tug sib txawv le caag los hais cov ntaub ntawv zoo kev txhawj xeeb. Ua ntej, nyob rau hauv tas li ntawd mus muab piv rau lawv tus kheej qhia pov ntawv xaiv tsa rau xaiv nyob rau hauv lub Catalist tswv cov ntaub ntawv, cov kev soj ntsuam kuj piv nws tus kheej-qhia tog, haiv neeg, tawm suab pov ntawv raws li txoj cai (xws li, sau npe los yog tsis sau npe) thiab pov ntawv xaiv tsa txoj kev (xws li, nyob rau hauv tus neeg, ntawv daim ntawv xaiv tsa, thiab lwm yam) rau cov neeg tseem ceeb nyob rau hauv cov Catalist databases. Rau cov plaub demographic variables, cov neeg tshawb fawb pom ntau npaum li cas feem ntau ntawm cov kev pom zoo ntawm daim ntawv ntsuam xyuas daim ntawv qhia thiab cov ntaub ntawv nyob rau hauv lub Catalist tswv cov ntaub ntawv tshaj rau kev xaiv tsa. Yog li ntawd, Catalist tswv cov ntaub ntawv cov ntaub ntawv zoo nkaus li muaj siab zoo cov ntaub ntawv rau qhov zoo lwm yam tshaj li pov ntawv xaiv tsa, tawm tswv yim hais tias nws tsis yog cov neeg pluag zuag qhia tag nrho. Ob, nyob rau hauv ib feem siv cov ntaub ntawv los ntawm Catalist, Ansolabehere thiab Hersh tsim peb sib txawv kev ntsuas ntawm zoo ntawm lub nroog pov ntawv xaiv tsa cov ntaub ntawv, thiab lawv pom tias tus nqi kwv yees tus nqi ntawm tshaj-qhia ntawm pov npav yog yeej tseem zoo ncaj rau ib yam ntawm cov ntaub ntawv no zoo ntsuas, ib tug nrhiav tau hais tias hais tias lub siab cov nqi ntawm cov tshaj-qhia tsis tau uas tau tsav los ntawm cov cheeb nroog nrog unusually tsawg cov ntaub ntawv zoo.

Muab cov creation ntawm no master pov ntawv xaiv tsa cov ntaub ntawv, lub thib ob qhov chaw ntawm tej uas tsis yog qhia lawv txuas tus daim ntawv ntsuam xyuas cov ntaub ntawv mus rau nws. Piv txwv li, yog hais tias qhov linkage yog ua li cas incorrectly nws yuav ua rau ib tug tshaj-kwv yees ntawm qhov txawv ntawm qhia thiab validated pov ntawv xaiv tsa tus cwj pwm (Neter, Maynes, and Ramanathan 1965) . Yog hais tias txhua tus neeg muaj ib tug ruaj khov, tshwj xeeb qhia tau tias yog nyob rau hauv ob qho tag nrho cov ntaub ntawv qhov chaw, ces linkage yuav tsis tseem ceeb. Nyob rau hauv lub US thiab feem ntau lwm lub teb chaws, txawm li cas los, yog tsis muaj universal qhia. Ntxiv mus, txawm yog hais tias muaj xws li ib cim cov neeg tej zaum yuav tos muab nws mus rau daim ntawv ntsuam xyuas kev soj ntsuam! Yog li, Catalist tau ua linkage siv imperfect identifiers, nyob rau hauv cov ntaub ntawv no plaub daim ntawm cov ntaub ntawv hais txog txhua tus neeg teb: lub npe, yog txivneej los pojniam, me nyuam yug xyoo, thiab lub tsev chaw nyob. Piv txwv li, Catalist yuav tsum tau txiav txim siab yog hais tias tus Homie J Simpson nyob rau hauv lub CCES yog tus tib neeg raws li lub Homer Jay Simpson nyob rau hauv lawv cov tswv cov ntaub ntawv cov ntaub ntawv. Nyob rau hauv kev xyaum, piv yog ib tug yooj yim thiab ntxhov txheej txheem, thiab, yuav ua rau teeb meem loj rau cov kev soj ntsuam, Catalist xam tau tias yog nws txuam cov txheej txheem yuav tsum tau proprietary.

Nyob rau hauv thiaj li yuav validate cov txuam algorithms, lawv tso siab rau ob txoj kev sib tw. Ua ntej, Catalist koom nyob rau hauv ib txoj kev sib txeeb uas tau khiav los ntawm ib tug neeg sab nraud, peb-tog: lub MITRE Corporation. MITRE muab tag nrho cov neeg ob nrov cov ntaub ntawv cov ntaub ntawv yuav tsum tau sib phim, thiab txawv pab pawg sib rov qab mus rau MITRE qhov zoo tshaj plaws txuam. Vim hais tias MITRE nws tus kheej paub qhov tseeb txuam lawv muaj peev xwm ua tau lub paab. Ntawm 40 cov tuam txhab uas competed, Catalist tuaj nyob rau hauv ob qhov chaw. Qhov no zoo ywj siab, peb-tog kev luj xyuas ntawm proprietary technology yog heev tsawg tsawg thiab incredibly zoo; nws yuav tsum tau muab peb ntseeg hais tias Catalist tus txuam cov txheej txheem no yeej tseem zoo nyob rau lub xeev-of-the-art. Tab sis yog lub xeev-of-the-art zoo txaus? Nyob rau hauv tas li ntawd mus rau qhov txuam sib txeeb, Ansolabehere thiab Hersh tsim lawv tus kheej txuam kev sib tw rau Catalist. Los ntawm ib tug ua ntej lawm project, Ansolabehere thiab Hersh tau sau pov ntawv xaiv tsa cov ntaub ntawv los ntawm Florida. Lawv muab ib co ntawm cov ntaub ntawv no nrog ib co ntawm lawv cov teb redacted rau Catalist thiab ces piv Catalist tus ceeb toom ntawm no teb mus rau lawv tiag tiag qhov tseem ceeb. Qhov zoo ces, Catalist tus qhia nyob ze rau qhov nyiaj txiav qhov tseem ceeb, qhia hais tias Catalist yuav phim ib nrab pov ntawv xaiv tsa cov ntaub ntawv mus rau lawv cov tswv cov ntaub ntawv cov ntaub ntawv. Cov ob txoj kev sib tw, ib tug los ntawm ib tug thib peb-tog thiab ib tug los Ansolabehere thiab Hersh, muab peb ntau cog qoob loo nyob rau hauv lub Catalist txuam algorithms, txawm tias peb yuav tsis saib xyuas lawv cov caij nyoog ua raws li peb tus kheej.

Muaj ntau yav dhau los cov me nyuam no mus validate pov npav. Rau ib tug txheej txheem cej luam ntawm uas cov ntaub ntawv no, mus saib Belli et al. (1999) , Berent, Krosnick, and Lupia (2011) , Ansolabehere and Hersh (2012) , thiab Hanmer, Banks, and White (2014) .

Nws yog ib qho tseem ceeb kom nco ntsoov tias txawm hais tias nyob rau hauv cov ntaub ntawv no soj ntsuam tau txhawb los ntawm lub zoo ntawm cov ntaub ntawv los ntawm Catalist, lwm yam kev soj ntsuam los ntawm coj mus muag muag khoom tau tsawg siab. Soj ntsuam tau pom tsis zoo thaum cov ntaub ntawv los ntawm ib tug daim ntawv ntsuam xyuas rau ib tug neeg-cov ntaub ntawv los ntawm Marketing Systems Group (uas nws tus kheej merged ua ke cov ntaub ntawv los ntawm peb cov chaw kho mob: Acxiom, Experian, thiab InfoUSA) (Pasek et al. 2014) . Uas yog, cov ntaub ntawv cov ntaub ntawv tsis phim daim ntawv ntsuam xyuas cov lus teb uas soj ntsuam yuav tsum muaj tseeb, lub datafile tau ploj lawm cov ntaub ntawv rau ib tug loj tus naj npawb ntawm cov lus nug, thiab cov uas ploj lawm cov ntaub ntawv tus qauv ze rau kom qhia daim ntawv ntsuam xyuas tus nqi (nyob rau hauv lwm yam lus uas ploj lawm cov ntaub ntawv yog systematic , tsis random).

Yog xav paub ntxiv nyob rau hauv cov ntaub ntawv linkage ntawm cov ntawv tshawb fawb thiab kev cov ntaub ntawv, saib Sakshaug and Kreuter (2012) thiab Schnell (2013) . Yog xav paub ntxiv nyob rau hauv cov ntaub ntawv linkage nyob rau hauv Feem ntau, saib Dunn (1946) thiab Fellegi and Sunter (1969) (keeb kwm lub) thiab Larsen and Winkler (2014) (niaj hnub). Zoo ib yam li le caag kuj tau tsim nyob rau hauv computer science nyob rau hauv lub npe xws li cov ntaub ntawv deduplication, Piv txwv li kev qhia kom paub, lub npe txuam, duplicate nrhiav kom tau, thiab theej tawm cov ntaub ntawv nrhiav kom tau (Elmagarmid, Ipeirotis, and Verykios 2007) . Tseem muaj kev ceev ntiag tug khaws cia le caag los sau linkage uas tsis yuav tsum tau cov kis ntawm tus kheej yam ntaub ntawv (Schnell 2013) . Soj ntsuam ntawm nyob Facebook tsim ib txoj kev los probabilisticsly txuas lawv cov ntaub ntawv mus pov ntawv xaiv tsa tus cwj pwm (Jones et al. 2013) ; no linkage twb ua los soj ntsuam ib tug xyaum ua tej yam uas kuv yuav qhia rau koj txog nyob rau hauv Tshooj 4 (Bond et al. 2012) .

Lwm cov piv txwv ntawm linking ib tug loj-scale social daim ntawv ntsuam xyuas rau tsoom fwv kev tswj cov ntaub ntawv los ntawm lub Health thiab Nyiaj Survey thiab lub Social Security Administration. Yog xav paub ntxiv nyob rau hauv uas txoj kev kawm no, xws li ntaub ntawv qhia txog kev tso cai los txoj kev, saib Olson (1996) thiab Olson (1999) .

Tus txheej txheem ntawm combining ntau qhov chaw ntawm kev cov ntaub ntawv mus rau hauv ib tug tswv datafile-tus txheej txheem uas Catalist neeg ua hauj lwm-yog ib qho nyob rau hauv lub statistical chaw ua hauj lwm ntawm ib co teb chaws tsoom fwv. Ob tug soj ntsuam los ntawm Statistics Sweden tau sau ib phau ntawv ncauj lus kom ntxaws rau lub npe (Wallgren and Wallgren 2007) . Rau ib qho piv txwv ntawm no mus kom ze nyob rau hauv ib lub nroog nyob rau hauv lub tebchaws United States (Olmstead County, Minnesota; lub tsev ntawm lub Mayo Clinic), saib Sauver et al. (2011) . Yog xav paub ntxiv nyob rau hauv uas tsis muaj peev xwm tshwm sim nyob rau hauv kev tswj cov ntaub ntawv, saib Groen (2012) .