S einhaue , Hans Wal e ; Décieux, Jean Philippe; Siege , Manuel; E e, And eas;
Zinn, Sabine
A icle — Published Ve sion
Es ablishing ap obabili y sample in ac isis con ex :
heexample o Uk ainian e ugees in Ge many in 2022
AS A Wi scha s- und Sozials a is isches A chi
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: S einhaue , Hans Wal e ; Décieux, Jean Philippe; Siege , Manuel; E e, And eas;
Zinn, Sabine (2024) : Es ablishing ap obabili y sample in ac isis con ex : heexample o Uk ainian
e ugees in Ge many in 2022, AS A Wi scha s- und Sozials a is isches A chi , ISSN 1863-8163,
Sp inge , Be lin, Heidelbe g, Vol. 18, Iss. 1, pp. 77-97,
h ps://doi.o g/10.1007/s11943-024-00338-0
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315634
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
ORIGINALVERÖFFENTLICHUNG
h ps://doi.o g/10.1007/s11943-024-00338-0
AS A Wi scha s- und Sozials a is isches A chi (2024) 18:77–97
Es ablishing a p obabili y sample in a c isis con ex :
he example o Uk ainian e ugees in Ge many in 2022
Hans Wal e S einhaue · Jean Philippe Décieux · Manuel Siege ·
And eas E e · Sabine Zinn
Recei ed: 6 July 2023 / Accep ed: 19 Feb ua y 2024 / Published online: 4 Ma ch 2024
© The Au ho (s) 2024
Abs ac Following Russia’s in asion o Uk aine in ea ly 2022, mo e han one
million e ugees ha e a i ed in Ge many. These Uk ainian e ugees di e in many
aspec s om Ge many’s pas o ced mig a ion expe iences and he e exis s an u gen
need o sound da a and in o ma ion o poli ics, p ac i ione s, and academics. In e-
sponse, he IAB-BiB/FReDA-BAMF-SOEP s udy was es ablished o p o ide high-
quali y longi udinal da a ollowing a egis e -based p obabili y sample. We de ail on
an app oach o sampling e ugees in b ie ime, making use o wo di e en eg-
is e s— he Ge man popula ion egis e and he cen al egis e o o eigne s—and
discuss he quali y o he inal sample wi h espec o po en ial selec i i y o pa -
icipa ion in he panel. O e all, we demons a e he bene i s and easibili y o es-
ablishing egis e -based samples e en in he con ex o a geopoli ical c isis and he
necessi y o sound da a wi hin b ie ime ho izons. We p o ide guidance ha can be
ollowed o simila e en s in he u u e.
Keywo ds Uk ainian Re ugees · P obabili y Sampling · Cen al Regis e o
Fo eigne s · Popula ion Regis e s
Hans Wal e S einhaue · Sabine Zinn
Socio-Economic Panel, Ge man Ins i u e o Economic Resea ch, Moh ens aße 58, 10117 Be lin,
Ge many
E-Mail: hws einhaue @diw.de
Jean Philippe Décieux · And eas E e
Mig a ion and Mobili y, Fede al Ins i u e o Popula ion Resea ch,
F ied ich-Ebe -Allee 4, 65185 Wiesbaden, Ge many
Manuel Siege
Resea ch Cen e, Fede al O ice o Mig a ion and Re ugees,
Neumeye s aße 22–26, 90411 Nü nbe g, Ge many
Sabine Zinn
Humbold Uni e si y Be lin, Un e den Linden 6, 10099 Be lin, Ge many
K
78 H. W. S einhaue e al.
1 In oduc ion
On 24 Feb ua y 2022, Russia s a ed i s mili a y in asion o Uk aine; he escala ion
o he a med con lic con inues o a ec la ge pa s o he coun y. In esponse
o he loss o secu i y and p o ec ion, millions o Uk ainians led he coun y in o
neighbo ing coun ies and membe s a es o he Eu opean Union (EU). Re ugees
mos ly led o Poland as neighbo ing coun y. Ge many is hos ing he second la ges
communi y o Uk ainian e ugees wi hin he EU. Be o e he Russian in asion in
Feb ua y 2022, immig an s om Uk aine cons i u ed a compa a i ely small g oup
in Ge many. A he end o 2021, abou 155,000 Uk ainian ci izens we e li ing
in Ge many; his e lec s a long- e m, decade-long inc ease—o abou 2.6% pe
yea —in he numbe o Uk ainians. Thei mig a ion olume was compa a i ely
small, wi h an a e age yea ly immig a ion o a ound 13,000 Uk ainian ci izens
a i ing in Ge many be ween 2012 and 2021 (see Fig. 1).
This pa e n changed undamen ally in 2022. Feb ua y 2022 saw 14,000
Uk ainian ci izens leeing he wa (mos o hem a ound he beginning o he
wa end o Feb ua y). In Ma ch 2022, abou 417,000 Uk ainian ci izens a i ed. Al-
hough numbe s o a i ing Uk ainian e ugees quickly dec eased he ea e , almos
64,000 e ugees a i ed in Ge many du ing June 2022. In less han i e mon hs, he
s ock o Uk ainian ci izens egis e ed inc eased almos se en imes, eaching 1.02
million Uk ainian ci izens egis e ed in Ge many a he end o June 2022.
Fig. 1 De elopmen o immig an lows (EMR) o Uk ainian ci izens o Ge many and s ocks (AZR) o
Uk ainian ci izens in Ge many, 2012–2022 (2012–2021: annual igu es, 2022: mon hly igu es). (Sou ce:
Special analysis om he Ge man Cen al Regis e o Fo eigne s (AZR) ( epo ing da e 30 No embe
2022) and Ge man Popula ion Regis e (EMR))
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 79
The ecep ion o Uk ainian e ugees and he p o ision o op ions o in eg a-
ion pose majo challenges o policymake s, adminis a ion, and socie y. While EU
membe s a es, including no leas Ge many, ha e lea ned a lo om p e ious la ge-
scale a i als o e ugees, he in lux o Uk ainians di e s om he pas in a leas
i e key cha ac e is ics (B ücke e al. 2023): Fi s , he demog aphic composi ion
di e s wi h women, child en, and elde ly people domina ing ecen o ced mig a-
ion lows om Uk aine, because adul men mus emain in Uk aine o mili a y
se ice. Second, e ugees om Uk aine a e g an ed a special legal s a us by Eu o-
pean law (A icle 5(1) o Eu opean Council Di ec i e 2001/55/EC adop ed in July
2001), which enables hem o ob ain a esidence i le ( esidence pe mi acco ding o
§ 24 o he Residence Ac ( empo a y p o ec ion)) in Ge many wi hou an asylum
p ocedu e (Fede al O ice o Mig a ion and Re ugees 2022). Thi d, he accep ance
o Uk ainian e ugees in he hos socie y seems o be highe han in p e ious la ge-
scale a i als o e ugees (D ažano á and Geddes 2022). Fou h, e ugees a i ing
in he pas we e usually alloca ed (S einhaue e al. 2019). Uk ainian e ugees, on
he o he hand, had he oppo uni y o choose hei own place o esidence, p o-
ided hey we e able o ind hei own accommoda ion— o example wi h ela i es,
iends, o acquain ances. Only e ugees who we e unable o p o ide hemsel es
wi h accommoda ion we e dis ibu ed geog aphically (Adam e al. 2021). Fi h,
he sho dis ance be ween Uk aine and Ge many—compa ed o o igin coun ies in
he Middle Eas and Cen al Asia, o example— educes he cos s and h ea s o
a el ou es. The g ea e ease o a eling is also suppo ed by wai ing ain a es
be ween Uk aine and Ge many o e ugees. Addi ionally, geog aphical p oximi y
makes ci cula mig a ion be ween o igin and hos coun ies mo e likely o happen
equen ly.
The olume and speed o o ced mig a ion om Uk aine, oge he wi h he sub-
s an ial di e ences exis ing be ween ecen Uk ainian e ugees compa ed o he
a i al o e ugees in he pas , equi e comp ehensi e knowledge and sound (longi-
udinal) da a o unde s and i s indi idual and socie al consequences. Whe eas se e al
ini ia i es quickly esponded o hose da a equi emen s wi h he implemen a ion o
ad-hoc su eys o Uk ainian e ugees based on eadily a ailable non-p obabili y
samples, he IAB-BiB/FReDA-BAMF-SOEP su ey was launched wi h he aim o
gene a ing high-quali y p obabili y-based longi udinal da a on Uk ainian e ugees
ecen ly a i ing in Ge many. The p ojec is a join wo k be ween he Ge man In-
s i u e o Employmen Resea ch (Ins i u ü A bei sma k und Be u s o schung,
IAB), he Ge man Fede al Ins i u e o Popula ion Resea ch (Bundesins i u ü
Be ölke ungs o schung, BiB), he Resea ch Cen e o he Ge man Fede al O ice
o Mig a ion and Re ugees (Fo schungszen um des Bundesam es ü Mig a ion
und Flüch linge, BAMF-FZ), and he Ge man Socio-Economic Panel (SOEP). The
success o his s udy depends on he quick and e ec i e c ea ion o a p obabili y
sample o Uk ainian e ugees in Ge many. Only wi h a p obabili y-based app oach
i is possible o gene alize he esul s o he s udy o Uk ainian e ugees in Ge many.
This pape p o ides de ails on how we c ea e a andom sample using wo di e en
adminis a i e egis e s: he Ge man popula ion egis e (Einwohne melde egis e ,
EMR) and he Ge man Cen al Regis e o Fo eigne s (Auslände zen al egis e ,
AZR). Using bo h egis e s in combina ion allowed o bene i ing om hei ad-
K
80 H. W. S einhaue e al.
an ages while balancing hei disad an ages. Speci ically, cen ally a ailable basic
in o ma ion om he AZR abou newly egis e ed Uk ainian ci izens om 24 Feb u-
a y 2022, onwa ds p o ided he basis o sampling municipali ies in Ge many hos -
ing Uk ainian e ugees in a i s phase. He e, he AZR p o ides imely in o ma ion
on he o e all numbe o Uk ainian e ugees egis e ed wi h ecep ion acili ies, he
police, o o eigne s’ au ho i ies a he municipal le el. In con as , hese numbe s
a e no cen ally a ailable om he EMRs, because EMRs a e main ained decen al
a he le el o municipali ies. The ad an age o he EMR, howe e , is ha i con ains
indi idual add ess da a o people egis e ed wi hin he municipali y, which was no
ye he case in he AZR a ha ime. Fo his eason, we d aw a sample o munic-
ipali ies in a i s phase using in o ma ion p o ided by he AZR on he numbe o
Uk ainian e ugees. Wi hin he sampled municipali ies, we ask he EMRs o lis all
Uk ainian na ionals aged 18 o 70 who egis e ed a e 24 Feb ua y 2022 oge he
wi h hei add esses. This p ocedu e builds on he example o he “Re ugees in
he Ge man Educa ional Sys em (ReGES)” s udy (S einhaue e al. 2019), bu ex-
ends i o a sample co e ing all Ge man ede al s a es and esponding o immedia e
mig a ion lows.
The pape p esen s he sampling app oach hen discusses i s s eng hs and weak-
nesses wi h pa icula emphasis on po en ial bias due o consen o panel pa ici-
pa ion. Sec ion 2 p o ides an o e iew o gene al sampling echniques o e ugee
popula ions, while Sec . 3 o e s an o e iew o ecen su eys o Uk ainian e ugees.
De ails o ou app oach on sampling e ugees a e discussed in he ollowing wo
sec ions: Sec . 4 p o ides in o ma ion on he egis a ion and alloca ion o Uk ainian
e ugees in Ge many and he sampling o municipali ies om he AZR. In Sec . 5,
he design o he IAB-BiB/FReDA-BAMF-SOEP is in oduced wi h espec o sam-
pling Uk ainian e ugees. Sec ion 7 de ails he ieldwo k and he esponse a es o
he s udy be o e Sec . 7 concludes. The pape shows, i s , ha Ge many has by now
implemen ed an e icien sys em o adminis a i e egis a ion o e ugees ha can
be success ully used o sampling in he con ex o geopoli ical c ises and esul ing
la ge-scale e ugee o mig a ion lows. Second, i shows ha he combina ion o
bo h egis e s— he EMR and he AZR—allows o es ablishing high-quali y p ob-
abili y samples despi e complica ed (mos ly da a p o ec ion ela ed) egula ions o
accessing hose egis e s and e en in con ex s when in o ma ion is u gen ly needed.
2 Sampling echniques o mig an and e ugee popula ions
Re ugees a e o ced mig an s. This dis inguishes hem in essen ial aspec s om o he
mig an s, such as labo mig an s o mig an s due o amily euni ica ion. Volun a y
mig a ion usually happens a e a long decision-making p ocess (Kley 2017). People
lea ing hei home coun y o li e elsewhe e ep esen only a e y small p opo ion
o he home popula ion (wo ldwide, abou 3% in 2015; see Willekens e al. 2016).
Re ugees, howe e , lee in a hu y and do no ollow a (pu ely) a ional plan ega d-
ing hei escape ou e, place o e uge, o abou hei u he li e cou se (Hunkle
e al. 2022). Re ugees o en lee o neighbou ing o nea by coun ies, whe e hey
cons i u e a majo p opo ion o he e ugee popula ion. The eason is ha wa , ex-
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 81
ensi e pe secu ion, o displacemen makes mo e people ea o li e and limb, hus
d i ing hem o ac . Focusing on ecen examples, 30% o Sy ian ci izens and almos
20% o Venezuelan ci izens ha e le hei home coun y o lee iolen un es and
des uc ion (Uni ed Na ions High Commissione o Re ugees 2022a).
In he hos ing coun y, ne e heless, e ugees a e usually s ill a small g oup
compa ed o he na i e popula ion. Mo eo e , hey a e highly likely o change hei
esidence equen ly (Bloch 1999); al hough local legal esidence equi emen s o en
play a key ole in e ugees’ eedom o mo emen (see El-Kayed and Hamann (2018)
o he Ge man case). Hence, acco ding o he de ini ion o Tou angeau (2014),
hey a e a ha d- o- each popula ion (see also Massey 2014; Wenzel e al. 2022).
Tou angeau (2014) de ines a popula ion as being ha d o each i i is ei he ha d o
sample,ha d oiden i y, ha d o ind o con ac ,ha d ope suade,ha d oin e iew,
o a combina ion o hese aspec s. Speci ically, a g oup is ha d o sample when he e
is ei he no sampling ame a ailable o he g oup o he g oup is small wi h espec
o hei ac ion in he popula ion. Ano he eason o a popula ion o be ha d o
sample is mobili y. Highly mobile people canno be easily loca ed a a ce ain place
o esidence. A popula ion is ha d o iden i y when i is s igma ized, sensi i e, o
i sc eening ques ions miss membe s o he popula ion esul ing in unde -co e age.
G oups a e ha d o ind o con ac when hei membe s a e mobile, no willing o be
iden i ied as pa o a g oup, o simply p o ec ed by ga ekeepe s. Gi en con ac is
es ablished, some indi iduals a e ha d o pe suade o pa icipa e in he su ey. This
is o en ela ed o busyness, al e na i es o spend hei ime, he su ey opic, o he
au ho i y issuing he su ey. Finally, some indi iduals a e ha d o in e iew because
o language p oblems o because o hei cogni i e o physical abili ies.
In he li e a u e, se e al me hods a e p oposed o d aw samples o ha d- o- each
popula ions (see And eß and Ca eja (2018) o mig an popula ions). The mos
p ominen s a egies a e loca ion sampling, snowball sampling, esponden d i en
sampling, con enience sampling, and (sc eened) egis e samples. The i s wo ap-
p oaches p o ide non-p obabili y samples. The la e lead o p obabili y samples. In
loca ion sampling esponden s a e ec ui ed in places whe e hey spend a no able
amoun o ime. In snowball sampling a a ge pe son gi es access o he su ey
ques ionnai e o o he a ge pe sons he o she can con ac . Bo h app oaches a e
commonly used o ec ui mig an g oups, including e ugees (e.g., Agadjanian and
Zo o a 2012; McKenzie and Mis iaen 2009). Howe e , hey yield non-p obabili y
samples whose su ey s a is ics equi e a model-based amewo k o be ex apola ed
o he popula ion le el. Compa ed o he design-based app oach he me hodology o
es ima ion becomes mo e complex. Because in o ma ion on he a ge popula ion is
usually no a ailable, i is also no possible o compensa e o selec ion biases (see
G o es 2006 and Kal on 2014), e.g., ega ding g oups ha a e commonly no well
co e ed by mig an su eys such as uneduca ed mig an s (Amio 2020). Fu he -
mo e, esea ch shows ha he quali y o non-p obabili y samples is signi ican ly
wo se and less obus compa ed o andom-based samples (see Co nesse e al. 2020;
MacInnis e al. 2018). Responden d i en sampling (RDS) is applied in a ious mi-
g an s udies (e.g., La o 2018). The basic idea o RDS is o sample people om
a ha d- o- each popula ion and make use o hei social ne wo ks, hus elying on he
sampled pe son o be well connec ed o o he people o he popula ion. Compa ed
K
82 H. W. S einhaue e al.
o snowball sampling, RDS does allow o d awing a andom sample, when ce ain
assump ions a e me , e.g., he e e al chains become long enough. Howe e , he
condi ions o gene a ing a andom sample o mig an s a e di icul o achie e wi h
his echnique (especially passing on su ey ques ions h ough long chains o e-
sponden s); see also (Tyldum 2021). Swi ching RDS o a web mode does no eally
o e come i s obs acles. On he con a y, i makes i ulne able o misuse and poses
a h ea o da a quali y (Sosenko and B amley 2022). In con enience sampling, in e -
iewe s selec pa icipan s o a su ey, o example based on hei p oximi y (e.g.,
being a a ecep ion cen e ) o ce ain cha ac e is ics (e.g., being use o Facebook).
Bu also, esponden s can selec hemsel es in o he su ey, e.g., an online su ey
p omo ed on social media.
To o e come hese d awbacks, andom samples based on egis e s a e, no only
o mig an s o e ugees, seen as a supe io me hod o gene a ing a sample. This
equi es, i s , egis e da a and, secondly, access o egis e da a. Whe he such da a
exis s and is also accessible o scien i ic pu poses a ies ac oss coun ies. In gen-
e al, he da a si ua ion is be e in Scandina ian coun ies han in o he Eu opean
coun ies o a ound he wo ld (see Webe and Saa ela 2019;Belle al.2015). In
Ge many, he wo mos comp ehensi e egis e s o sampling mig an s and e ugees
a e he Ge man popula ion egis e and he cen al egis e o o eigne s. In gene al,
he popula ion egis e cons i u es he mos comp ehensi e sampling ame wi h he
legal obliga ion o egis e in he local egis a ion o ice wi hin wo weeks a e
changing add ess in Ge many. Howe e , a majo obs acle o sampling indi iduals
a he ede al le el is ha he Ge man popula ion egis e is no o ganized cen ally.
The egis e , main ained a he le el o he municipali y, con ains add esses and ba-
sic pe sonal demog aphic in o ma ion (e.g., gende , da e o bi h, na ionali y, da e
o mig a ion, see § 3 Bundesmeldegese z o he ull lis ) o almos all pe sons who
a e o icially esiding in Ge many; hus, also o all o icially egis e ed e ugees and
mig an s. The egis e can be used and accessed o scien i ic pu poses based on
§ 34 and § 46 Bundesmeldegese z. He e, he in o ma ion accessible is limi ed and
each in o ma ion mus be subs an ia ed. Each egis a ion o ice can decide whe he
o p o ide he desi ed in o ma ion. Recen examples o using he popula ion egis-
e o es ablishing mig an samples include he p ojec “Socio-cul u al in eg a ion
p ocesses among New Immig an s in Eu ope” (Diehl e al. 2016), he “Ge man
Emig a ion and Remig a ion Panel S udy” (E e e al. 2021), and he panel o he
Ge man Cen e o In eg a ion and Mig a ion Resea ch “DeZIM-Panel” (Dollmann
e al. 2022). Howe e , because o he decen alized s uc u e o Ge many’s pop-
ula ion egis e , i s use mus always ely on wo-s age sampling. In his sampling
echnique, a andom sample o municipali ies (se ing as p ima y sampling uni s,
PSUs) wi h enough mig an s is selec ed a he i s s age and indi idual add esses
( ep esen ing he seconda y sampling uni s, SSUs) a e d awn a he second s age.
The second egis e used o sampling mig an s and e ugees in Ge many is he
cen al egis e o o eigne s, which documen s all pe sons who a e no Ge man
na ionals and s ay in Ge many o mo e han 90 days (Babka on Gos omski and
Pupe e 2008). I ecei es i s in o ma ion mos ly om (local) immig a ion o ices in
Ge many (“Auslände behö den”), whose a ea o esponsibili y coincides wi h ha o
he Ge man municipali ies (“Gemeinde”) and dis ic s (“K eise”). Thus, wo-s age
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 83
samples can be selec ed om his egis e . Recen examples o using he AZR in-
clude he s udy “Fo ced Mig a ion and T ansna ional Family A angemen s” (Saue
e al. 2022) as well as he es ablishmen and e eshing o he IAB-BAMF-SOEP
s udy (B ücke e al. 2016; Kühne e al. 2019 and S einhaue e al. 2022). S udies
using he AZR as a sampling ame also ollow a wo-s age sampling echnique,
usually sampling local immig a ion o ices (PSUs) be o e sampling he indi iduals
egis e ed a hose o ices (SSUs). When d awing ou sample, howe e , add esses
we e only a ailable in he AZR o e ugees who we e unde going asylum p oceed-
ings. Because Uk ainian e ugees do no ha e o unde go an asylum p ocedu e, no
add esses we e a ailable o hem. Mo eo e , ob aining egis e ed add esses om
he immig a ion au ho i ies would ha e aken a compa a i ely long ime.
3 Re iew o exis ing su eys o Uk ainian e ugees
Cu en ly, mos o he ew exis ing su eys abou Uk ainian e ugees use non-p ob-
abili y sampling o c ea e hei sample—wi h all he d awbacks men ioned ea lie .
A b ie o e iew o s udies on Uk ainian e ugees is p esen ed in Table 1p o iding
de ails on he hos coun y, he ield pe iod he su ey was conduc ed, he numbe
o esponden s as well as he sampling design applied by he s udy. Wi h espec o
compa a i e samples o Uk ainian e ugees ac oss di e en hos ing coun ies, he
Uni ed Na ions High Commissione o Re ugees (UNHCR) and pa ne s collec ed
da a in Czech Republic, Hunga y, Moldo a, Poland, Romania, and Slo akia om
4871 Uk ainian e ugees using a loca ion sampling app oach be ween 16 May and
15 June 2022. He e, in e iews we e mos ly conduc ed a loca ions such as bo -
de a eas and ansi zones o in o ma ion and assis ance poin s (Uni ed Na ions
High Commissione o Re ugees 2022b). Simila ly, he Eu opean Union Agency
Table 1 B ie o e iew on exis ing s udies on Uk ainian Re ugees
S udy/Conduc o s Coun y Field pe iod Responden s Sampling design
UNHCR & pa -
ne s
Czech Republic, Hun-
ga y, Moldo a, Poland,
Romania, and Slo akia
16 May
o 15 June
2022
4871 Loca ion sampling
EUAA & OECD Eu opean Union 11 Ap il o
7 June 2022
2369 Con enience
sampling
BMI Ge many Ma ch 2022 App ox.
2000
Loca ion sam-
pling, con enience
sampling
GESIS Ge many, Poland Ap il o
May 2022
App ox.
1300
Con enience
sampling
Uk AiA Aus ia Ma ch o
June 2022
App ox.
1000
Con enience
sampling
Uk AiA K aków, Poland Ma ch o
June 2022
500 Loca ion sampling
Uk ainian
Re ugees in Poland
Su ey 2022
Poland June o
Augus
2022
App ox.
1800
S a i ied wo-
s age andom
sampling
K
84 H. W. S einhaue e al.
o Asylum (EUAA) oge he wi h he O ganiza ion o Economic Co-ope a ion and
De elopmen (OECD) collec ed in o ma ion on 2369 Uk ainian e ugees ia con e-
nience sampling using an online mode o da a collec ion du ing he ime be ween
11 Ap il and 7 June 2022 (Eu opean Union Agency o Asylum 2022).
In Ge many, he Fede al Minis y o he In e io and Communi y (BMI) launched
an ea ly su ey in Ma ch 2022 a he egis a ion o ices o h ee cen al hubs in
Be lin, Hambu g, and Munich, whe e hey in e iewed e ugees. Addi ionally, he
su ey was ad e ised on he homepages o he BMI, BAMF, and by he mobile
phone app Ge many4Uk aine.de (Fede al Minis y o he In e io and Communi y
2022). This mul isou ce con enience sampling app oach esul ed in almos 2000
in e iews. Addi ionally, a web-based su ey on Uk ainians s aying in Ge many
o Poland was un by he Leibniz Ins i u e o he Social Sciences (GESIS) in
Ap il and May 2022 (Pö zschke e al. 2022). The s udy ec ui ed a ound 1300
e ugees h ough ad e s on social media. Bo h s udies p o ide an ini ial pic u e o
he a es, li ing si ua ions, a i udes, p oblems, and needs o he Uk ainian e ugees
in Ge many. Howe e , as bo h s udies do no use andom samples, nei he can make
gene al s a emen s abou he si ua ion o Uk ainian e ugees in Ge many. Fu he ,
bo h s udies consis o only one wa e. Such designs canno p o ide any insigh s
conce ning he si ua ion o Uk ainian e ugees and how i is changing o e ime.
Wi h espec o o he majo hos coun ies o Uk ainian e ugees, Aus ia launched
a apid esponse su ey (Uk ainian a i als in Aus ia, Uk AiA) using con enience
sampling o quickly lea n abou he socio-demog aphics o Uk ainian e ugees as
well as hei educa ional esou ces and in en ions o s ay in Aus ia o e u n. The
su ey was conduc ed be ween Ma ch and June 2022 using bo h pen and pape in e -
iews (PAPI) and compu e assis ed web in e iews (CAWI). The su ey conduc ed
mo e han 1000 in e iews wi h adul esponden s, also collec ing in o ma ion on
hei pa ne s and child en (Kohlenbe ge e al. 2022). Pa allel o he s udy con-
duc ed in Aus ia, 500 Uk ainian e ugees we e su eyed in K aków, Poland, also
using loca ion sampling a di e en egis a ion spo s (P˛edziwia e al. 2022). Mo e-
o e , Poland oge he wi h he Wo ld Heal h O ganiza ion (WHO) implemen ed he
Uk ainian Re ugees in Poland Su ey 2022. They use a s a i ied wo-s age andom
sampling design o sample Uk ainian e ugees. A he i s s age, loca ions (i.e.,
PSUs) we e s a i ied by bo de c ossing egions and PSUs we e andomly sampled.
Pe sons (i.e., SSUs) aged 18 yea s and olde we e sampled using sys ema ic sam-
pling wi hin he PSU, i hey had al eady s ayed in Poland o a leas wo weeks.
Roughly 1800 sampled pe sons also p o ided in o ma ion hi d pe sons hey we e
a eling wi h, hus yielding in o ma ion on abou 5000 Uk ainian e ugees (Beqi i
and Cie piał-Wolan 2022).
4 Regis a ion p ocedu e o Uk ainian e ugees in Ge many
Knowing ha comp ehensi e egis e s a e he bes choice o c ea ing a andom
sample o Uk ainian e ugees is one side o he coin, bu inding and accessing such
a egis e wi hin a easonable amoun o ime and c ea ing an app op ia e sample
in a imely manne is he o he . A undamen al issue he e is how o icial au ho i ies
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 91
Table 3 Final disposi ion codes
o he g oss sample acco ding
o AAPOR. (Sou ce: IAB-BiB/
FReDA-BAMF-SOEP-Da a;
Au ho s’ own calcula ion)
Final disposi ion code Numbe
1.1 Comple e in e iew 10,395
2.11 Re usal 63
2.12 B eak-o o pa ial 2008
2.2 Non-Con ac 24,992
2.31 Dea h 7
2.32 Physically o men ally unable/
incompe en
26
2.4 New add ess a e ield pe iod 259
3.1 No hing known abou add ess 10,076
4.1 Sc een ou 97
4.2 Mo ed ab oad 77
in he in e im. Fu he de ails on inal disposi ion codes a e a ailable in Table 3.
A o al o 8695 in e iews we e submi ed online, wi h an addi ional 1700 submi ed
as PAPI.
The ollowing Table 4p o ides he numbe o e ugees in he g oss sample as
well as o he i s wa e in 2022 by ede al s a e, sex, age g oup, and ma i al s a us.
The in o ma ion displayed o he g oss sample and he i s wa e in 2022 is he
in o ma ion as p o ided by he EMR. Ce ainly, he da a highligh s a no able con-
cen a ion o e ugees in No h Rhine-Wes phalia and Ba a ia, pa icula ly a ibu ed
o signi ican popula ions in Düsseldo and Munich, espec i ely. These u ban cen-
e s, being majo des ina ions o Uk ainian e ugees, con ibu e subs an ially o he
o e all numbe s in hei espec i e ede al s a es. Fo a huge po ion o he sample
he ma i al s a us is unknown. This is mainly because EMRs did no p o ide his
in o ma ion. Among known s a uses, he e is di e si y, wi h a majo i y o he sample
being single o ma ied.
Fo he g oss sample, in o ma ion p o ided by he EMR (see Table 4)is he
only in o ma ion a ailable o compa e e ugees who decided o pa icipa e in he
panel and hose who e used o do so. We model he decision o pa icipa e in he
panel using a logi model. The dependen a iable (y) o he model is he decision
o pa icipa e in he panel (y= 1 i inal disposi ion code is 1.1 and y= 0 i inal
disposi ion code is one o 2.11, 2.12, 2.2, 2.32, 2.4, and 3.1). We exclude e ugees
who died (code 2.31), who we e sc eened ou o he popula ion (code 4.1), and
who mo ed ab oad (code 4.2) om he analysis because hey do no belong o he
desi ed popula ion. In he model we es ima e, we con ol o ede al s a e, age, sex,
and ma i al s a us by inse ing dummy- a iables. Figu e 5displays he coe icien
plo o he model including he dummy a iables which signi ican ly in luence
he decision o pa icipa e in he panel s udy. The model inds male e ugees and
e ugees sampled in Be lin and Hambu g o be less likely o pa icipa e in he panel.
Re ugees being sampled in Saa land and e ugees being ma ied a e mo e likely o
pa icipa e in he panel s udy.
The weigh s accompanying he da a a e gene a ed h ough a h ee-s ep p o-
cess. Ini ially, design weigh s o he wo-phase design a e calcula ed ollowing
he me hodology ou lined by Sä ndal e al. (2003, Chap. 9). The design weigh s a e
K
92 H. W. S einhaue e al.
Table 4 To al numbe o e ugees in he g oss sample and he i s wa e o he IAB-BiB/FReDA-BAMF-
SOEP-Su ey. (Sou ce: IAB-BiB/FReDA-BAMF-SOEP-Da a; Au ho s’ own calcula ion)
G oss sample Re usals Panel
Fede al s a e
Schleswig Hols ein 971 767 204
Hambu g 3397 2743 654
Lowe Saxony 2625 2054 571
B emen 341 267 74
No h Rhine-Wes phalia 9801 7560 2241
Hesse 3404 2692 712
Rhineland-Pala ina e 2036 1559 477
Baden-Wü embe g 3473 2669 804
Ba a ia 8584 6613 1971
Saa land 289 195 94
Be lin 7421 6059 1362
B andenbu g 642 517 125
Mecklenbu g-Wes e n Pome ania 734 559 175
Saxony 1785 1434 351
Saxony-Anhal 1354 1046 308
Thü ingen 1143 871 272
Sex
Unknown 10 9 1
Male 10,086 8110 1976
Female 37,904 29,486 8418
Age g oup
Unknown 1604 1233 371
Olde han 17 up o 20 2459 1875 584
Olde han 20 up o 25 4025 3127 898
Olde han 25 up o 30 4586 3681 905
Olde han 30 up o 35 6318 4930 1388
Olde han 35 up o 40 7387 5758 1629
Olde han 40 up o 45 5886 4588 1298
Olde han 45 up o 50 4203 3251 952
Olde han 50 up o 55 3196 2519 677
Olde han 55 up o 60 2681 2137 544
Olde han 60 up o 65 3318 2656 662
Olde han 65 up o 70 2337 1850 487
Olde han 70 0 0 0
Ma i al s a us
Unknown 28,323 22,346 5977
Single 8445 6609 1836
Ci il pa ne ship 6 4 2
Ma ied 8548 6515 2033
Di o ced 1949 1534 415
Widowed 729 597 132
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 93
Fig. 5 Coe icien Plo o he model es ima ing he decision o pa icipa e in he panel o he IAB-BiB/
FReDA-BAMF-SOEP-Su ey. (Sou ce: IAB-BiB/FReDA-BAMF-SOEP-Da a; Au ho s’ own calcula ion)
in ended o accoun o he complexi ies in oduced by he wo-phase sampling as
well as he sys ema ic (pps) selec ion. Design weigh s a e hen adjus ed o accoun
o non- esponse. This adjus men conside s cha ac e is ics p o ided by he EMR
( e e enced in Table 4). This s ep add esses po en ial biases in oduced by non- e-
sponse. Subsequen ly, aking is applied o ma gins o sex, age, ede al s a e, and
mon h o immig a ion, o ensu e ha he sample dis ibu ion aligns wi h he popu-
la ion dis ibu ion, as p o ided by he AZR e ec i e No embe 2022. The esul ing
weigh s, compu ed h ough his p ocess, yield an e ec i e sample size o 8231 and
a easonable design e ec o 1.26, compu ed acco ding o Kish (1992).
7 Conclusions
A e Russia’s in asion o Uk aine in ea ly 2022, mo e han one million e ugees a -
i ed in Ge many. These Uk ainian e ugees di e in many aspec s om Ge many’s
pas o ced mig a ion expe iences including hei demog aphics, legal s a us, pe -
cei ed accep ance, alloca ion, and he con inui y o ci cula mig a ion lows, no
leas because o g ea e geog aphical p oximi y. To lea n abou hese ecen ly a i -
ing e ugees, hei needs, esou ces, as well as challenges ahead, a su ey allowing
o gene aliza ion o his popula ion was u gen ly needed. To mee his need quickly,
ou ins i u ions joined hei expe ise and esou ces o c ea e a p obabili y sample
o he popula ion o Uk ainian e ugees using wo di e en egis e s: he Ge man
K
94 H. W. S einhaue e al.
popula ion egis e and he cen al egis e o o eigne s. The app oach we p esen ed
in his pape can be used o d aw a sample in he same way o any hi d coun y
na ionals.
The sampling design encompasses a wo-phase me hodology in ol ing sys ema ic
(pps) sampling. This app oach may be deemed in ica e due o ce ain d awbacks
associa ed wi h sys ema ic (pps) sampling, no ably he inabili y o compu e second-
o de inclusion p obabili ies, hus impeding classical a iance es ima ion (Wol e
2007). None heless, al e na i e me hodologies, such as jackkni e o boo s ap me h-
ods, eme ge as iable op ions o a iance es ima ion wi hin he con ex o complex
designs.
While u ilizing he AZR and he EMR in a wo-phase sampling app oach, ou
s udy is con ined o he popula ion o Uk ainian e ugees aged 18–70, egis e ed
be ween 24 Feb ua y 2022 and he ime o sampling in he EMR. A compa ison
be ween he g oss sample d awn om he EMR and he co esponding popula ion
egis e ed in he AZR (e ec i e No embe 2022) e eals a la ge cong uence. The
p esen ed analyses show ha , e en wi hin a geopoli ical c isis esul ing in la ge
in lows o e ugees, he exis ing egis e s in Ge many cons i u e a comp ehensi e
sampling ame. Following ime and esou ce cons ain s, a delibe a e decision was
aken o concen a e on mo e popula ed a eas ins ead o u al a eas, aiming owa ds
a la ge g oss sample.
Compa ing he g oss and he ne sample o Uk ainian e ugees wi h he a ge
popula ion o Uk ainian e ugees aged 18–70 ha we e egis e ed by he end o May
2022 in Ge many p o ides e idence o he o e all success o his sampling app oach
and he high-quali y o his p obabili y sample. The pape shows he bene i s and
easibili y o es ablishing egis e -based samples e en in con ex s o geopoli ical
c isis and p o iding sound da a wi hin b ie ime ho izons. Fo poli ics and p ac i-
ione s, he da a p o ides comp ehensi e in o ma ion o e idence-based decision-
making much ea lie han in he pas . Fo mig a ion schola s he esul ing su ey
da a is highly aluable because his su eying o he a ge popula ion al eady s a s
be o e selec i e e u n and onwa d mo emen s by he o ced mig an s ha e aken
place. Mo eo e , he da a can be used in u u e esea ch and compa e esul s o ou
s udy wi h hose o o he s o compa able measu emen s. This will be o pa icu-
la in e es o esea che s compa ing indings om p obabili y and non-p obabili y
samples.
Acknowledgemen s The au ho s hank all o he membe s o he IAB-BiB/FReDA-BAMF-SOEP p ojec
eam (in alphabe ical o de ): He be B ücke , Ma in Buja d, Ad iana Ca dozo, Ma kus M. G abka, Yuliya
Kosyako a, Am ei Maddox, Nadja Milewski, Robe Nade i, Wenke Niehues, Nina Ro he , Leno e Saue ,
Sophia Schmi z, Sil ia Schwanhäuse , C. Ka ha ina Spieß, and Ke s in Tanis as well as he membe s o
he p ojec eam a in as consis ing o Do is Hess, Michael Ruland, and Sab ina To eg oza. Addi ional
hanks o Jan Ebe le om he Fede al S a is ical O ice and many colleagues om local egis a ion o ices
p o iding in aluable insigh s in o he de ails as well as hei local speci ics o he egis a ion p ocedu es in
Ge many. Finally, he au ho s hank wo anonymous e iewe s o hei aluable commen s and sugges ions
imp o ing he pape conside ably.
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Con lic o in e es H.W. S einhaue , J.P. Décieux, M. Siege , A. E e and S. Zinn decla e ha hey ha e
no compe ing in e es s.
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 95
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as
you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Com-
mons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle
a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he
ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly
om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.
0/.
Re e ences
Adam F, Föbke S, Imani D, P a enbach C, Weiss G, Wiegand C-C (2021) “Los in ansi ion”? In eg a-
ion o e ugees in o he local housing ma ke in Ge many. J U ban A 43(6):831–850. h ps://doi.
o g/10.1080/07352166.2018.1562302
Agadjanian V, Zo o a N (2012) Sampling and su eying ha d- o- each popula ions o demog aphic e-
sea ch: a s udy o emale labo mig an s in moscow, ussia. Demog Res 26(5):131–150. h ps://doi.
o g/10.4054/DemRes.2012.26.5
Amio M (2020) Immig a ion, local c owd-ou and unde co e age bias. (CEP discussion pape 1669).
London, UK: Cen e o Economic Pe o mance, LSE. h ps://cep.lse.ac.uk/pubs/download/dp1669.
pd . Accessed 1 Dec 2022
And eß H-J, Ca eja R (2018) Sampling mig an s in six Eu opean coun ies: how o de elop a compa a i e
design? Comp Mig S ud. h ps://doi.o g/10.1186/s40878-018-0099-x
Babka on Gos omski C, Pupe e M (2008) Zu allsbe agung on Auslände n au Basis des Auslän-
de zen al egis e s: E ah ungen bei de Rep äsen a i be agung „Ausgewähl e Mig an eng uppen in
Deu schland 2006/2007“. Me hoden Da en Anal 2(2):149–177
Bell M, Cha les-Edwa ds E, Kupiszewska D, Kupiszewski M, S illwell J, Zhu Y (2015) In e nal mig a ion
da a a ound he wo ld: Assessing con empo a y p ac ice. Popul Space Place 21(1):1–17. h ps://doi.
o g/10.1002/psp.1848
Beqi i M, Cie piał-Wolan M (2022) Uk ainian e ugees in Poland su ey 2022—P elimina y indings.
Poland. h ps://da a.unhc .o g/en/documen s/download/97052. Accessed 2 Dec 2022
Bloch A (1999) Ca ying ou a su ey o e ugees: some me hodological conside a ions and guidelines.
J Re ug S ud 12(4):367–383. h ps://doi.o g/10.1093/j s/12.4.367
Bogumil J, Bu gi M, Kuhlmann S, Ha ne J, Heube ge M, K önke C (2018) Besse e Ve wal ung in de
Mig a ions- und In eg a ionspoli ik: Handlungsemp ehlungen ü Ve wal ungen und Gese zgebung
im öde alen Sys em. Nomos, Baden-Baden h ps://doi.o g/10.5771/9783845295862
B ücke H, Ro he N, Schupp J (2016) IAB-BAMF-SOEP-Be agung on Ge lüch e en: Übe blick und e -
s e E gebnisse. Poli ikbe a ung kompak 116. Be lin. h ps://www.diw.de/documen s/publika ionen/
73/diw_01.c.547162.de/diwkompak _2016-116.pd . Accessed 19 Sep 2022
B ücke H, E e A, G abka M, Kosyako a Y, Niehues W, Ro he N, Tanis K (2023) Uk ainian e ugees
in Ge many: e idence om a la ge ep esen a i e su ey. Comp Popul S ud 48:395–424. h ps://doi.
o g/10.12765/CPoS-2023-16
Co nesse C, Blom A, Du win D, K osnick J, De Leeuw E, Legleye S, Wenz A (2020) A e iew o concep-
ual app oaches and empi ical e idence on p obabili y and nonp obabili y sample su ey esea ch.
J Su S a Me hodol 8(1):4–36. h ps://doi.o g/10.1093/jssam/smz041
Diehl C, Lubbe s M, Mühlau P, Pla L (2016) S a ing ou : New mig an s’ socio-cul u al in eg a-
ion ajec o ies in ou Eu opean des ina ions. E hnici ies 16(2):157–179. h ps://doi.o g/10.1177/
1468796815616158
Dollmann J, Maye S, Lie z A, Siegel M, Köhle J (2022) Se ing up an o line ec ui ed online access
panel wi h an o e sampling o immig an s and hei descendan s: he Ge man deZIM.panel h ps://
doi.o g/10.31235/os .io/mdpnx
D ažano á L, Geddes A (2022) A i udes owa ds Uk ainian e ugees and go e nmen al esponses
in 8 Eu opean coun ies. h ps://www.asilep ojec .eu/a i udes- owa ds-uk ainian- e ugees-and-
go e nmen al- esponses-in-8-eu opean-coun ies/ (C ea ed Sep embe ). Accessed 3 Jan 2023
El-Kayed N, Hamann U (2018) Re ugees’ access o housing and esidency in Ge man ci ies: in e nal bo de
egimes and hei local a ia ions. Soc Incl 6(1):135–146. h ps://doi.o g/10.17645/si. 6i1.1334
K
96 H. W. S einhaue e al.
E e A, Décieux J, E linghagen M, Audi o J, Sande N, Schneide N, Wi e N (2021) Su eying ac oss
bo de s: he expe iences o he Ge man emig a ion and emig a ion panel s udy. In: E linghagen M,
E e A, Schneide N, Wi e N (eds) The global li es o Ge man mig an s. Sp inge , Cham, pp 21–39
h ps://doi.o g/10.1007/978-3-030-67498-4_2
Eu opean Union Agency o Asylum (2022) Su eys o a i ing mig an s om Uk aine. h ps://euaa.
eu opa.eu/si es/de aul / iles/publica ions/2022-06/2022_06_14_EUAA_SAM_UKR_Fac shee .pd .
Accessed 13 Oc 2022
Fede al Minis y o he In e io and Communi y (2022) Ge lüch e e aus de Uk aine. h ps://www.
bmi.bund.de/Sha edDocs/downloads/DE/ e oe en lichungen/nach ich en/2022/um age-uk aine-
luech linge.pd ?__blob=publica ionFile& =3. Accessed 13 Oc 2022
Fede al O ice o Mig a ion and Re ugees (2022) Ques ions and answe s on en e ing Uk aine and s aying
in Ge many. h ps://www.bam .de/Sha edDocs/Anlagen/DE/AsylFluech lingsschu z/ aq-uk aine-en.
pd ?__blob=publica ionFile& =16. Accessed 3 Jan 2023
G o es R (2006) Non esponse a es and non esponse bias in household su eys. Public Opin Q 70(5):
646–675. h ps://doi.o g/10.1093/poq/n l033
Hunkle C, Scha e T, Sue baum M, Yanasmayan Z (2022) Spa ial and social im/mobili y in o ced mi-
g a ion: e isi ing class. J E hn Mig S ud. h ps://doi.o g/10.1080/1369183X.2022.2123431
Jacobsen J, Siege M (2023) Es ablishing a panel s udy o e ugees in Ge many: i s wa e e-
sponse and panel a i ion om a compa a i e pe spec i e. Field Me hods. h ps://doi.o g/10.1177/
1525822X231204817
Kal on G (2014) P obabili y sampling me hods o ha d- o-sample popula ions. In: Tou angeau R, Ed-
wa ds B, Johnson T, Wol e K, Ba es N (eds) Ha d- o-su ey popula ions. Camb idge Uni e si y
P ess, Camb idge, pp 401–423 h ps://doi.o g/10.1017/CBO9781139381635.024
Kish L (1992) Weigh ing o unequal Pi. J O S a 8(2):183–200
Kley S (2017) Facili a o s and cons ain s a each s age o he mig a ion decision p ocess. Popul S ud
71(Sup1):35–49. h ps://doi.o g/10.1080/00324728.2017.1359328
Kohlenbe ge J, Bube -Ennse I, Rengs B, Se z I, Riede e B (2022) Uk AiA Abschlussbe ich S ad Wien.
Aus ia. h ps://www.uk aia.a /wp-con en /uploads/2022/08/uk aia_ inal_ epo _ci y_o _ ienna.pd .
Accessed 2 Dec 2022
Kühne S, Jacobsen J, K oh M (2019) Sampling in imes o high immig a ion: he su ey p ocess o he
IAB-BAMF-SOEP su ey o e ugees. Su ey me hods: insigh s om he ield. h ps://doi.o g/10.
13094/SMIF-2019-00005
La o S (2018) Collec ing da a om mig an s in Ghana: Lessons lea ned using esponden -d i en sam-
pling. Demog Res 38(36):1017–1058. h ps://doi.o g/10.4054/DemRes.2018.38.36
MacInnis B, K osnick J, Ho A, Cho M-J (2018) The accu acy o measu emen s wi h p obabili y and
Nonp obabili y su ey samples: eplica ion and ex ension. Public Opin Q 82(4):707–744. h ps://doi.
o g/10.1093/poq/n y038
Massey D (2014) Challenges o su eying immig an s. In: Tou angeau R, Edwa ds B, Johnson T, Wol e K,
Ba es N (eds) Ha d- o-su ey popula ions. Camb idge Uni e si y P ess, Camb idge, pp 270–292
h ps://doi.o g/10.1017/CBO9781139381635.017
McKenzie D, Mis iaen J (2009) Su eying mig an households: a compa ison o census-based, snowball
and in e cep poin su eys. J R S a Soc Se A 172(2):339–360
P˛edziwia K, B zozowski J, Naho niuk O (2022) Re ugees om Uk aine in K aków. h ps://owim.uek.
k akow.pl/wp-con en /uploads/use - iles/ epo s/OWIM e ugees.pd . Accessed 2 Dec 2022
Pö zschke S, Weiß B, Hebel A, Piepenbu g J, Poppek O (2022) Ge lüch e e aus de Uk aine – E s e desk ip-
i e E gebnisse eine Onlinebe agung in Deu schland und Polen h ps://doi.o g/10.34879/gesisblog.
2022.60
Sä ndal C-E, Swensson B, W e man J (2003) Model assis ed su ey sampling. Sp inge , New Yo k
Saue L, K aus E, Kassam K, Schüh e S, Pupe e M, Wol e S, Schneeklo h U (2022) Fo ced
mig a ion and ansna ional amily a angemen s—E i ean and Sy ian e ugees in Ge many
(T ansFAR): me hodology epo . BiB da a and echnical epo s 1/2022, Wiesbaden. h ps://d-
nb.in o/1256734772/34. Accessed 12 Jan 2023
Sosenko F, B amley G (2022) Sma phone-based Responden D i en Sampling (RDS): A me hodological
ad ance in su eying small o ‘ha d- o- each’ popula ions. PLoS One 17(7):e270673. h ps://doi.o g/
10.1371/jou nal.pone.0270673
S einhaue H, Zinn S, Will G (2019) Sampling e ugees o an educa ional longi udinal su ey. Su ey
me hods: insigh s om he ield. h ps://doi.o g/10.13094/SMIF-2019-00007
K
Es ablishing a p obabili y sample in a c isis con ex : he example o Uk ainian e ugees in... 97
S einhaue H, Siege s R, Siege M, Jacobsen J, Zinn S (2022) Sampling, non esponse, and weigh ing
o he 2020 e eshmen sample (M6) o he IAB-BAMF-SOEP e ugee panel. SOEP su ey pape
1104. Be lin. h ps://www.econs o .eu/handle/10419/261440. Accessed 1 Dec 2022
The Ame ican Associa ion o Public Opinion Resea ch (2016) Final disposi ions o case codes and ou -
come a es o su eys
Tillé Y (2006) Sampling algo i hms. Sp inge , New Yo k
Tou angeau R (2014) De ining ha d- o-su ey popula ion. In: Tou angeau R, Edwa ds B, Johnson T,
Wol e K, Ba es N (eds) Ha d- o-su ey popula ions. Camb idge Uni e si y P ess, Camb idge,
pp 2–20 h ps://doi.o g/10.1017/CBO9781139381635.003
Tyldum G (2021) Su eying mig an popula ions wi h esponden -d i en sampling. Expe iences om su -
eys o eas -wes mig a ion in Eu ope. In J Soc Res Me hodol 24(3):341–353. h ps://doi.o g/10.10
80/13645579.2020.1786239
Uni ed Na ions High Commissione o Re ugees (2022a) Global ends—Fo ced displacemen in 2021.
Copenhagen. h ps://www.unhc .o g/media/global- ends- epo -2021. Accessed 2 Jan 2023
Uni ed Na ions High Commissione o Re ugees (2022b) Li es on hold: p o iles and in en ions o e ugees
om Uk aine. h ps://da a.unhc .o g/en/documen s/download/94176. Accessed 13 Oc 2022
Webe R, Saa ela J (2019) Ci cula mig a ion in a con ex o ee mobili y: E idence om linked popula-
ion egis e da a om Finland and Sweden. Popul Space Place 25(4):e2230. h ps://doi.o g/10.1002/
psp.2230
Wenzel L, Husen O, Sande mann P (2022) Su eying di e se Subpopula ions in e ugee s udies: e lec ions
on sampling, implemen a ion, and ansla ion s a egies d awn om expe iences wi h a egional quan-
i a i e su ey on e ugee pa en s in Ge many. J Re ug S udie. h ps://doi.o g/10.1093/j s/ eac043
Willekens F, Massey D, Rayme J, Beauchemin C (2016) In e na ional mig a ion unde he mic oscope.
Science 352(6288):897–899. h ps://doi.o g/10.1126/science.aa 6545
Wol e KM (2007) In oduc ion o a iance es ima ion. Sp inge , New Yo k
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.
K
