Ando , Ma k A.; Baue , Thomas K.; Eße , Jana; Schmid , Ch is oph M.; Tombe g,
Lukas
A icle — Published Ve sion
Who Ge s Vaccina ed? Cogni i e and Non‐Cogni i e
P edic o s o Indi idual Beha iou in Pandemics
Ox o d Bulle in o Economics and S a is ics
P o ided in Coope a ion wi h:
John Wiley & Sons
Sugges ed Ci a ion: Ando , Ma k A.; Baue , Thomas K.; Eße , Jana; Schmid , Ch is oph M.; Tombe g,
Lukas (2024) : Who Ge s Vaccina ed? Cogni i e and Non‐Cogni i e P edic o s o Indi idual
Beha iou in Pandemics, Ox o d Bulle in o Economics and S a is ics, ISSN 1468-0084, Wiley,
Hoboken, NJ, Vol. 87, Iss. 3, pp. 562-585,
h ps://doi.o g/10.1111/obes.12644
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323874
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/
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 87, 3 (2025) 0305-9049
doi: 10.1111/obes.12644
Who Ge s Vaccina ed? Cogni i e and Non-Cogni i e
P edic o s o Indi idual Beha iou in Pandemics
MARK A. ANDOR,†,‡ THOMAS K. BAUER,†,‡,§ JANA EßER,†
CHRISTOPH M. SCHMIDT†,‡,§ and LUKAS TOMBERG†
†RWI - Leibniz Ins i u e o Economic Resea ch, Hohenzolle ns aße 1-3 Essen, 45128, Ge many
‡Ruh Uni e si y Bochum, Uni e si ¨
a ss aße 150 Bochum, 44801, Ge many
§IZA - Ins i u e o Labo Economics, Schaumbu g-Lippe-S . 5-9 Bonn, 53113, Ge many
Abs ac
This s udy in es iga es di e en cogni i e and non-cogni i e cha ac e is ics associa ed
wi h indi iduals’ willingness o ge accina ed agains Co id-19 and hei ac ual
accina ion s a us. Ou empi ical analysis is based on da a ob ained om h ee su ey
wa es conduc ed in 2021 among abou 2,000 indi iduals li ing in he Ge man s a e o
No h Rhine-Wes phalia. We ind ha indi iduals wi h a high le el o ai eac ance – a
pe sonali y cha ac e is ic ha en ails he pe sonal endency o pe cei e pe suasion a emp s
as es ic ing one’s eedom – display a signi ican ly lowe willingness o ge accina ed.
They also end o ge inocula ed la e o ne e . Mo eo e , neu o icism, locus o con ol,
and s a is ical nume acy appea o be associa ed wi h he willingness o ge accina ed,
bu hese esul s a e less p onounced and less obus . Ou esul s indica e ha accina ion
campaigns and policies could be imp o ed by speci ically add essing hose wi h a high
le el o ai eac ance.
I. In oduc ion
S a ing a he end o 2019, he Co id-19 pandemic sp ead ou o he en i e wo ld quickly
and o ced many go e nmen s o implemen public heal h measu es. To decele a e
in ec ion and hospi aliza ion dynamics, au ho i ies manda ed, o example, he use o ace
masks and compliance wi h egula ions o keep dis ance. As accines became a ailable
in 2020, pa icula hope was placed on accina ion campaigns, since inocula ion is shown
o p e en in ec ions wi h Co id-19 om u ning in o se e e cases and may e en p e en
in ec ions al oge he (CDC, 2022). The e ec i eness o public heal h measu es, including
JEL Classi ica ion numbe s: D01, I12, I18.
We g a e ully acknowledge inancial suppo by he Minis e ium ¨
u Wi scha , Inno a ion, Digi alisie ung und
Ene gie des Landes No d hein-Wes alen. We hank he Edi o Climen Quin ana-Domeque, he anonymous
e e ees, Manuel F ondel, and Colin Vance o help ul commen s and sugges ions. Fu he mo e, we hank
ou p ojec collabo a o s Philipp B eidenbach, Maximilian Di ks, Ka ja Fels, Nils Ch is ian H¨
onow, Ma hias
Kaeding, Delia Niehues, S e an Rump , and To s en Schmid . Be o e ob aining ou esea ch da a, we p e egis e ed
ou hypo heses a he Ame ican Economic Associa ion’s RCT egis y (AEARCTR0007710).
562
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s use, dis ibu ion and
ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed.
P edic o s o indi idual beha iou in pandemics 563
accina ion campaigns, depends c i ically on popula accep ance o and adhe ence o hese
measu es. Howe e , in many coun ies, he e has been eluc ance o e usal o ollow
he heal h guidelines and e en s ong p o es s agains some o hese measu es, especially
agains accina ion equi emen s.
The sea ch o explana ions o his eluc ance and e usal has led o a la ge body
o esea ch since he ou b eak o he pandemic. Beha iou al esea ch has iden i ied
impo an cul u al and g oup-le el cha ac e is ics including a lack o us (Ba gain and
Aminjono , 2020), poli ical pola iza ion (Allco e al.,2020) and cul u al indi idualism
(Bazzi, Fiszbein, and Geb esilasse, 2021; Chen, F ey, and P esiden e, 2021) as impo an
de e minan s o eluc ance o ollow heal h guidelines. On he indi idual le el, Al a o
e al. (2022) and Fang e al. (2022) iden i ied p osociali y and social p e e ences as ac o s
os e ing compliance.
Wi h his pape , we complemen hese lines o esea ch by ocusing on he ole o
cogni i e and non-cogni i e cha ac e is ics. Cogni i e cha ac e is ics such as s a is ical
nume acy can be expec ed o be ele an o accina ion decisions because hey may
con ibu e o an indi idual’s abili y o ca ego ize he dispa a e s a is ical in o ma ion
eleased du ing he pandemic and o make an in o med decision based on ha
in o ma ion. In addi ion, non-cogni i e cha ac e is ics, such as pe sonali y ai s, may
play an impo an ole in accina ion decisions, o example, by mode a ing ecipien s’
eac ions o pe suasi e messages (Hi sh, Kang, and Bodenhausen, 2012), by shaping an
indi idual’s p oneness o expe ience ce ain emo ions (Widige and Ol manns, 2017), o
by de e mining an indi idual’s endency o beha e p osocially (Ando e al.,2022).
Speci ically, we examine he associa ion o ai eac ance, neu o icism, locus o con ol
belie s, as well as s a is ical nume acy wi h indi iduals’ accina ion s a us and willingness
o ge accina ed agains Co id-19.1Ou analysis is based on h ee wa es o su ey da a
o mo e han 2,000 indi iduals li ing in he Ge man s a e o No h Rhine-Wes phalia
collec ed be ween May and No embe 2021. As ou main ou come, we obse e he sel -
epo ed accina ion s a us and, i he pe son is no ye accina ed, he s a ed willingness
o ge accina ed. Fo b e i y, we e e o ou main ca ego ical ou come a iable he ea e
as he willingness o ge accina ed, which includes he e ealed p e e ence o al eady
being accina ed as one ca ego y.
Ou esul s indica e ha he willingness o ge accina ed agains Co id-19 dec eases
signi ican ly wi h an inc easing le el o ai eac ance, a pe sonali y cha ac e is ic ha
en ails he pe sonal endency o pe cei e pe suasion a emp s as es ic ing one’s eedom.
Neu o icism, locus o con ol and s a is ical nume acy also appea o be associa ed wi h
he willingness o ge accina ed, bu hese esul s a e less p onounced and less obus
wi h ega d o he econome ic speci ica ion han he esul s ega ding ai eac ance. The
es ima ion o an Accele a ed Failu e Time (AFT) model u he shows ha indi iduals
wi h a high le el o ai eac ance ge accina ed signi ican ly la e han hose wi h a low
o medium le el o ai eac ance, while he emaining cha ac e is ics do no seem o be
associa ed wi h he iming o accina ion.
1Be o e ob aining ou esea ch da a, we p e egis e ed ou hypo hesis ha hese cha ac e is ics may be associa ed
wi h indi idual compliance wi h p o ec i e measu es agains Co id-19 (Ando , 2021).
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
564 Bulle in
The esul s ega ding ai eac ance con ibu e o p io esea ch in medicine and
psychology, which has shown ha ai eac ance is nega i ely associa ed wi h indi idual
e o s o ollow o icial heal h ecommenda ions (D´
ıaz and Co a, 2022)aswellas
wi h accina ion in en ions (D a˙zkowski and T epanowski, 2022). We complemen hese
indings by showing ha ai eac ance is s ongly associa ed wi h ac ual accina ion
beha iou and ha his associa ion is o high economic signi icance, as he likelihood o
being accina ed agains Co id-19 is 9 pe cen age poin s lowe in he g oup o esponden s
wi h high ai eac ance. Ou esul s u he sugges ha highe le els o neu o icism,
which ha e been shown o be ela ed o sel -p o ec i e beha iou o adhe ence o public
p o ec ion policies (e.g. K oencke e al.,2020), do no seem o play a majo ole wi h
ega d o accina ion beha iou . The same holds wi h ega d o he oles o locus o con ol
(De e eux, Mille , and Ki shenbaum, 2021; Olagoke, Olagoke, and Hughes, 2021)and
s a is ical nume acy (Lau e al.,2022). O e all, ou esul s unde pin he impo ance
o conside ing eac ance when designing heal h policies and heal h communica ion
campaigns (Reynolds-Tylus, 2019; Ball and Wozniak, 2022). Speci ically, we conclude
ha campaigns and policies in ended o encou age he un accina ed o ge inocula ed
should be speci ically a ge ed a hose wi h a high le el o ai eac ance.
The pape p oceeds as ollows. In he nex sec ion, we p o ide a de ailed desc ip ion
o ou da a se . In sec ion III, we p esen empi ical e idence on he ela ionship be ween
cogni i e and non-cogni i e ai s and indi iduals’ accina ion s a us and willingness o
ge accina ed. Sec ion IV p esen s he esul s o an AFT model o analyse he du a ion
un il i s accina ion. Sec ion Vpu s he empi ical esul s in o pe spec i e by e lec ing
on ou key indings in compa ison o he exis ing li e a u e. In sec ion VI, we discuss
limi a ions o ou s udy. Sec ion VII concludes.
II. Da a
Ou empi ical analysis is based on da a ob ained om h ee su ey wa es o abou
2,000 indi iduals li ing in he Ge man s a e o No h Rhine-Wes phalia conduc ed by
RWI – Leibniz Ins i u e o Economic Resea ch in coope a ion wi h he su ey ins i u e
o sa h oughou he yea 2021. The i s su ey ook place be ween 7 and 16 May (Wa e
1), ollowed by wo addi ional wa es collec ed be ween 26 July and 21 Augus (Wa e
2) and 20 Oc obe and 11 No embe (Wa e 3), espec i ely. Fo Wa e 2 and Wa e 3,
we e-sampled as many pa icipan s om Wa e 1 as possible o es ablish a panel da a
s uc u e. The su ey popula ion consis ed o a sub-sample o o sa’s household panel.
o sa main ains a la ge panel o po en ial su ey pa icipan s ha is ep esen a i e o
he popula ion o Ge man-speaking in e ne use s. The da a collec ion occu ed as pa
o a esea ch p ojec ha aimed a e alua ing he epidemiological, economic and social
e ec s o a con olled edemp ion o he se e e lock-down measu es imposed o igh he
Co id-19 pandemic in ce ain municipali ies o No h Rhine-Wes phalia.2These so-called
Model Municipali ies ha e been o e -sampled du ing he collec ion o he da a, i.e., hal
o he su ey pa icipan s li e in hese municipali ies. The su eys we e conduc ed online
2These municipali ies included he ci ies o Essen, Hamm, Cologne, K e eld, M¨
onchengladbach, Lennes ad ,
Lipps ad , M¨
uns e , and Soes , as well as he coun ies o Coes eld, D¨
u en, Pade bo n and Wa endo .
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 565
using a ool ha can be accessed on compu e s, able s and sma phones. The a ge g oup
we e pe sons aged 18 and abo e.
App oxima ely 3,000 esponden s pa icipa ed pe su ey wa e. Fo ou es ima ion
sample, we conside only hose indi iduals who pa icipa ed in all h ee wa es, which
a e 2,317 pa icipan s. Addi ionally, we exclude 100 pa icipan s who did no answe all
ques ions ele an o ou analysis, i.e., ques ions on hei willingness o ge accina ed
(elici ed in all h ee su ey wa es) and hei mon h o accina ion i hey ha e al eady
ecei ed hei i s accina ion dose (elici ed in Wa e 2), ques ions on cogni i e and
non-cogni i e cha ac e is ics, as well as on age, gende and educa ion le el (elici ed in
Wa e 1). To ob ain in o ma ion on indi iduals’ willingness o ge accina ed, we asked
in May 2021: ‘Vaccina ions agains Co ona s a ed in Ge many a he end o Decembe .
Ha e you al eady been accina ed agains Co ona,will you ge accina ed as soon as
you ha e he chance,o would you a he wai o no ge accina ed a all?’, allowing he
ollowing answe s:
•‘I ha e al eady been accina ed agains Co ona’ (Vaccina ed)
•‘Will ge accina ed as soon as I ha e he chance’ (Willing)
•‘Will a he wai ’ (Wai ing)
•‘Will no be accina ed a all’ (Unwilling)
•‘Don’ know / no speci ied’
We coun he 13 esponden s who epo ed ha hey did no know o p e e ed no
o speci y hei accina ion s a us as pa o he 100 excluded pa icipan s who did no
answe all ques ions ele an o ou analysis. In addi ion, we dele e 37 esponden s om
he sample as hey p o ided inconsis en esponses wi h ega d o hei accina ion da es.3
The inal es ima ion sample consis s o 2,180 indi iduals (see Figu e 1 o a isual
ep esen a ion o he exclusions).4
3These esponden s s a ed in May ha hey had al eady been accina ed bu indica ed in Wa e 2 ha hey had
been accina ed in June o July.
4In he ollowing, we examine whe he he e a e di e ences be ween he pa icipan s who a e included and
pa icipan s who a e no included, o example because hey did no pa icipa e in all h ee su ey wa es. Fi s ,
pa icipan s who we e excluded due o i em non- esponse do no di e om he inal es ima ion sample wi h
espec o gende and age, he wo a iables ha a e a ailable o all pa icipan s ega dless o esponse beha iou
(Table A2). Those excluded because o answe ing ques ions on hei accina ion s a us o hei psychological
cha ac e is ics wi h ‘Don’ know / no speci ied’ end o be sligh ly less educa ed han he inal es ima ion sample
(Table A3). Table A1 shows ha hose pa icipan s who did no pa icipa e in Wa es 2 and 3 a e signi ican ly
less likely o be accina ed bu mo e likely o be willing o ge accina ed in Wa e 1. This may be due o hei
signi ican ly lowe age. Addi ionally, pa icipan s who did no pa icipa e in Wa es 2 and 3 epo ed lowe le els o
neu o icism. As a obus ness check, we he e o e conduc an addi ional analysis (Table A14) wi h he la ge sample
o 2,893 indi iduals who pa icipa ed in Wa e 1, ega dless o hei pa icipa ion in subsequen wa es. Speci ically,
we es ima e ou main eg ession om Table A2 – in which we examine he associa ion be ween he cogni i e and
non-cogni i e cha ac e is ics and he willingness o ge accina ed using mul inomial logi models – wi h he la ge
sample. The esul s emain quali a i ely he same, in pa icula we ind a obus nega i e associa ion be ween ai
eac ance and being accina ed. Fu he mo e, we ind e idence o signi ican associa ions be ween he accina ion
s a us o he willingness o ge accina ed and neu o icism, locus o con ol, and s a is ical nume acy. Fo ou
analyses in he main body o he pape , we ocus on he sample o 2,180 esponden s who pa icipa ed in all h ee
su ey wa es, as his allows a di ec compa ison o he analyses on accina ion s a us ac oss all h ee su ey wa es
and he analyses on accina ion iming due o he cons an sample ac oss all di e en speci ica ions.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
566 Bulle in
FIGURE 1. Exclusions om he inal es ima ion sample
FIGURE 2. Vaccina ion s a us and willingness o ge accina ed in May 2021
Figu e 2shows he dis ibu ion o he willingness o ge accina ed in May 2021 as i
appea s in ou inal sample. A majo i y o 65% o he esponden s was al eady accina ed
in May 2021, and mos o he esponden s who we e no ye accina ed we e a leas
willing o ge accina ed (28% o he sample). Only a mino i y o abou 4% and 3% o
he indi iduals in ou sample p e e ed o wai o epo ed o no ge accina ed a all,
espec i ely. Mo e de ailed desc ip i e in o ma ion on he dis ibu ion o he willingness
o ge accina ed ac oss socio-economic g oups is p esen ed in Table A4.
In Wa e 2, conduc ed in July and Augus 2021, we asked he esponden s who
we e al eady accina ed a his poin in ime (94.6%) abou he mon h in which hey
ecei ed hei i s accina ion dose. These accina ion da es we e in luenced by a p io i y
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 567
FIGURE 3. Kaplan–Meie plo o accina ion mon hs as s a ed in July/Augus 2021 di e en ia ed by he
accina ion in en ions s a ed in May 2021
scheme imposed by he Ge man go e nmen o ensu e ha he mos ulne able g oups o
in ec ions we e accina ed i s . F om he s a o he accina ion campaign in Decembe
2020 un il 6 Ap il 2021, pe sons aged 80 o olde as well as pe sons who wo k in nu sing,
elde ly ca e, and heal h ca e we e p io i ized. F om 7 Ap il 2021, o 6 May 2021, he
p io i iza ion was ex ended o pe sons aged 70 o olde as well as o ch onically ill pe sons
and hei close ela i es o ca egi e s, con ac pe sons o p egnan pe sons, as well as
pe sons in public se ice, such as he police, i e depa men , and school se ice. F om
7 May 2021, o 6 June 2021, he p io i iza ion was u he ex ended o pe sons aged 60
o olde and o u he p o essional g oups ha ha e a lo o con ac wi h cus ome s o
wo k in sys em- ele an u ili ies. F om 7 June 2021, he e was no longe any special
p io i iza ion and e e yone could ge accina ed.
Figu e 3con as s he accina ion in en ions elici ed in Wa e 1 wi h he accina ion
iming elici ed in Wa e 2. I appea s ha he la ge majo i y o hose who s a ed in May
ha hey will ge accina ed as soon as possible indeed go accina ed in May, June o
July. Only a small sha e o his g oup was s ill no accina ed a he ime o Wa e 2
(3.4%). In con as , 57.5% o hose who s a ed ha hey would a he wai be o e ge ing
accina ed in Wa e 1 and 82.3% o hose who s a ed no in en ion o ge accina ed a all
we e s ill no accina ed in Wa e 2.
Table 1p esen s summa y s a is ics o he ou come and explana o y a iables used in
ou empi ical analysis. The la e can be ca ego ized in o wo g oups: Va iables desc ibing
he cogni i e and non-cogni i e cha ac e is ics o he indi iduals in ou sample, which a e
in he ocus o his s udy, and socio-economic cha ac e is ics. We measu ed i e di e en
cogni i e and non-cogni i e cha ac e is ics ha we conside ed ele an o accina ion
decisions based on he scien i ic li e a u e5as well as heo e ical conside a ions: ai
eac ance, neu o icism, locus o con ol and s a is ical nume acy.
5On ai eac ance, see D´
ıaz and Co a (2022) and Ball and Wozniak (2022). On neu o icism, see K oencke
e al. (2020). On locus o con ol, see De e eux e al. (2021) and Olagoke e al. (2021). Rega ding isk and heal h
li e acy in he con ex o pandemics and accina ion, s a is ical nume acy is conside ed a key elemen (Riise
e al.,2020; G ee e al.,2021).
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
568 Bulle in
TABLE 1
Summa y s a is ics based on he i s su ey wa e (May 2021)
Va iable Explana ion Mean
(I) Ou come a iables
Vaccina ed Dummy: 1 i esponden is al eady accina ed 0.65
Willing Dummy: 1 i esponden is willing o ge accina ed as soon as possible 0.28
Wai ing Dummy: 1 i esponden p e e s o wai 0.04
Unwilling Dummy: 1 i esponden is unwilling o ge accina ed 0.03
(II) Cogni i e and non-cogni i e cha ac e is ics
T ai eac ance
Low Dummy: 1 i esponden has a low le el o ai eac ance 0.24
Medium Dummy: 1 i esponden has a medium le el o ai eac ance 0.32
High Dummy: 1 i esponden has a high le el o ai eac ance 0.44
Neu o icism
Low Dummy: 1 i esponden has a low le el o Neu o icism 0.33
Medium Dummy: 1 i esponden has a medium le el o Neu o icism 0.29
High Dummy: 1 i esponden has a high le el o Neu o icism 0.38
Locus o Con ol
Low Dummy: 1 i esponden has a low in e nal locus o con ol 0.30
Medium Dummy: 1 i esponden has a medium in e nal locus o con ol 0.31
High Dummy: 1 i esponden has a high in e nal locus o con ol 0.39
Lipkus e al. nume acy scale
Low Dummy: 1 i esponden sco es low on he selec ed i ems o he nume acy
scale
0.30
High Dummy: 1 i esponden sco es high on he selec ed i ems o he nume acy
scale
0.70
Be lin nume acy es Dummy: 1 i esponden passed he Be lin Nume acy Tes (sho e sion) 0.25
(III) Socio-economic cha ac e is ics
Female Dummy: 1 i esponden is a woman 0.47
Age Age o esponden 57.8
(s anda d de ia ion) (14.4)
College deg ee Dummy: 1 i esponden has a college deg ee 0.34
T ai eac ance is p oxied using h ee selec ed i ems o he Hong Psychological
Reac ance Scale (Hong and Faedda, 1996). The pe sonali y cons uc ha his scale is
supposed o measu e is psychological eac ance, de ined as ‘ he mo i a ional s a e ha
is hypo hesized o occu when a eedom is elimina ed o h ea ened wi h elimina ion’
(B ehm and B ehm, 1981, p. 37) and ‘ eac ance p oduces a desi e o es o e one’s
a i udinal o beha io al eedom’ (Shen and Dilla d, 2005, p. 74). While eac ance can
be iewed as a mo i a ional s a e, i is a gued ha i can also be iewed as a pe sonali y
ai ha de e mines a pe son’s p oneness o eel and ac in a eac an manne (Shen and
Dilla d, 2005).
Neu o icismwas e ie edusing heneu o icism- ela edi emso heBigFi eIn en o y
(BFI-S) ha is used in he Ge man Socio-Economic Panel (Schupp and Ge li z, 2008).
This pe sonali y ai has wo poles: neu o icism on one side and emo ional s abili y on
he o he . Pe sons who sco e high on neu o icism ha e a s onge endency o expe ience
nega i e emo ions such as anxie y, ange , guil and dep ession (Widige , 2009, p. 129).
Locus o con ol was cap u ed using he o iginal i ems o he Psychological Coping
Resou ces componen o he Mas e y Module by Pea lin and Schoole (1978), which is,
o example, also used o measu e locus o con ol in Cobb-Cla k and Schu e (2013).
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 569
Locus o con ol cap u es an indi idual’s belie abou how s ongly e en s in he li e a e
shaped by own beha iou (Ro e , 1966;Ga zandKa el,1993). Indi iduals who belie e
ha e en s s ongly depend on own beha iou ha e a high in e nal locus o con ol, while
hose belie ing ha e en s a e mainly de e mined by a e o luck, o ins ance, ha e an
ex e nal locus o con ol.
S a is ical nume acy was measu ed in wo ways. Fi s , we u ilized h ee selec ed i ems
o Lipkus, Samsa, and Rime ’s (2001) ex ended nume acy scale, which we e amed in a
heal h con ex . These i ems we e ela i ely simple pe cen age calcula ion asks. The e o e,
we used a second indica o o s a is ical nume acy by employing he sho e sion o he
Be lin Nume acy Tes p oposed by Cokely e al. (2012), which is a mo e demanding es
o s a is ical nume acy.6
Fo ou empi ical analysis, we i s calcula e he indi idual mean esponses on he
Like scales o all i ems o each cha ac e is ic. Second, we di ide he sample o each
cha ac e is ic in o h ee app oxima ely equally sized ca ego ies: Low ( he mean o he
esponses is below he lowe e cile), medium ( he mean o he esponses is be ween he
lowe and he uppe e cile), high ( he mean o he esponses is abo e he uppe e cile).7
Excep ions a e he measu es o s a is ical nume acy. The measu e based on Lipkus
e al. (2001) yields only one bina y esponse (co ec o inco ec /’don’ know’) pe i em.
To agg ega e he i ems, we calcula e he sum o he co ec answe s pe esponden ,
which e eals ha almos 70% o he esponden s answe ed all i ems co ec ly. Since
his 70% includes bo h he lowe and he uppe e cile, he e a e only wo ca ego ies o
his measu e: Low s a is ical nume acy (a leas one inco ec o ‘don’ know’ esponse)
and high s a is ical nume acy (all esponses a e co ec ). Simila ly, he sho o m o he
Be lin Nume acy es p oduces only one bina y esul (co ec o inco ec /’don’ know’).
The e o e, we use a dummy a iable ha is 1 i he co ec answe o he es was gi en,
which is ue o 25% o he indi iduals in ou sample, and 0 i he answe was w ong o
he answe op ion ‘don’ know’ was chosen.
Table A5 compa es he s udy sample wi h he popula ion o he Ge man s a e o No h
Rhine-Wes phalia. The esul s show ha indi iduals aged 19–45 a e unde ep esen ed in
ou sample, while he 45–65 and 65–75 age g oups a e o e ep esen ed. The p opo ion
o women in ou sample is sligh ly lowe , while he p opo ion o college g adua es is
highe han in he gene al popula ion. Consis en wi h he highe a e age age o he s udy
popula ion, he e a e mo e ma ied and ewe employed pe sons in he sample and ewe
pe sons wi h child en li ing in he household han in he popula ion. The a e age income
ca ego y in he sample is sligh ly lowe han he a e age ne household income in he
popula ion. Finally, he p opo ion o accina ed indi iduals in ou sample is qui e high
6See Cokely e al. (2012) o a de ailed discussion on he di e ences be ween he wo measu emen ins umen s.
7Since he mean esponses end o consis o a limi ed numbe o disc e e alues due o he small numbe o i ems
pe ai , he e is a high p obabili y o ies in which one o hese disc e e alues lies exac ly on he e cile bounda y.
In his case, we assign all esponden s wi h his alue o he highe g oup. Fo his eason, he espec i e g oups a e
no exac ly o he same size. In Appendix E, we conduc obus ness checks and o m he g oups in an al e na i e
way: The low ca ego y consis s o hose indi iduals whose ounded mean esponse on he Like scale o all i ems o
a cha ac e is ic is ei he 1 o 2. The medium ca ego y consis s o hose whose mean esponse is 3 (on 5-poin Like
scales) o 3, 4 o 5 (on 7-poin Like scales). The high ca ego y consis s o he emainde . This app oach esul s
in mo e unequally sized g oups, whe e he low and high ca ego ies end o consis o hose wi h a a he ex eme
ealiza ion o he espec i e cha ac e is ic.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
576 Bulle in
TABLE 3
Coe icien s o Accele a ed Failu e Time models (Weibull dis ibu ion) on he ime un il accina ion
(1) (2) (3) (4) (5) (6)
Medium ai eac ance 0.020 0.020
(0.013) (0.012)
High ai eac ance 0.088** 0.087**
(0.012) (0.012)
Medium neu o icism 0.008 0.001
(0.013) (0.013)
High neu o icism −0.008 −0.015
(0.012) (0.012)
Medium locus o con ol 0.004 0.010
(0.013) (0.013)
High locus o con ol −0.015 −0.004
(0.012) (0.012)
Low on Lipkus e al. nume acy scale −0.002 0.002
(0.012) (0.012)
Be lin nume acy es passed 0.011 0.013
(0.014) (0.013)
College deg ee −0.046** −0.049** −0.046** −0.048** −0.050** −0.049**
(0.011) (0.011) (0.011) (0.011) (0.011) (0.011)
Female −0.002 −0.002 −0.004 −0.003 −0.003 0.000
(0.010) (0.011) (0.010) (0.011) (0.011) (0.011)
Age≥80 −0.305** −0.315** −0.310** −0.312** −0.311** −0.304**
(0.031) (0.032) (0.033) (0.033) (0.033) (0.030)
80>Age≥70 −0.191** −0.203** −0.202** −0.202** −0.200** −0.190**
(0.012) (0.012) (0.012) (0.012) (0.013) (0.013)
70>Age≥60 −0.111** −0.118** −0.114** −0.116** −0.115** −0.111**
(0.014) (0.014) (0.014) (0.014) (0.014) (0.014)
Cons an 1.844** 1.897** 1.899** 1.896** 1.892** 1.845**
(0.012) (0.012) (0.012) (0.010) (0.011) (0.017)
ln(p) 1.569** 1.551** 1.552** 1.551** 1.551** 1.571**
(0.021) (0.022) (0.022) (0.022) (0.022) (0.021)
No es: The sample is es ic ed o hose esponden s who pa icipa ed in all h ee wa es. Vaccina ion mon hs we e
elici ed in Wa e 2 and he co a ia es we e elici ed in Wa e 1. He e, we con ol o age ca ego ies a he han
con inuous age, which is due o he na u e o he accina ion p io i iza ion scheme ha made he accine accessible
in a s agge ed way acco ding o he exac age cu o s used he e o o m he age g oups. Robus s anda d e o s a e
epo ed in pa en heses. Numbe o obse a ions: 2,180.
** and * indica e s a is ical signi icance a he 1% and 5% le el, espec i ely.
Again, we compa e he e ec o high ai eac ance o he e ec o he educa ion le el
and ind ha he 9% inc ease in he du a ion o i s accina ion in he high ai eac ance
g oup is abou wice as la ge in absolu e e ms as he 5% dec ease in he du a ion o i s
accina ion in he college deg ee g oup.
The esul s assuming a Log-Logis ic dis ibu ion co obo a e he main esul s om
he Weibull model (Table A20). The excep ion is ha in his model, a high in e nal locus
o con ol seems o be signi ican ly ela ed o a sho e ime un il accina ion. Ye , his
e ec becomes non-signi ican in he combined model (Column (6) o Table A20). This
sugges s ha he esul s conce ning locus o con ol a e less obus han hose conce ning
eac ance.
An addi ional obus ness check is p esen ed in Table A21. This obus ness check
examines he ex en o possible ime agg ega ion bias, which a ises because he obse ed
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 577
accina ion da es a e disc e e in na u e due o he mon hly obse a ion hy hm, e en
hough he model assumes con inuous da a (Pe e sen, 1991). To assess he sensi i i y o
ou esul s in his ega d, we ollow Allison (2010) and ecode ou da a o be in e al
censo ed, i.e., a he han assigning esponden s a speci ic mon h-indica o as he ou come
a iable, we assign hem an in e al whe e he uppe in e al bound is he mon h-indica o
used be o e and he lowe in e al bound is he p e ious mon h-indica o plus an epsilon
o 0.0001. The la e has he pu pose o a oiding ze os o esponden s who we e
accina ed in he i s mon h obse ed. Responden s who we e no accina ed a he end
o he obse a ion pe iod a e assigned an in e al ha has an open uppe in e al bound,
while he lowe in e al bound akes he alue o he la es obse ed mon h. This in e al
censo ed da a can also be analysed by a Weibull AFT model. The esul s ob ained by using
his in e al censo ed model a e quali a i ely simila o he ones in Table 3(Table A21).
In endency, he coe icien s o he in e al censo ed model a e somewha la ge , bu he e
a e no changes in he signs o signi icance o he esul s.
As a inal obus ness check, we examine he ex en o which he esul s a e d i en by
indi iduals who we e no ye accina ed a he ime he accina ion da es we e elici ed and
who he e o e en e he AFT model as igh -censo ed obse a ions. To do his, we omi
hese indi iduals om he es ima ion sample (Table A22). The signs and signi icance o
he esul ing coe icien s a e simila o he esul s in Table 3. A high le el o ai eac ance
is s ill signi ican ly associa ed wi h a la e accina ion da e, al hough he magni ude o
his ela ionship is smalle han in he model ha includes un accina ed indi iduals.
V. Pu ing he esul s in o pe spec i e: ai eac ance and accina ion
Ou esul s sugges ha ai eac ance is he s onges p edic o o accina ion beha iou .
To pu ou inding in o pe spec i e, we iden i ied nine s udies ha also in es iga e he
ela ionship be ween ai eac ance and a i udes owa ds accina ion (see Table 4 o an
o e iew). These s udies suppo ou inding by showing ha ai eac ance is a p edic o
o hesi ancy o ge accina ed in samples om he USA (Alba acin e al.,2021), Poland
(D a˙zkowski and T epanowski, 2022), Tu key (Salali e al.,2022), and Finland (So e i
e al.,2023). Compa ed o ou s udy, which analyses ac ual accina ion beha iou agains
Co id-19 a se e al poin s in ime, he ocus o hese s udies was mo e on inding ou wha
he hypo he ical willingness o ge accina ed was be o e he accines agains Co id-19
we e a ailable (D a˙zkowski and T epanowski, 2022; So e i e al.,2023), how high he
in en ion o be accina ed is among hose who we e no accina ed a he ime o he s udy,
i.e., he sel - epo ed likelihood o being accina ed in he u u e (Salali e al.,2022), o
i was asked abou he willingness o ge accina ed agains a hypo he ical new disease
(Alba acin e al.,2021). Resea ch has also shown ha ai eac ance no only p edic s
hesi ancy o ge accina ed onesel in he con ex o he Co id-19 pandemic, bu also
co ela es wi h opinions abou childhood accina ion in gene al p io o he pandemic
(Ho nsey, Ha is, and Fielding, 2018; Finkels ein e al.,2020; So e i e al.,2020).
Fu he mo e, wo s udies ha e in es iga ed whe he ai eac ance mode a es he
e ec s o expe imen al manipula ions on a i udes owa ds accina ion: Alba acin
e al. (2021) ound ha indi iduals wi h highe ai eac ance do no eac di e en ly
o manda o y Co id-19 accine equi emen s han indi iduals wi h lowe ai eac ance.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
578 Bulle in
TABLE 4
Li e a u e o e iew on he ela ionship be ween ai eac ance and accina ion beha iou and
a i udes
Re e ence S udy Coun y
Sample
Size
Ou come
a iable
Rela ion be ween
he ou come
a iable and
ai eac ance P- alue
Salali
e al. (2022)
1 Tu key 1,013 Vaccina ion in en ion Nega i e 0.001
Alba acin
e al. (2021)
1 USA 357 Vaccina ion in en ion N.a. <0.001
D a˙zkowski
and
T epanowski(2022)
1 Poland 551 Vaccina ion in en ion Nega i e <0.01
Finkels ein
e al. (2020)
1 USA 300 Vaccina ion p io i y o
child en
Nega i e <0.01
Ho nsey
e al. (2018)
1 In e na ional,
25 coun ies
5,323 A i udes owa d acci-
na ion
Nega i e <0.001
So e i
e al. (2020)
1 Finland 770 In luenza accina ion
beha iou
Nega i e 0.001
Vaccine a i udes (gen-
e al)
Nega i e <0.001
Vaccine a i udes
(In luenza)
Nega i e <0.001
W igh and
Rune (2023)
1 Aus alia 1,050 Belie s abou accine
sa e y and e icacy
Nega i e <0.05
So e i
e al. (2023)
1 Finland 199 Vaccina ion willing-
ness
Nega i e <0.001
Vaccina ion a i udes Nega i e <0.001
2 Finland 293 Vaccina ion willing-
ness
Nega i e <0.001
3 Finland 398 S a e eac ance Nega i e <0.001
Vaccina ion willing-
ness in =1
Nega i e <0.001
Change in accina ion
willingness be ween
=1and =2
Nega i e 0.782
No es: Fo he ou come a iables, highe sco es indica e mo e posi i e a i udes and belie s, and highe in en ions
and p obabili ies o be accina ed.
By con as , So e i e al. (2023) ound ha equi emen s o ge accina ed agains
a hypo he ical disease as well as addi ional in o ma ion on he accine inc eases he
willingness o ge accina ed among indi iduals wi h low ai eac ance, bu no among
hose wi h high ai eac ance. Nei he s udy ound e idence o accine equi emen s
back i ing, i.e., accine equi emen s did no educe he willingness o ge accina ed.
The di e en and he e o e puzzling indings o Alba acin e al. (2021) and So e i
e al. (2023) ega ding eac ance a e consis en wi h he esul s o Lau in, Kay, and
Fi zsimons (2012). Thei indings indica e ha eac ance may be pa icula ly impo an
when ules a e no absolu e, i.e., when i is no ce ain whe he a ule would come in o
e ec o when a ule is no c edibly en o ced. By con as , people end o a ionalize and
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 579
hus accep ules ha a e absolu e. Consequen ly, he associa ion be ween he willingness
o ge accina ed and eac ance may be weake unde manda o y accina ion. Du ing he
Co id-19 pandemic, No h Rhine-Wes phalia came closes o manda o y accina ion wi h
he so-called ‘2G egula ion’, which es ic ed access o places and e en s o social li e
o only hose who had been accina ed o had eco e ed. This egula ion i s came in o
e ec on 24 No embe 2021 (MAGS, 2021), i.e., a e Wa e 3 was conduc ed. Du ing
ou s udy pe iod, he p essu e o ge accina ed was p ima ily based on non-absolu e
ules, i.e., on social p essu e, public appeals, and inc easing hu dles o un accina ed
indi iduals, e.g., compulso y es s. Howe e , hese measu es s ill allowed un accina ed
indi iduals o pa icipa e in social li e.
Beyond he opic o accina ion beha iou , i is e iden ha ai eac ance plays
a signi ican ole in de e mining beha iou in aspec s o social li e. T ai eac ance,
as ope a ionalized by Hong and Faedda (1996), has been demons a ed o p edic
eal beha iou in mul iple domains such as isky sexual ac i i y, d ug use (Mille
and Quick, 2010), and beha iou in close ela ionships (Cha and, Dal on, and
Fi zsimons, 2007). Fu he mo e, i has been shown ha eac ance mode a es he
ela ionship be ween elec onic su eillance and coun e p oduc i e beha iou (Yos
e al.,2019), be ween p omo ional a ou s and consume spending (Be ini and
Aydinli, 2020), and be ween ho els’ en i onmen al ini ia i es and consume s’ in en ion
o be en i onmen ally conscious (Wang, K ishna, and McFe an, 2017).
To summa ize, his sec ion suppo s wo poin s. Fi s , we no e ha ou esul s
a e consis en wi h p e ious indings on he ela ionship be ween ai eac ance and
accina ion in en ions, which u he s eng hens ou conclusions. Second, we can
highligh ou con ibu ion o he li e a u e as we obse e ac ual accina ion beha iou
agains Co id-19 a he han hypo he ical beha iou o in en ions. Fu he mo e, we look
a di e en cogni i e and non-cogni i e cha ac e is ics in one s udy a he han ocusing
on a single ai .
VI. Limi a ions
This s udy is subjec o limi a ions ha a e ypical o su ey-based co ela ion s udies.
These include he possibili y ha he sample di e s om he popula ion in ce ain
obse able and unobse able ac o s. Howe e , as discussed in sec ion II, we expec ha
he elabo a e o line ec ui men p ocess o he su ey panel om which he esponden s
we e andomly d awn educes he se e i y o his p oblem compa ed o pu e online
su eys.
Fu he mo e, sel - epo ed heal h measu emen s a e suscep ible o a ious ypes o
epo ing bias (see, e.g., Bound, 1991; Ke kho s and Lindeboom, 1995; Kap eyn, Smi h,
and an Soes , 2007; Black e al.,2017a; Black, Johns on, and Suziedely e, 2017b;
Da illas, de Oli ei a, and Jones, 2023), which may also be he case wi h ou measu e o
accina ion s a us. Fo example, due o he high social p essu e and public a en ion on
accina ion agains Co id-19 du ing he s udy pe iod, sel - epo s on accina ion s a us
may ha e been subjec o social desi abili y bias (Wol e e al.,2022). Howe e , we
belie e such biases o be unlikely o in alida e ou esul s. Fi s , he esponse op ions
‘Will ge accina ed as soon as I ha e he chance’ o ‘Will a he wai ’ allow esponden s
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
580 Bulle in
o answe u h ully ha hey ha e no ye been accina ed wi hou di ec ly e ealing ha
hey ac ually do no wan o ge accina ed a all. This would mean ha we would expec
some o hose who chose one o hese op ions o be ac ually no willing o ge accina ed
and may only ha e chosen hese op ions because hey a e pe cei ed as less socially
disapp o ed o han he op ion ‘Will no be accina ed a all’. As a obus ness check, we
he e o e e-es ima ed ou main analyses using a bina y dependen a iable ha akes he
alue 1 only i someone has al eady been accina ed (Table A17), wi h all h ee o he
op ions coded as 0. The obus ness check con i ms ou esul s. Addi ionally, he accine
was s ill in sho supply, pa icula ly in May 2021, meaning ha he majo i y o Ge mans
had no ye been accina ed. The e o e, i canno be assumed ha a esponden who was
no accina ed would ha e el compelled by social desi abili y o make a alse s a emen
abou hei accina ion s a us. Thi d, he su ey was conduc ed anonymously online,
and pa icipan s we e in o med ha hei in o ma ion could no be aced back o hem.
Finally, ou main esul on ai eac ance emains e iden when conside ing he iming
o esponden s’ accina ion, whe e social desi abili y o o he mo i a ed mis epo ing is
e en less likely (Table 3).
Mo eo e , ou esul s desc ibe he associa ions be ween accina ion and he di e en
cogni i e and non-cogni i e cha ac e is ics, a he han causal e ec s. Ideally, u u e
esea ch can p o ide causal e idence o he e ec s o speci ic cha ac e is ics on
accina ion beha iou . Howe e , o policy make s and also o socie y, i migh be
e en mo e impo an o build upon he esul s o ou and ela ed s udies and y o use
knowledge on he p edic o s o accina ion beha iou o a ge accina ion campaigns. We
show ha ai eac ance in pa icula is an impo an p edic o o accina ion beha iou .
Indi iduals wi h a high le el o ai eac ance display a signi ican ly lowe willingness
o ge accina ed.11 In e en ions could a ge especially hose wi h high ai eac ance
and should, o example, y o o e come pe cep ions o eedom h ea , which may
be e y p onounced among hese indi iduals (see, o ins ance, he li e a u e e iew by
Reynolds-Tylus, 2019). Such app oaches could be accompanied by scien i ic esea ch o
e alua e he causal e ec s o a ge ed in e en ions, o example by means o andomized
con olled ield expe imen s, and o in es iga e how hese e ec s a e mode a ed by
indi idual di e ences in ai eac ance.
11 Like p obably all s udies, we canno obse e all ac o s ha may a ec accina ion decisions. Fo example, we
lack in o ma ion on he heal h s a us and medical his o y o he pa icipan s, al hough i seems plausible ha his
has an in luence on he willingness o be accina ed. We would expec people in poo heal h o be mo e willing o
be accina ed (al hough his could also be he e ogeneous and equi es i s own in es iga ion). Assuming his is he
case, i is an omi ed a iable ha could dis o ou esul s. In pa icula , i people who a e eac an a e also mo e
likely o ha e a poo e heal h s a us o , con e sely, people wi h a poo e heal h s a us a e mo e likely o be eac an ,
i could a ec he es ima ion esul s. While we conside i use ul o in es iga e hese ela ionships, i is impo an
o emphasize ha his does no imply a limi a ion ega ding he p edic i e powe o ai eac ance. We show a
obus associa ion ha people wi h highe ai eac ance a e less likely o be accina ed, which can be used o a ge
in e en ions. This is ue e en i he unde lying heal h s a us would ac ually be ully media ing his associa ion.
Ul ima ely, howe e , we would no assume ha he co ela ion be ween ai eac ance and he willingness o ge
accina ed is only due o he heal h s a us.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 581
VII. Conclusions
Using da a ob ained om h ee su ey wa es o abou 2,000 indi iduals li ing in
he Ge man s a e o No h Rhine-Wes phalia, his pape in es iga es he associa ion o
di e en cogni i e and non-cogni i e cha ac e is ics wi h an indi iduals’ willingness
o ge accina ed agains Co id-19 as well as hei accina ion s a us. The empi ical
esul s indica e ha ai eac ance is he s onges p edic o o accina ion beha iou .
The p obabili y o being accina ed agains Co id-19 is 9 pe cen age poin s lowe o
indi iduals wi h a high le el o ai eac ance, compa ed o hose wi h a low o medium
le el o ai eac ance. In addi ion, he ime o accina ion is gene ally 9% highe o such
indi iduals. Fu he mo e, we ind en a i e e idence ha highe s a is ical nume acy, a
highe in e nal locus o con ol, and a mo e p onounced le el o neu o icism a e ela ed
o accina ion in en ions. Howe e , hese esul s a e less obus han hose ega ding ai
eac ance.
Wi h ega d o policy implica ions, ou esul s indica e ha accina ion campaigns
and policies could be imp o ed by speci ically add essing ce ain g oups o people, in
pa icula hose wi h a high le el o ai eac ance. To c ea e such a ge ed measu es,
we can build on he li e a u e p eda ing he Co id-19 pandemic, which discusses
ways o o e come eac ance in heal h communica ion (see, o ins ance, he li e a u e
e iew by Reynolds-Tylus, 2019). Fo example, u u e accina ion campaigns could use
na a i es ha p omo e empa hy, as Shen (2010) has shown ha s a e empa hy can
a enua e eac ance ega ding pe suasi e messages ela ed o smoking and alcohol use.
Mo eo e , pe suasi e messages may di e in he ex en o which hey a e pe cei ed
as eedom h ea ening. Acco ding o he esul s o a me a-analysis by Rains (2013),
he use o less eedom- h ea ening language may educe eac an eac ions o such
messages.
Mo e gene ally, by demons a ing ha eac ance has high explana o y powe , and
by using an easy- o-implemen measu e o ai eac ance, ou s udy may p o ide an
impe us o using his concep in o he beha iou al economic s udies o explain why
some g oups a e no eached by ce ain ypes o in e en ions. The Co id-19 pandemic
was a si ua ion in which he s a e s ongly a emp ed o in luence indi idual beha iou
in he sho e m. A simila si ua ion a ose in he win e o 2022/2023 when Eu ope
expe ienced an ene gy sho age due o he wa in Uk aine, and Eu opean go e nmen s
u ged p i a e households o sa e as much ene gy as possible. Bu e en beyond immedia e
c ises, he e a e many go e nmen in e en ions a ge ing indi idual beha iou al changes,
o example in he a eas o p e en i e heal h ca e and e i emen sa ings. In all
hese si ua ions, eac ance can be an impo an ac o limi ing he e ec i eness o
in e en ions.
Acknowledgmen
Open Access unding enabled and o ganized by P ojek DEAL.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
582 Bulle in
Final Manusc ip Recei ed: Augus 2023
Re e ences
Alba acin, D., Jung, H., Song, W., Tan, A. and Fishman, J. (2021). ‘Ra he han inducing psychological
eac ance, equi ing accina ion s eng hens in en ions o accina e in us popula ions’, Scien i ic Repo s,
Vol. 11, p. 20796.
Al a o, L., Faia, E., Lame sdo , N. and Saidi, F. (2022). ‘Heal h ex e nali ies and policy: he ole o social
p e e ences’, Managemen Science, Vol. 325, pp. 6751–6761.
Allco , H., Boxell, L., Conway, J., Gen zkow, M., Thale , M. and Yang, D. (2020). ‘Pola iza ion and
public heal h: Pa isan di e ences in social dis ancing du ing he co ona i us pandemic’, Jou nal o Public
Economics, Vol. 191, 104254.
Allison, P. D. (2010). Su i al Analysis Using SAS: A P ac ical Guide 2nd ed., SAS P ess, Ca y, No h
Ca olina.
Ando , M. A. (2021). ‘Co ona NRW – Fu he Analyses’, AEA RCT Regis y, May 21. h ps://doi.o g/10
.1257/ c .7710-1.0.
Ando , M. A., Cox, J., Ge s e , A., P ice, M., Somme , S. and Tombe g, L. (2022). Locus o Con ol and
P osocial Beha io , Na ional Bu eau o Economic Resea ch (NBER) Wo king Pape No. w30359.
Ball, H. and Wozniak, T. R. (2022). ‘Why do some ame icans esis co id-19 p e en ion beha io ? An
analysis o issue impo ance, message a igue, and eac ance ega ding co id-19 messaging’, Heal h
Communica ion, Vol. 37, pp. 1812–1819.
Ba gain, O. and Aminjono , U. (2020). ‘T us and compliance o public heal h policies in imes o co id-19’,
Jou nal o Public Economics, Vol. 192, 104316.
Bazzi, S., Fiszbein, M. and Geb esilasse, M. (2021). ‘”Rugged indi idualism” and collec i e (in)ac ion du ing
he co id-19 pandemic’, Jou nal o Public Economics, Vol. 195, 104357.
Be ini, M. and Aydinli, A. (2020). ‘Consume eac ance o p omo ional a o s’, Jou nal o Re ailing,Vol.
96, pp. 578–589.
Black, N., Johns on, D. W., Shields, M. A. and Suziedely e, A. (2017a). ‘Who p o ides inconsis en epo s
o hei heal h s a us? The impo ance o age, cogni i e abili y and socioeconomic s a us’, Social Science
& Medicine, Vol. 191, pp. 9–18.
Black, N., Johns on, D. W. and Suziedely e, A. (2017b). ‘Jus i ica ion bias in sel - epo ed disabili y: new
e idence om panel da a’, Jou nal o Heal h Economics, Vol. 54, pp. 124–134.
Bo ga, L. G., Cla k, A. E., D’Amb osio, C. and Lepin eu , A. (2022). ‘Cha ac e is ics associa ed wi h co id-19
accine hesi ancy’, Scien i ic Repo s, Vol. 12, p. 12435.
Bound, J. (1991). ‘Sel - epo ed e sus objec i e measu es o heal h in e i emen models’, Jou nal o Human
Resou ces, Vol. 26, pp. 106–138.
B ehm, S. S. and B ehm, J. W. (1981). Psychological Reac ance: A Theo y o F eedom and Con ol, Academic
P ess, New Yo k, New Yo k.
Came on, A. C. and T i edi, P. K. (2005). Mic oeconome ics: Me hods and Applica ions, Camb idge
Uni e si y P ess, New Yo k, New Yo k.
Cascini, F., Pan o ic, A., Al-Ajlouni, Y., Failla, G. and Riccia di, W. (2021). ‘A i udes, accep ance and
hesi ancy among he gene al popula ion wo ldwide o ecei e he co id-19 accines and hei con ibu ing
ac o s: A sys ema ic e iew’, eClinicalMedicine, Vol. 40, 101113.
CDC. (2022). Co id-19 Vaccines a e E ec i e, Cen e s o Disease Con ol and P e en ion (CDC), A lan a,
Geo gia. las upda e: June 28, 2022. h ps://www.cdc.go /co ona i us/2019-nco / accines/e ec i eness
/wo k.h ml, e ie ed on 12 Decembe 2022.
Cha and, T. L., Dal on, A. N. and Fi zsimons, G. J. (2007). ‘Nonconscious ela ionship eac ance: when
signi ican o he s p ime opposing goals’, Jou nal o Expe imen al Social Psychology, Vol. 43, pp. 719–726.
Chen, C., F ey, C. B. and P esiden e, G. (2021). ‘Cul u e and con agion: indi idualism and compliance wi h
co id-19 policy’, Jou nal o Economic Beha io & O ganiza ion, Vol. 190, pp. 191–200.
Cobb-Cla k, D. A. and Schu e , S. (2013). ‘Two economis s’ musings on he s abili y o locus o con ol’,
The Economic Jou nal, Vol. 123, pp. F358–F400.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 583
Cokely, E. T., Galesic, M., Schulz, E., Ghazal, S. and Ga cia-Re ame o, R. (2012). ‘Measu ing isk li e acy:
he Be lin Nume acy Tes ’, Judgmen and Decision Making, Vol. 7, pp. 25–47.
Cox, D. R. (1972). ‘Reg ession models and li e- ables’, Jou nal o he Royal S a is ical Socie y: Se ies B
(Me hodological), Vol. 34, pp. 187–202.
C ossley, T. F., Fishe , P. and Low, H. (2021). ‘The he e ogeneous and eg essi e consequences o co id-19:
e idence om high quali y panel da a’, Jou nal o Public Economics, Vol. 193, 104334.
Da illas, A., de Oli ei a, V. H. and Jones, A. M. (2023). ‘Is inconsis en epo ing o sel -assessed heal h
pe sis en and sys ema ic? E idence om he UKHLS’, Economics & Human Biology, Vol. 49, 101219.
De e eux, P. G., Mille , M. K. and Ki shenbaum, J. M. (2021). ‘Mo al disengagemen , locus o con ol, and
belie in a jus wo ld: indi idual di e ences ela e o adhe ence o co id-19 guidelines’, Pe sonali y and
Indi idual Di e ences, Vol. 182, 111069.
D´
ıaz, R. and Co a, F. (2022). ‘Reac ance, mo ali y, and disgus : he ela ionship be ween a ec i e disposi ions
and compliance wi h o icial heal h ecommenda ions du ing he co id-19 pandemic’, Cogni ion and
Emo ion, Vol. 36, pp. 120–136.
D a˙zkowski, D. and T epanowski, R. (2022). ‘Reac ance and pe cei ed disease se e i y as de e minan s o
co id-19 accina ion in en ion: an applica ion o he heo y o planned beha io ’, Psychology, Heal h &
Medicine, Vol. 27, pp. 2171–2178.
Fang, X., F eye , T., Ho, C.-Y., Chen, Z. and Goe e, L. (2022). ‘P osociali y p edic s indi idual beha io and
collec i e ou comes in he co id-19 pandemic’, Social Science & Medicine, Vol. 308, 115192.
Finkels ein, S. R., Boland, W. A., Vallen, B., Connell, P. M., She man, G. D. and Feems e , K. A. (2020).
‘Psychological eac ance impac s a ings o pedia ician accine- ela ed communica ion quali y, pe cei ed
accine sa e y, and accina ion p io i y among US pa en s’, Human Vaccines & Immuno he apeu ics,Vol.
16, pp. 1024–1029.
Ga z, M. and Ka el, M. J. (1993). ‘Indi idual change in pe cei ed con ol o e 20 yea s’, In e na ional Jou nal
o Beha io al De elopmen , Vol. 16, pp. 305–322.
G ee , M. L., Sample, S., Jensen, H. K., McBain, S., Lipschi z, R. and Sex on, K. W. (2021). ‘Co id-19 is
connec ed wi h lowe heal h li e acy in u al a eas’, S udies in Heal h Technology and In o ma ics,Vol.
281, pp. 804–808.
Hi sh, J. B., Kang, S. K. and Bodenhausen, G. V. (2012). ‘Pe sonalized pe suasion: ailo ing pe suasi e
appeals o ecipien s’ pe sonali y ai s’, Psychological Science, Vol. 23, pp. 578–581.
Hong, S.-M. and Faedda, S. (1996). ‘Re inemen o he Hong psychological eac ance scale’, Educa ional
and Psychological Measu emen , Vol. 56, pp. 173–182.
Ho nsey, M. J., Ha is, E. A. and Fielding, K. S. (2018). ‘The psychological oo s o an i- accina ion a i udes:
a 24-na ion in es iga ion’, Heal h Psychology, Vol. 37, pp. 307–315.
Hume , E., Jesse , A., Plene , P. L., P obs , T. and Pieh, C. (2021). ‘Educa ion le el and co id-19 accina ion
willingness in adolescen s’, Eu opean Child & Adolescen Psychia y, Vol. 32, pp. 537–539.
Judge, T. A., E ez, A., Bono, J. E. and Tho esen, C. J. (2002). ‘A e measu es o sel -es eem, neu o icism, locus
o con ol, and gene alized sel -e icacy indica o s o a common co e cons uc ?’, Jou nal o Pe sonali y
and Social Psychology, Vol. 83, pp. 693–710.
Kap eyn, A., Smi h, J. P. and an Soes , A. (2007). ‘Vigne es and sel - epo s o wo k disabili y in he Uni ed
S a es and The Ne he lands’, Ame ican Economic Re iew, Vol. 97, pp. 461–473.
Ke kho s, M. and Lindeboom, M. (1995). ‘Subjec i e heal h measu es and s a e dependen epo ing e o s’,
Heal h Economics, Vol. 4, pp. 221–235.
Kleinbaum, D. G. and Klein, M. (2012a). ‘Pa ame ic su i al models’, in Kleinbaum M. and Klein M. (eds),
Su i al Analysis: A Sel -Lea ning Tex , Sp inge , New Yo k.
Kleinbaum, D. G. and Klein, M. (2012b). ‘The s a i ied cox p ocedu e’, in Kleinbaum M. and Klein M. (eds),
Su i al Analysis: A Sel -Lea ning Tex , Sp inge , New Yo k.
K oencke, L., Geukes, K., U esch, T., Kupe , N. and Back, M. (2020). ‘Neu o icism and emo ional isk du ing
he co id-19 pandemic’, Jou nal o Resea ch in Pe sonali y, Vol. 89, 104038.
K oszne , R. S. and S ahan, P. E. (1999). ‘Wha d i es de egula ion? Economics and poli ics o he elaxa ion
o bank b anching es ic ions’, Qua e ly Jou nal o Economics, Vol. 114, pp. 1437–1467.
Lau, N. T., Wilkey, E. D., Sol anlou, M., Lagac´
e Cusiac, R., Pe e s, L., T emblay, P., Go in, C., Al es, I. S.,
Ribne , A. D., Thompson, C., Van Hoo , J., Bahnmuelle , J., Al a ez, A., Bellon, E., Coolen, I., Olli ie , F.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
584 Bulle in
and Ansa i, D. (2022). ‘Nume acy and co id-19: examining in e ela ionships be ween nume acy, heal h
nume acy and beha iou ’, Royal Socie y Open Science, Vol. 9, 201303.
Lau in, K., Kay, A. C. and Fi zsimons, G. J. (2012). ‘Reac ance e sus a ionaliza ion: di e gen esponses o
policies ha cons ain eedom’, Psychological Science, Vol. 23, pp. 205–209.
Lipkus, I. M., Samsa, G. and Rime , B. K. (2001). ‘Gene al pe o mance on a nume acy scale among highly
educa ed samples’, Medical Decision Making, Vol. 21, pp. 37–44.
MAGS. (2021). Neue Co ona-Schu z e o dnung ab 24. No embe 2021, Minis y o Labo , Heal h and
Social A ai s o he S a e o No h Rhine-Wes phalia (MAGS), D¨
usseldo . h ps://www.mags.n w
/p essemi eilung/neue-co onaschu z e o dnung-ab-24-no embe -2021, e ie ed on Feb ua y 28, 2024.
Mille , C. H. and Quick, B. L. (2010). ‘Sensa ion seeking and psychological eac ance as heal h isk p edic o s
o an eme ging adul popula ion’, Heal h Communica ion, Vol. 25, pp. 266–275.
Olagoke, A. A., Olagoke, O. O. and Hughes, A. M. (2021). ‘In en ion o accina e agains he no el 2019
co ona i us disease: he ole o heal h locus o con ol and eligiosi y’, Jou nal o Religion and Heal h,
Vol. 60, pp. 65–80.
Pea lin, L. I. and Schoole , C. (1978). ‘The s uc u e o coping’, Jou nal o Heal h and Social Beha io ,Vol.
19, pp. 2–21.
Pe e sen, T. (1991). ‘Time-agg ega ion bias in con inuous- ime haza d- a e models’, Sociological
Me hodology, Vol. 21, pp. 263–290.
Rains, S. A. (2013). ‘The na u e o psychological eac ance e isi ed: a me a-analy ic e iew’, Human
Communica ion Resea ch, Vol. 39, pp. 47–73.
Reynolds-Tylus, T. (2019). ‘Psychological eac ance and pe suasi e heal h communica ion: a e iew o he
li e a u e’, F on ie s in Communica ion, Vol. 4, p. 56.
Riise , K., Helse h, S., Ha alds ad, K., To bjo nse, A. and Richa dsen, K. R. (2020). ‘Adolescen s’ heal h
li e acy, heal h p o ec i e measu es, and heal h- ela ed quali y o li e du ing he co id-19 pandemic’, PLoS
One, Vol. 18, e0238161.
Ro e , J. B. (1966). ‘Gene alized expec ancies o in e nal e sus ex e nal con ol o ein o cemen ’,
Psychological Monog aphs: Gene al and Applied, Vol. 80, pp. 1–28.
Salali, G. D., Uysal, M. S., Bozyel, G., Akpina , E. and Aksu, A. (2022). ‘Does social in luence a ec
co id-19 accina ion in en ion among he un accina ed?’, E olu iona y Human Sciences, Vol. 4, e32.
Schupp, J. and Ge li z, J.-Y. (2008). Big Fi e In en o y-SOEP (BFI-S), GESIS - Leibniz Ins i u e o he
Social Sciences, Mannheim/Cologne.
Shen, L. (2010). ‘Mi iga ing psychological eac ance: he ole o message-induced empa hy in pe suasion’,
Human Communica ion Resea ch, Vol. 36, pp. 397–422.
Shen, L. and Dilla d, J. P. (2005). ‘Psychome ic p ope ies o he Hong psychological eac ance scale’,
Jou nal o Pe sonali y Assessmen , Vol. 85, pp. 74–81.
Solon, G., Haide , S. J. and Woold idge, J. M. (2015). ‘Wha a e we weigh ing o ?’, Jou nal o Human
Resou ces, Vol. 50, pp. 301–316.
So e i, A., Ka lsson, L. C., M¨
aki, O., An olk, J., Wa is, O., Ka lsson, H., Ka lsson, L., Lind el , M. and
Lewandowsky, S. (2020). ‘T ai eac ance and us in doc o s as p edic o s o accina ion beha io , accine
a i udes, and use o complemen a y and al e na i e medicine in pa en s o young child en’, PLoS One,
Vol. 15, e0236527.
So e i, A., Ka lsson, L. C., M¨
aki, K. O., Hol o d, D., Fasce, A., Schmid, P., An olk, J., Ka lsson, L., Ka lsson,
H., Nol i, S., Ka uki i, M., Lind el , M. and Lewandowsky, S. (2023). ‘T ai eac ance as psychological
mo i a ion o ejec accina ion: wo longi udinal s udies and one expe imen al s udy’, Applied Psychology:
Heal h and Well-Being, Vol. 16, pp. 597–614.
Wang, W., K ishna, A. and McFe an, B. (2017). ‘Tu ning o he ligh s: consume s’ en i onmen al e o s
depend on isible e o s o i ms’, Jou nal o Ma ke ing Resea ch, Vol. 54, pp. 478–494.
Widige , T. A. (2009). ‘Neu o icism’, in Lea y M. R. and Hoyle R. H. (eds), Handbook o Indi idual
Di e ences in Social Beha io , Guil o d P ess, New Yo k, New Yo k.
Widige , T. A. and Ol manns, J. R. (2017). ‘Neu o icism is a undamen al domain o pe sonali y wi h
eno mous public heal h implica ions’, Wo ld Psychia y, Vol. 16, pp. 144–145.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
P edic o s o indi idual beha iou in pandemics 585
Wol e , F., Maye l, J., Ande sen, H. K., Wieland, T. and Junke mann, J. (2022). ‘O e es ima ion o co id-19
accina ion co e age in popula ion su eys due o social desi abili y bias: esul s o an expe imen al
me hods s udy in Ge many’, Socius, Vol. 8, 23780231221094749.
W igh , D. and Rune, K. T. (2023). ‘Unde lying mo i a o s o an i- accina ion a i udes amongs egional
Sunshine Coas pa en s in Aus alia’, Heal h P omo ion Jou nal o Aus alia, Vol. 34, pp. 579–586.
Yos , A. B., Beh end, T. S., Howa dson, G., Badge Da ow, J. and Jensen, J. M. (2019). ‘Reac ance o
elec onic su eillance: a es o an eceden s and ou comes’, Jou nal o Business and Psychology, Vol. 34,
pp. 71–86.
Suppo ing In o ma ion
Addi ional Suppo ing In o ma ion may be ound in he online Appendix:
Appendix S1. Supplemen a y In o ma ion.
©2024 The Au ho (s). Ox o d Bulle in o Economics and S a is ics published by Ox o d Uni e si y and John Wiley & Sons L d.
