Sonnen ag, Sabine; Ka k, Roni ; Venz, Lau a
A icle — Published Ve sion
Leade suppo o eco e y: A mul i‐le el app oach o
employee psychological de achmen om wo k
Jou nal o Occupa ional and O ganiza ional Psychology
P o ided in Coope a ion wi h:
John Wiley & Sons
Sugges ed Ci a ion: Sonnen ag, Sabine; Ka k, Roni ; Venz, Lau a (2024) : Leade suppo o eco e y:
A mul i‐le el app oach o employee psychological de achmen om wo k, Jou nal o Occupa ional
and O ganiza ional Psychology, ISSN 2044-8325, Wiley Pe iodicals, Inc., Hoboken, NJ, Vol. 97, Iss. 4,
pp. 1762-1788,
h ps://doi.o g/10.1111/joop.12538
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/313778
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/
1762
|
J Occup O gan Psychol. 2024;97:1762–1788.
wileyonlinelib a y.com/jou nal/joop
Recei ed: 15 Janua y 2024
|
Accep ed: 14 July 2024
DOI: 10.1111/joop.12538
RESEARCH ARTICLE
Leade suppo o eco e y: A mul i- le el app oach
o employee psychological de achmen om wo k
Sabine Sonnen ag1 | Roni Ka k2,3 | Lau a Venz4
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.
© 2024 The Au ho (s). Jou nal o Occupa ional and O ganiza ional Psychology published by John Wiley & Sons L d on behal o The B i ish
Psychological Socie y.
1Depa men o Psychology, Uni e si y o
Mannheim, Mannheim, Ge many
2Depa men o Psychology, Ba - Ilan Uni e si y,
Rama Gan, Is ael
3Exe e School o Business, Exe e Uni e si y,
Exe e , UK
4Ins i u e o Managemen and O ganiza ion,
Leuphana Uni e si y Lünebu g, Lünebu g,
Ge many
Co espondence
Sabine Sonnen ag, Depa men o Psychology,
School o Social Sciences, Uni e si y o
Mannheim, A5, 6, C104, D- 68159 Mannheim,
Ge many.
Email: sabine.sonnen [email protected]
Funding in o ma ion
SHRM Founda ion
Abs ac
This esea ch examines he ole o leade s o employee
eco e y. We hypo hesize ha leade suppo o eco e y
(empa hy o eco e y, espec o bounda ies, and ole
modelling) ela es posi i ely o employees' psychological de-
achmen om wo k du ing non- wo k ime ha , in u n,
p edic s well- being ou comes. We a gue ha leade suppo
o eco e y can only be e ec i e when he leade - membe
exchange (LMX) ela ionship quali y is su icien ly high.
In a se ies o scale- de elopmen and scale- alida ion s ud-
ies, we demons a ed he cons uc and con en alidi y o a
new measu e o leade suppo o eco e y. We es ed ou
hypo heses wi h dia y da a collec ed om 152 employees.
Respec o bounda ies was posi i ely ela ed o employees'
psychological de achmen om wo k du ing non- wo k ime
a he pe son le el. Psychological de achmen om wo k
was posi i ely ela ed o low emo ional exhaus ion and a
high mo ning eco e y s a e, bo h a he pe son and he day
le el. LMX mode a ed he ela ionship be ween leade sup-
po o eco e y (o e all measu e), empa hy o eco e y,
and espec o bounda ies on he one hand and psycho-
logical de achmen on he o he hand, such ha he ela ion-
ships became non- signi ican when LMX was lowe . The
s udy sugges s ha leade s in high- quali y ela ionships can
con ibu e o employee eco e y – a p ocess ha helps o
main ain employee well- being.
KEYWORDS
leade ship, mul ile el me hods, eco e y, well- being
|
1763
LEADER SUPPORT FOR RECOVERY
BACKGROUND
Reco e y om daily wo k is c ucial o employee heal h and well- being (Sonnen ag e al., 2022; S eed
e al., 2021). Du ing s ess ul imes, i is pa icula ly impo an ha employees men ally de ach om
wo k du ing a e - wo k hou s and ha hey gain some men al dis ance om wo k so ha hey can s a
he nex wo kday e eshed and wi h new ene gy (Clin on e al., 2017; Liu e al., 2021). Resea ch has
iden i ied se e al ac o s ha acili a e e sus hinde psychological de achmen om wo k, including
speci ic leisu e ac i i ies (Hahn e al., 2012; Liu e al., 2021), job demands (Ge meys & De Gie e , 2018),
and in e pe sonal job s esso s (Rod íguez- Muñoz e al., 2017).
Al hough he e a e a ew excep ions (Benne e al., 2016; Temen e al., 2020), he ole o leade ship
o employee eco e y in gene al and o employee psychological de achmen in pa icula has been
la gely neglec ed. Leade s, howe e , migh be e y impo an o employee eco e y and psychological
de achmen , espec i ely, because hey in luence employees' pe spec i es on wo k and on wo k- home
bounda ies (Kim & Beeh , 2020; Liao e al., 2016). Mo e speci ically, leade s migh acili a e hei em-
ployees' eco e y by being mind ul o signs o s ain and deple ion in hei employees, acknowledging
hei eco e y needs, and by eminding hem abou he impo ance o eco e y. Howe e , leade s also
migh unde mine hei employees' eco e y by iola ing bounda ies be ween wo k and non- wo k li e,
o ins ance, by con ac ing hem du ing leisu e ime.
To be e unde s and he ole leade s play in employee eco e y we, i s , de elop and alida e a
scale o measu e leade suppo o eco e y, and, second, sugges and es a model ha in oduces
leade suppo o eco e y as an impo an p edic o o employee eco e y (Figu e 1). Speci ically,
we p opose ha leade suppo o eco e y is posi i ely ela ed o ollowe s' psychological de ach-
men om wo k du ing a e - wo k hou s and ha psychological de achmen , in u n, is associa ed
wi h lowe le els o exhaus ion and highe le els o mo ning eco e y s a es. We suppose ha leade
suppo o eco e y may no be e ec i e unde all ci cums ances and examine leade - membe
exchange (LMX; Liden e al., 1997) as a mode a o o he ela ionship be ween leade suppo o
eco e y and employee psychological de achmen om wo k. To cap u e he subs an ial wi hin-
pe son a ia ion in eco e y- ela ed cons uc s and well- being (Podsako e al., 2019), we collec
daily su ey da a. We include leade suppo o eco e y and LMX as pe son- le el a iables in he
mul i- le el s udy design.
We seek o make se e al con ibu ions. Fi s , ou s udy adds o he eco e y li e a u e by exam-
ining he ole o leade ship o employee eco e y. Un il now, he as majo i y o s udies in es i-
ga ing p edic o s o eco e y ocused on job demands and job esou ces (e.g., job con ol; Benne
e al., 2018; S eed e al., 2021). We ex end he p edic o space by showing ha also he leade , as an
impo an aspec o he in e pe sonal wo k en i onmen , plays a co e ole in acili a ing employees'
eco e y p ocesses.
Second, we sugges leade suppo o eco e y as a speci ic new mul idimensional cons uc . As
such, we highligh ha leade suppo o eco e y encompasses mul iple ace s ha leade s need o con-
side when hey wan o suppo hei ollowe s' eco e y p ocesses. We build on ea lie wo k by Benne
P ac i ione poin s
• Leade s play an impo an ole in employees' eco e y.
• Employees can be e psychologically de ach om hei wo k du ing non- wo k ime when
leade s espec bounda ies be ween wo k and non- wo k li e. Respec ing bounda ies should
be a op p io i y o leade s.
• Leade suppo o eco e y canno be e ec i e in low- quali y LMX ela ionships. Leade s
should suppo hei employees' eco e y and s i e o build good ela ionships wi h hem.
1764
|
SONNENTAG e al.
e al. (2016) ha demons a ed ha leade s' expec a ions ha employees wo k du ing leisu e ime can
unde mine employee eco e y. Howe e , we go beyond Benne e al.'s app oach and a gue ha leade
suppo o eco e y is mo e han jus no expec ing employees o show excessi e wo k commi men
du ing non- wo k ime. Ra he , leade suppo o eco e y includes beha iou s ha explici ly p omo e
eco e y, o ins ance by exp essing empa hy o eco e y and ac ing as a ole model. To be able o ex-
amine he mul idimensional na u e o leade suppo o eco e y, we de elop and alida e sho scales
ha cap u e a ious aspec s o he cons uc .
Thi d, we con ibu e o he b oade li e a u e on leade ship and well- being (Ha ms e al., 2017;
Mon ano e al., 2017) by ex ending he knowledge abou he bene i s o leade - membe exchange
(LMX). While Benne e al. (2016) epo ed a di ec nega i e associa ion be ween high LMX and em-
ployee eco e y, we demons a e ha a mo e nuanced pe spec i e on LMX is needed. Al hough a high
LMX migh lead o ollowe o e commi men and neglec o eco e y (Benne e al., 2016; Ha is &
Kacma , 2006), a high LMX is needed so ha ollowe s can us ha hei leade exp esses suppo o
eco e y ou o a bene olen mo i e.
CORE CONCEPTS
We de ine leade suppo o eco e y as a leade 's beha iou s ha “encou age employees o eco e
du ing nonwo k ime” (Benne e al., 2016, p. 16441). Building on his b oad de ini ion, we concep ual-
ize leade 's suppo o eco e y as a mul idimensional cons uc comp ising empa hy o eco e y, e-
spec o bounda ies, and ole modelling as h ee essen ial building blocks. We posi ion ou
concep ualiza ion o leade suppo o eco e y wi hin he b oade li e a u e on supe iso social sup-
po . Social suppo in gene al can be de ined as he p o ision o “psychological o ma e ial esou ces
… o a ocal indi idual by pa ne s in some o m o social ela ionship” (Jolly e al., 2021, p. 292) and
comp ises emo ional, in o ma ional, and ins umen al suppo as i s co e dimensions (House e al., 1988;
Jolly e al., 2021).
Emo ional suppo e e s o he exp ession o empa hy and communica es ha he a ge pe son
is ca ed o and app ecia ed (Cobb, 1976; Semme e al., 2008). In he con ex o leade suppo o
eco e y, emo ional suppo means ha leade s show empa hy o hei employees' eco e y needs.
In o ma ional suppo e e s o gi ing ad ice and cogni i e guidance and he eby helping o de ine
and unde s and di icul si ua ions (Cohen & Wills, 1985). One impo an way by which leade s can
1Al hough Benne e al.'s de ini ion is e y sui able o cap u ing he essence o leade suppo o eco e y as a b oad mul i- dimensional
concep , Benne e al.'s measu e is compa ably na ow and ocusses on jus one aspec o he b oade concep (i.e., expec a ion o wo k du ing
pe sonal ime).
FIGURE 1 Concep ual model.
|
1765
LEADER SUPPORT FOR RECOVERY
in o ma ionally suppo employee eco e y is by p o iding a pe sonal example abou how hey in eg a e
eco e y in o a busy li e, ha is, by ole modelling. Gene ally, ins umen al suppo includes inancial
help and he p o ision o o he esou ces and angible suppo (Cohen & Wills, 1985). I may also in-
clude ac ions ha gi e “ he ecipien inc eased ime o ac i i ies such as elaxa ion o en e ainmen ”
(Cohen & Wills, 1985, p. 313). Applied o leade suppo o eco e y, he mos impo an ins umen al
suppo may no be he p o ision o any ma e ial esou ce, bu o e ain om in uding in o employees'
ee ime, ha is, o espec bounda ies be ween wo k and non- wo k ime.
Empa hy o eco e y
Empa hy – he endency o unde s and and expe ience in e nal s a es o o he pe sons (Cla k e al., 2019)
– plays an impo an ole in o ganiza ional li e and is an impo an aspec o e ec i e leade ship (Bu ch
e al., 2016). In line wi h a mo e gene al de ini ion o leade empa hy, we de ine leade empa hy o
eco e y as a leade 's “abili y o accu a ely ecognize and unde s and” (Meinecke & Kau eld, 2019,
p. 487) ha employees need eco e y. Empa hy o eco e y e e s o he emo ional aspec s o leade
social suppo . I means ha leade s a e a en i e o hei employees' eco e y needs, ha hey ac-
knowledge ha eco e y is impo an and ha hey le hei employees know ha hey unde s and hei
eco e y needs.
Respec o bounda ies
Acknowledging employees' eco e y needs is impo an , bu i is no su icien o e ec i ely suppo -
ing employee eco e y. To help employees eco e , leade s need o espec he bounda ies be ween
employees' wo k and non- wo k li es. Pe de ini ion, eco e y is impeded when wo k in udes in o non-
wo k hou s (Meijman & Mulde , 1998). These in usions can ake se e al o ms, wi h comple ing wo k-
ela ed asks and being con ac ed o wo k pu poses belonging o he mos p ominen in usions (Bu s
e al., 2015; Ten B ummelhuis & T ougakos, 2014). Ob iously, leade s a e an impo an sou ce o hese
in usions, as hey se expec a ions abou ask accomplishmen and es ablish no ms abou job- ela ed
communica ion du ing non- wo k hou s. To enable eco e y, leade s need o hold he bounda ies a ound
hei subo dina es' non- wo k ime. Typically, leade s who espec bounda ies e ain om assigning
asks ha need o be comple ed li e ally “o e nigh ” and do no con ac hei employees du ing a e -
wo k hou s o on he weekend. Respec o bounda ies is a co e componen o he ins umen al aspec
o leade social suppo . A leade 's beha iou is mos ins umen al o employee eco e y when he
leade p o ides ime and “peace o mind” by no in uding in o employees' pe iods o es . Benne
e al.'s (2016) measu e o supe iso y suppo o eco e y cap u es his dimension o leade suppo
o eco e y as i s main ea u e using i ems such as “I communica e wi h subo dina es du ing non-
wo k ime” and “The mo e p oduc i e employees wo k du ing non- wo k ime” (p. 1654; bo h i ems
e e se- coded).
Role modelling
Role modelling is an impo an s a egy by which leade s can in luence hei employees (B own &
T e iño, 2014; Ya e & Ka k, 2011). In he eco e y con ex , ole modelling means ha leade s se
a posi i e example o eco e y p ac ices, by communica ing abou hei own eco e y needs and
ac i i ies. Fo ins ance, when ole modelling eco e y, leade s ell hei ollowe s abou how hey
in eg a e eco e y in o hei daily li es, how hey manage o ake ime o eco e y, and which e-
co e y ac i i ies hey enjoy. Role modelling co esponds o he in o ma ional aspec o leade social
suppo . Acco ding o social lea ning heo y (Bandu a, 1977), employees obse e how leade s speak
1766
|
SONNENTAG e al.
abou hei pe sonal eco e y, ex ac in o ma ion om hei leade s' epo s abou eco e y, and
d aw conclusions abou he alue o eco e y and how hey hemsel es could implemen eco e y
in o hei li es. Thus, leade s' ole modelling o eco e y will ha e a posi i e impac on employees'
eco e y. Bene i s o leade ole modelling ha e been demons a ed in he wo k- amily li e a u e:
Fo ins ance, when supe iso s ole model wo k- amily balance, employees expe ience less wo k-
amily con lic (Hamme e al., 2009).
Leade suppo o eco e y as a pe son- le el cons uc
In line wi h ea lie esea ch (Benne e al., 2016; Nie e al., 2021), we ea leade suppo o eco e y
as a pe son- le el a iable. The deg ee o which leade s exp ess empa hy o eco e y needs, espec em-
ployees' bounda ies, and ole model eco e y di e s om leade o leade . O cou se, leade s mus enac
hese beha iou s on ce ain days, and he pe son- le el deg ee o suppo o eco e y is he agg ega e o
leade beha iou s occu ing on indi idual days. We assume ha his o e all pe son- le el suppo o e-
co e y is mo e ele an han wha migh happen on each indi idual day because leade s migh no ha e
con ac wi h hei employees e e y day (Ba nes e al., 2015) and migh no add ess he eco e y opic
e e y day. Mos likely, leade s do no need o alk abou eco e y e e y day; o employees, i should be
mo e impo an o know ha hei leade gene ally suppo s eco e y. Leade s' suppo o eco e y can
also be e ec i e on days when leade s and employees do no mee o do no alk abou eco e y.
HYPOTHESES DEVELOPMENT
Leade suppo o eco e y and employee eco e y a he pe son le el
Leade suppo o eco e y will help employees de ach om wo k du ing non- wo k ime. Psychological
de achmen om wo k du ing non- wo k ime (i.e., empo al men al disengagemen om wo k;
Sonnen ag & F i z, 2007) is a co e eco e y expe ience and, as such, plays a c ucial ole o employee
well- being (S eed e al., 2021). Psychological de achmen does no only mean a oiding wo k du ing non-
wo k hou s bu also e aining om wo k- ela ed hough s. Such hough s would keep he psychobio-
logical sys em ac i a ed (B osscho e al., 2006) and he e o e impede es o a ion.
Leade suppo o eco e y will be posi i ely ela ed o employees' psychological de achmen o
se e al easons. Fi s , when leade s show empa hy o eco e y, hey acknowledge ha eco e y is im-
po an . They gi e hei employees he eeling ha hey unde s and hei eco e y needs and ake hem
se iously. When employees expe ience ha hei eco e y needs a e aken se iously, hey eel unde s ood
(Longmi e & Ha ison, 2018) so ha hey can see hei own eco e y needs as being legi ima e. This
expe ience, in u n, should help hem o espond o hese needs by p io i izing eco e y du ing non-
wo k pe iods. They know ha hey need no bo he abou wo k du ing a e - wo k hou s, which makes
de achmen om wo k mo e likely. When leade empa hy o eco e y is low, howe e , employees can-
no be su e ha hei eco e y needs a e legi ima e and ha hey a e “allowed” o de ach om wo k.
Acco dingly, hey will be less likely o de ach.
Second, espec ing bounda ies acili a es employees' psychological de achmen om wo k. When
leade s do no espec bounda ies by, o ins ance, alloca ing u gen asks sho ly be o e he end o he
wo kday o by con ac ing employees a e wo k, hey keep employees men ally connec ed o wo k and,
consequen ly, employees' psychological de achmen will be low. Impo an ly, no only leade s' ac ual
bounda y- iola ing beha iou s bu also he possibili y ha he leade migh c oss he bounda y and
in ude in o non- wo k ime may keep wo k- ela ed hough s ac i a ed and impede psychological de ach-
men . Howe e , when employees know ha hei leade s espec he bounda ies, employees can ully
disconnec om wo k du ing non- wo k ime. Empi ical indings om Benne e al. (2016) suppo his
assump ion. When supe iso s endo sed s a emen s such as “I communica e wi h subo dina es du ing
|
1767
LEADER SUPPORT FOR RECOVERY
non- wo k ime”, employees we e less likely o belong o a lea ing- wo k- behind eco e y p o ile (i.e., a
p o ile cha ac e ized by a high le el o psychological de achmen ).
Thi d, leade ole modelling will help employees wi h psychological de achmen om wo k. When
leade s model pu ing a en ion on eco e y, hey con ey ha eco e y is a desi able goal ha will in-
spi e employees o s i e o eco e y as well, o ins ance by e aining om job- ela ed hough s and
ac i i ies du ing non- wo k ime. Mo eo e , when ole modelling eco e y, leade s may e en ell how
hey achie e a s a e o eco e y, ansmi ing help ul in o ma ion o hei employees, who in u n will
lea n how o implemen hese s a egies hemsel es (Mo gen o h e al., 2015). Pu suing eco e y goals
and knowing how o eco e will make i mo e likely ha employees achie e psychological de achmen
om wo k du ing non- wo k ime.
Hypo hesis 1. A he pe son le el, leade suppo o eco e y is posi i ely ela ed o
psychological de achmen om wo k.
Leade suppo o eco e y will no only be ela ed o employees' psychological de achmen om
wo k bu may be e lec ed in highe employee well- being, such as lowe le els o exhaus ion and highe
le els o eeling eco e ed in he mo ning. Acco dingly, we p opose ha leade suppo o eco e y will
ha e indi ec e ec s on employee exhaus ion and he mo ning s a e o being eco e ed ia psychological
de achmen om wo k.
Exhaus ion is a s ess- ela ed expe ience ha “ e e s o eelings o being o e ex ended and deple ed
o one's emo ional and physical esou ces” (Maslach e al., 2001, p. 399). Psychological de achmen
om wo k du ing non- wo k ime should ela e nega i ely o exhaus ion. Employees who de ach om
wo k o ge abou hei wo k demands du ing non- wo k ime. Thus, wo k is no longe men ally p esen
o hem, and hei psychobiological sys ems ha ha e been axed du ing wo k a e no longe ac i a ed
and can e u n o hei p e- s esso le els (Meijman & Mulde , 1998). As a consequence, exhaus ion
declines, esul ing in lowe exhaus ion le els du ing a e - wo k hou s. In line wi h his p edic ion,
esea ch ound nega i e pe son- le el co ela ions be ween psychological de achmen and exhaus ion
(Head ick e al., 2022). Taken oge he , because leade suppo o eco e y should be posi i ely ela ed
o psychological de achmen om wo k and because psychological de achmen om wo k should be
nega i ely ela ed o exhaus ion, we expec a nega i e indi ec e ec om leade suppo o eco e y o
exhaus ion ia psychological de achmen om wo k.
Hypo hesis 2a. A he pe son le el, leade suppo o eco e y is nega i ely ela ed o
exhaus ion ia psychological de achmen om wo k.
The s a e o being eco e ed is “an indica o o success ul eco e y esul ing om ac i i ies o ex-
pe iences pu sued du ing p e ious leisu e ime” (Binnewies e al., 2009, p. 70). I esul s om he suc-
cess ul es o a ion o esou ces ha had been deple ed du ing wo k and exp esses i sel in a eeling o
being e eshed and ene ge ic. We p opose ha leade suppo o eco e y will ha e a se ial indi ec
e ec on an employee's a e age mo ning s a e o being eco e ed, ia psychological de achmen om
wo k and low exhaus ion. When employees success ully de ach om wo k and consequen ly expe ience
lowe exhaus ion le els, hey will bene i om he de achmen p ocess he nex mo ning as well. Taken
oge he , because leade suppo o eco e y should be posi i ely ela ed o psychological de achmen
om wo k, psychological de achmen om wo k should be nega i ely ela ed o exhaus ion, and ex-
haus ion, in u n, should be nega i ely ela ed o he mo ning s a e o being eco e ed, we expec a
posi i e se ial indi ec e ec om leade suppo o eco e y o he mo ning s a e o being eco e ed
ia psychological de achmen om wo k and exhaus ion.
Hypo hesis 2b. A he pe son le el, leade suppo o eco e y is posi i ely ela ed o
he mo ning s a e o being eco e ed ia psychological de achmen om wo k and bed ime
exhaus ion.
1768
|
SONNENTAG e al.
The mode a ing ole o leade - membe - exchange
Resea ch sugges s ha he ela ionship quali y be ween leade s and hei subo dina es has an impac on
how employees espond o hei leade 's beha iou s (G eenbaum e al., 2018; Sch iesheim e al., 1998).
We p opose ha he ela ionship be ween leade and employee (leade - membe exchange; LMX) mod-
e a es he associa ion be ween leade suppo o eco e y and employee psychological de achmen om
wo k du ing non- wo k ime.
LMX is a leade ship concep elying on co e p inciples o social exchange (Dulebohn e al., 2012).
Acco ding o he LMX app oach, leade s de elop dis inc exchange ela ionships wi h hei employ-
ees: Low- quali y ela ionships ocus on economic exchanges as p esc ibed in employmen con ac s.
High- quali y ela ionships go beyond hese basic economic exchanges and include he exchange o
a b oade se o esou ces, such as “deep mu ual a ec ion and admi a ion” (Liden e al., 1997, p.
84). A high- quali y ela ionship will no only ha e a posi i e impac on employees' job pe o mance
and sa is ac ion (Dulebohn e al., 2012; Ma in e al., 2016), bu i will also help leade suppo o
eco e y o un old i s ull po en ial. Employees in high- quali y LMX ela ionships us hei lead-
e s, implying ha hey ha e “con idence in ( he) cha ac e o ( he) leade ” (Di ks & Fe in, 2002,
p. 613). Applied o he con ex o eco e y, hey ha e con idence ha he leade is since e when
exp essing a posi i e iew on eco e y and when encou aging employees o ake ime o eco e y
and empo a ily o ge abou wo k. This con idence implies ha employees can allow hemsel es
o ake eco e y se iously and o men ally de ach om wo k when hey a e o he job. This us in
hei leade being ypical o employees in high- quali y LMX ela ionships le s employees “accep
ulne abili y” (Rousseau e al., 1998, p. 398) when demons a ing ha hey do no hink abou hei
job 24/7 because hey can us ha hei leade will accep – o e en welcome – his empo al de-
achmen om wo k.
The e o e, he suppo he leade exp esses o eco e y will be mo e in luen ial o he eco -
e y p ocesses o employees in high- quali y ela ionships. Employees in low- quali y ela ionships,
howe e , will no necessa ily us ha hei leade s ha e good in en ions when emphasizing he
impo ance o eco e y. Leade s' a i udes abou eco e y will be less ele an o hem because hey
eel mo e dis ance owa ds he leade . Consequen ly, o employees in low- quali y ela ionships,
hei leade 's suppo o eco e y will ha e less impac . Acco dingly, we sugges ha he quali y o
he LMX ela ionship is a mode a o in he ela ionship be ween leade suppo o eco e y and
psychological de achmen om wo k. In o he wo ds, he quali y o he LMX ela ionship needs o
be easonably high so ha employees' de achmen om wo k can bene i om leade suppo o
eco e y.
Hypo hesis 3. A he pe son le el, LMX mode a es he ela ionship be ween leade sup-
po o eco e y and psychological de achmen om wo k such ha he posi i e ela ion-
ship be ween leade suppo o eco e y and psychological de achmen b eaks down (i.e.,
becomes non- signi ican ) when LMX is low.
Linking Hypo heses 2 and 3, we p opose ha he indi ec ela ionship o leade suppo o eco e y
wi h exhaus ion and he mo ning s a e o being eco e ed ia psychological de achmen depends on
LMX. Employees in a high- quali y LMX ela ionship bene i om leade suppo o eco e y in he
sense ha hey show highe de achmen om wo k and, in u n, will expe ience lowe le els o exhaus-
ion and highe le els o being eco e ed in he mo ning. Because hey us in hei leade , hey will eel
encou aged ha i is psychologically sa e o de ach om wo k, and his highe le el o de achmen will
esul in mo e a ou able le els o exhaus ion and a s a e o being eco e ed. Employees in a low- quali y
LMX ela ionship, howe e , will no be able o ake ad an age o hei leade 's suppo o eco e y.
Consequen ly, hei le el o psychological de achmen om wo k will no be a ec ed by leade suppo
o eco e y. In u n, i will also be less likely ha leade suppo o eco e y is e lec ed in low exhaus-
ion and a high mo ning eco e y s a e. We p opose:
|
1769
LEADER SUPPORT FOR RECOVERY
Hypo hesis 4. A he pe son le el, we expec a condi ional indi ec e ec such ha he
indi ec e ec be ween leade suppo o eco e y and exhaus ion and he mo ning s a e
o being eco e ed ia psychological de achmen om wo k b eaks down (i.e., becomes
non- signi ican ) when LMX is low.
Day- le el p ocesses: psychological de achmen om wo k, bed ime
exhaus ion, and he mo ning s a e o being eco e ed
In addi ion o he pe son- le el p ocesses desc ibed in Hypo heses 1 o 4, eco e y p ocesses will be
obse ed a he day le el as well. S udies ha e shown ha psychological de achmen happening a
he day le el is associa ed wi h subsequen day- le el well- being ou comes (Liu e al., 2021; Sonnen ag
e al., 2008). We p opose ha day- le el psychological de achmen om wo k is ela ed o low bed ime
exhaus ion and, in u n, o a high mo ning s a e o being eco e ed.
As desc ibed abo e, when de aching om wo k du ing a e - hou s, s ain symp oms ha ha e eme ged
du ing he wo kday a e alle ia ed and exhaus ion dec eases. No de aching om wo k, howe e , means
o con inue hinking abou wo k- ela ed ma e s ha will u he d aw on a pe son's ene ge ic esou ces.
Consequen ly, exhaus ion le els will emain high. Indeed, day- le el s udies demons a ed nega i e associ-
a ions be ween psychological de achmen om wo k and exhaus ion (B edeho s e al., 2023) and simila
ou comes (Ge meys & De Gie e , 2018). We seek o eplica e hese indings in ou s udy:
Hypo hesis 5a. A he day le el, psychological de achmen om wo k is nega i ely e-
la ed o bed ime exhaus ion.
We p opose ha day- le el psychological de achmen om wo k should be associa ed wi h a
high mo ning s a e o being eco e ed h ough low bed ime exhaus ion. As a gued wi h espec o
Hypo hesis 5a, psychological de achmen om wo k should be nega i ely ela ed o bed ime exhaus-
ion. Low bed ime exhaus ion, in u n, should con ibu e o he eeling o being e eshed in he nex
mo ning because i implies ha one inishes he e ening in a less deple ed s a e ha mos p obably
ansla es in o a ela i ely ene ge ic s a e in he nex mo ning. In con as , high bed ime exhaus ion
educes he likelihood o being ully eco e ed he nex mo ning.
Hypo hesis 5b. A he day le el, psychological de achmen om wo k is posi i ely e-
la ed o he mo ning s a e o being eco e ed ia low bed ime exhaus ion.
OVERVIEW OF STUDIES
To assess leade suppo o eco e y, we ollowed p inciples o scale de elopmen and alida ion
(Colqui e al., 2019; Hinkin, 1998; Lambe & Newman, 2023). Speci ically, we de eloped measu es o
empa hy o eco e y, espec o bounda ies, and ole modelling wi hin a b oade esea ch p ojec on
leade s' ole o eco e y ha also aimed a examining leade s' ole o aking aca ions, encou aging
b eaks a wo k, and p o iding b eak in as uc u e a he wo kplace. We s a ed he scale- de elopmen
p ocess by elici ing leade s' b oade iew on employee eco e y in a quali a i e s udy (S udy 1). In
S udies 2a, 2b, 3a, and 3b, we de eloped and es ed he cons uc alidi y o he h ee speci ic scales
assessing empa hy o eco e y, espec o bounda ies, and ole modelling, using a con i ma o y ac o
analy ic app oach (Lambe & Newman, 2023). In S udy 4, we es ed con en alidi y. S udy 5 is ou
main s udy in which we es ed ou hypo heses.2
2This is he i s publica ion om S udies 1, 2a, 2b, 3a, 4, and 5. Resul s om S udy 3b ocusing on a di e en esea ch ques ion and using
di e en a iables ha e been al eady published (Sonnen ag & Schi ne , 2019).
1776
|
SONNENTAG e al.
de achmen om wo k, exhaus ion, and s a e o being eco e ed. Speci ically, we modelled he h ee
ac o s cap u ing leade suppo o eco e y as a highe - o de pe son- le el ac o and LMX as a
sepa a e pe son- le el ac o . A he day le el, we speci ied he wi hin- pe son componen s o psycho-
logical de achmen om wo k, exhaus ion, and s a e o being eco e ed. The i o his wo- le el
model was good, χ2 = 1233.671, d = 527,4 p < .001, CFI = .930, RMSEA = .036, SCF = .9872,
AIC = 32,286.745, BIC = 32,972.316. This model i he da a be e han al e na i e models such as
an al e na i e (non- nes ed) model in which all ou leade ship a iables loaded on one common
pe son- le el ac o , χ2 = 1444.395, d = 533, p < .001, CFI = .910, RMSEA = .040, AIC = 32,486.251,
BIC = 33,142.015, and an al e na i e model in which all a iables assessed in he daily su eys loaded
on one common ac o a he day and he pe son le el, χ2 = 6034.064, d = 537,5 p < .001, CFI = .458,
RMSEA = .098, SCF = .9796, AIC = 36,959.940, BIC = 37,595.832, Sa o a- Ben le Δχ2 = 8104.388,
Δd = 10, p < .001.
Analy ic s a egy
Because ou da a had a mul i- le el s uc u e (i.e., days nes ed wi hin pe sons) and because we wan ed o
make ull use o he a ailable da a (Newman, 2014), we analysed ou da a wi h a wo- le el pa h model
wi h Bayesian es ima ion, using Blimp 3.0.63 ha allows o model- based impu a ion o missing da a
(Ende s e al., 2020).
We es ed Hypo heses 1 o 4 in he be ween- pe son pa o he model (i.e., a he pe son le el) and
Hypo heses 5a and 5b in he wi hin- pe son pa o he model (i.e., a he day le el). Because he high-
o de ac o model o leade suppo o eco e y i he da a easonably well, we es ed ou hypo heses
(1) wi h an o e all measu e o leade suppo o eco e y, o which we a e aged sco es on i s h ee
aspec s (empa hy, espec o bounda ies, and ole modelling) weigh ed by he espec i e ac o loadings
and (2) sepa a ely o he h ee aspec s o leade suppo o eco e y.
To adequa ely es he day- le el media ion p oposed in Hypo heses 5a and 5b, we modelled he
ela ionships be ween psychological de achmen , e ening exhaus ion, and he mo ning s a e o
being eco e ed a he pe son and day le el (P eache e al., 2010). A he pe son le el, we used man-
i es pe son means o he a iables assessed a he day le el. A he day le el, we included he day o
da a collec ion as a con ol a iable because pa icipa ing in he s udy migh ha e an impac on e-
sponden s' e e yday beha iou (Gab iel e al., 2019). Fo his pu pose, we added a con inuous a i-
able wi h s udy day coded om 1 o 10 o he analysis. When p edic ing bed ime exhaus ion and
mo ning s a e o being eco e ed a he day (i.e., wi hin- pe son) le el, we con olled o bed ime
exhaus ion and he mo ning s a e o being eco e ed, espec i ely, om he p e ious day o da a
collec ion.6 A he day le el, we speci ied pa hs in ol ing he con ol a iables as ixed and all o he
pa hs as andom. Fo es ing indi ec e ec s, we elied on he p oduc o coe icien app oach
(MacKinnon e al., 2002).
Resul s
We s a ed ou hypo heses es s wi h a main- e ec model wi hou any in e ac ion e ms. As he uppe
pa o Table 2 shows, he o e all measu e o leade suppo o eco e y was un ela ed o psychological
de achmen om wo k. Consequen ly, he indi ec e ec s om he o e all measu e o leade suppo
o eco e y o exhaus ion ia psychological de achmen , median = −.041, SD = .032, 95% CI [−.121,
4We ixed wo nega i e non- signi ican esidual a iances o 0.
5We ixed he same esidual a iances o 0 as in he compa ison model.
6When excluding all con ol a iables om he analysis, esul s did no change.
|
1777
LEADER SUPPORT FOR RECOVERY
TABLE 2 Uns anda dized coe icien s om mul i- le el modelling including main e ec s only.
Psychological de achmen Bed ime exhaus ion S a e o being eco e ed
Median SD 2.5% 97.5% Median SD 2.5% 97.5% Median SD 2.5% 97.5%
Pe son le el
In e cep 3.559 .077 3.406 3.708 2.777 .070 2.642 2.916 3.255 .066 3.128 3.387
Suppo o eco e y (o e all) .272 .139 −.004 .544 .053 .123 −.187 .295 .131 .106 −.078 .335
LMX −.140 .135 −.401 .127 −.282 .117 −.413 −.051 .012 .102 −.188 .216
Psychological de achmen – – – – −.168 .075 −.317 −.024 .089 .065 −.041 .214
Bed ime exhaus ion – – – – – – – – −.467 .072 −.607 −.323
Residual a iance .819 .102 .649 1.050 .638 .078 .507 .810 .464 .057 .370 .591
Va iance explained by
coe icien s
3.5% 10.7% 31.0%
Day le el
In e cep 3.559 .088 3.382 3.729 2.753 .079 2.595 2.906 3.210 .078 3.054 3.361
Day o da a collec ion .002 .010 −.019 .022 −.000 .008 −.017 .016 .011 .009 −.007 .028
E ening exhaus ion (p e ious
day)
– – – – .043 .051 −.056 .144 – – – –
S a e o being eco e ed
(p e ious day)
– – – – – – – – .023 .042 −.059 .105
Psychological de achmen – – – – −.212 .041 −.298 −.136 .014 .049 −.084 .107
Bed ime exhaus ion – – – – – – – – −.141 .053 −.248 −.038
Residual a iance .638 .034 .576 .710 .402 .023 .358 .445 .426 .027 .377 .481
Va iance explained by
coe icien s
.0% 3.0% 1.4%
No e: All ela ionships we e es ed in one o e all model.
1778
|
SONNENTAG e al.
.003], and he mo ning s a e o being eco e ed ia psychological de achmen and exhaus ion, me-
dian = .019, SD = .016, 95% CI [−.001, .059], we e no signi ican .
When compu ing models o he h ee sepa a e aspec s o leade suppo o eco e y,7 we ound
ha espec o bounda ies was posi i ely ela ed o psychological de achmen om wo k, me-
dian = .203, SD = .090, 95% CI [.025, .382]. The indi ec nega i e e ec om espec o bounda ies
o exhaus ion ia psychological de achmen was signi ican , median = −.033, SD = .023, 95% CI
[−.088, −.001]. The se ial indi ec posi i e e ec om espec o bounda ies o he mo ning s a e
o being eco e ed ia psychological de achmen and exhaus ion was signi ican , median = .015,
SD = .011, 95% CI [.0004, .0435]. In con as o espec o bounda ies, bo h empa hy and ole mod-
elling we e un ela ed o psychological de achmen om wo k, and he indi ec e ec s o empa hy
and ole modelling, espec i ely, on exhaus ion and he mo ning s a e o being eco e ed we e non-
signi ican . Taken oge he , we ound pa ial suppo o Hypo heses 1 and 2: Resul s o espec o
bounda ies we e in line wi h he hypo heses, bu esul s o empa hy, ole modelling, and he o e all
measu e we e no .
In he nex s ep, we included he in e ac ion e m be ween leade suppo o eco e y and LMX
in he model (Table 3). The in e ac ion e m be ween he o e all measu e o leade suppo o e-
co e y and LMX was a signi ican p edic o o psychological de achmen om wo k. Simple slope
analysis showed ha when LMX was high (+1 SD), leade suppo o eco e y was posi i ely e-
la ed o psychological de achmen om wo k, median = .488, SD = .170, 95% CI [.166, .826], bu
when LMX was low (−1 SD), leade suppo o eco e y was no , median = .079, SD = .163, 95%
CI [−.244, .398]. This pa e n o indings is in line wi h Hypo hesis 3. Figu e 2 (Panel a) shows he
simple slopes.
The indi ec nega i e e ec om leade suppo o eco e y o exhaus ion ia psychological
de achmen was signi ican when LMX was high (+1 SD), median = −.073, SD = .048, 95% CI
[−.188, −.003], bu non- signi ican when LMX was low (−1 SD), median = −.010, SD = .030, 95%
CI [−.080, .045]. The se ial indi ec posi i e e ec om leade suppo o eco e y o he mo ning
s a e o being eco e ed ia psychological de achmen and exhaus ion was signi ican o high LMX,
median = .034, SD = .023, 95% CI [.001, .091], and non- signi ican o low LMX, median = .004,
SD = .014, 95% CI [−.021, .039]. O e all, he signi ican condi ional indi ec e ec s p o ide suppo
o Hypo hesis 4.
In addi ion o his analysis elying on he o e all measu e o leade suppo o eco e y, we an
he mode a o analysis o he h ee sepa a e measu es as well.8 Tes ing Hypo hesis 3 in sepa a e
analyses showed signi ican mode a o e ec s o empa hy, median = .190, SD = .088, 95% CI [.017,
.363], and espec o bounda ies, median = .226, SD = .085, 95% CI [.061, .394]. The in e ac ion
e ec be ween ole modelling and LMX was no signi ican , median = .175, SD = .095, 95% CI
[−.016, .329].
Subsequen simple slope analysis showed ha when LMX was high (+1 SD), empa hy o eco e y
was posi i ely ela ed o psychological de achmen om wo k, median = .378, SD = .143, 95% CI
[.101, .664], bu no when LMX was low (−1 SD), median = .058, SD = .137, 95% CI [−.212, .329].
Simila ly, when LMX was high (+1 SD), espec o bounda ies was posi i ely ela ed o psycholog-
ical de achmen om wo k, median = .464, SD = .134, 95% CI [.205, .732], bu no when LMX was
low (−1 SD), median = .082, SD = .102, 95% CI [−.118, .281]. Panels b and c in Figu e 2 show he
simple slopes.
Tes s o condi ional indi ec e ec s o empa hy o eco e y we e no signi ican . Speci ically, he
indi ec nega i e e ec om empa hy o eco e y o exhaus ion ia psychological de achmen was no
signi ican , nei he o high (+1 SD), median = −.049, SD = .037, 95% CI [−.138, .003], no o low (−1
7Tables epo ing main e ec s o he h ee sepa a e aspec s o leade suppo o eco e y a e a ailable online in he Suppo ing In o ma ion
(Tables S4–S6).
8Tables epo ing he in e ac ion e ec o he h ee sepa a e aspec s o leade suppo o eco e y a e a ailable online in he Suppo ing
In o ma ion (Tables S7–S9).
|
1779
LEADER SUPPORT FOR RECOVERY
TABLE 3 Uns anda dized coe icien s om mul i- le el modelling including he in e ac ion e ec be ween leade suppo o eco e y (o e all) and LMX.
Psychological de achmen Bed ime exhaus ion S a e o being eco e ed
Median SD 2.5% 97.5% Median SD 2.5% 97.5% Median SD 2.5% 97.5%
Pe son le el
In e cep 3.452 .090 3.276 3.629 2.783 .083 2.620 2.946 3.276 .076 3.128 3.423
Suppo o eco e y (o e all)
(SUPREC)
.285 .140 .016 .558 .052 .124 −.189 .230 .133 .105 −.075 .338
LMX −.073 .137 −.341 .193 −.284 .118 −.517 −.056 .001 .104 −.199 .209
SUPREC × LMX .243 .107 .035 .454 −.015 .097 −.205 .175 −.035 .080 −.195 .120
Psychological de achmen – – – – −.162 .076 −.311 −.012 .086 .067 −.046 .218
Bed ime exhaus ion – – – – – – – – −.468 .072 −.611 −.327
Residual a iance .792 .097 .631 1.007 .642 .079 .510 .819 .465 .057 .372 .598
Va iance explained by coe icien s 7.9% 11.0% 31.1%
Day le el
In e cep 3.560 .089 3.385 3.734 2.763 .081 2.596 2.917 3.206 .078 3.056 3.362
Day o da a collec ion .002 .010 −.018 .022 .000 .008 −.016 .017 .011 .009 −.007 .029
E ening exhaus ion (p e ious day) – – – – .040 .049 −.054 .137 – – – –
S a e o being eco e ed (p e . day) – – – – – – – – .023 .042 −.058 .105
Psychological de achmen – – – – −.212 .039 −.292 −.139 .015 .048 −.081 .106
Bed ime exhaus ion – – – – – – – – −.139 .054 −.247 −.034
Residual a iance .639 .034 .577 .710 .401 .023 .359 .449 .426 .026 .377 .479
Va iance explained by coe icien s .0% 3.0% 1.3%
No e: All ela ionships we e es ed in one o e all model.
1780
|
SONNENTAG e al.
SD) LMX, median = −.006, SD = .023, 95% CI [−.061, .036]. The se ial indi ec nega i e e ec om
empa hy o eco e y o he mo ning s a e o being eco e ed ia psychological de achmen and exhaus-
ion was non- signi ican o bo h high, median = .022, SD = .017, 95% CI [−.002, .066] and low LMX,
median = .003, SD = .011, 95% CI [−.017, .029].
Tes s o condi ional indi ec e ec s o espec o bounda ies we e signi ican : The indi ec nega i e
e ec om espec o bounda ies o exhaus ion ia psychological de achmen was signi ican when
LMX was high (+1 SD), median = −.080, SD = .044, 95% CI [−.183, −.011], bu no when LMX was
low (−1 SD), median = −.013, SD = .021, 95% CI [−.063, .023]. The se ial indi ec nega i e e ec om
espec o bounda ies o he mo ning s a e o being eco e ed ia psychological de achmen and ex-
haus ion was posi i e and signi ican o high LMX, median = .037, SD = .022, 95% CI [.005, .089], and
non- signi ican o low LMX, median = .006, SD = .010, 95% CI [−.011, .031]. Fo ole modelling, none
o he condi ional indi ec e ec s was signi ican .
To sum up, ou analyses p o ide suppo o Hypo hesis 3 when examining empa hy o eco e y,
espec o bounda ies, and he o e all measu e o leade suppo o eco e y. Resul s o espec o
bounda ies and he o e all measu e we e in line wi h Hypo hesis 4. Fo ole modelling, howe e , nei he
Hypo hesis 3 no Hypo hesis 4 ecei ed empi ical suppo .
Rega ding he hypo heses speci ied o he day le el, analysis showed ha psychological de ach-
men om wo k was nega i ely ela ed o e ening exhaus ion, median = −.212, SD = .039, 95% CI
[−.292, −.139]. Mo eo e , e ening exhaus ion was nega i ely ela ed o he mo ning s a e o being
eco e ed, median = −.139, SD = .054, 95% CI [−.247, −.034], and he indi ec e ec om psycho-
logical de achmen o he mo ning s a e o being eco e ed ia exhaus ion was posi i e and signi -
ican , median = .029, SD = .013, 95% CI [.007, .057]. These indings a e in line wi h Hypo heses 5a
and 5b.
GENERAL DISCUSSION
In his pape , we examined he ole o leade suppo o eco e y o employees' psychological de ach-
men om wo k and downs eam exhaus ion and s a e o being eco e ed. Based on a se ies o scale-
de elopmen and scale- alida ion s udies, we p esen ed a sho h ee- dimensional measu e o leade
suppo o eco e y ha has good cons uc and con en alidi y. In ou main s udy (S udy 5), we ound
ha espec o bounda ies bu nei he o e all suppo o eco e y no empa hy o ole modelling
p edic ed employees' psychological de achmen om wo k. LMX mode a ed he ela ionship be ween
leade suppo o eco e y and psychological de achmen : Only when LMX was high, o e all leade
suppo o eco e y was posi i ely associa ed wi h psychological de achmen om wo k. When LMX
was low, howe e , leade suppo o eco e y was un ela ed o psychological de achmen . Looking a
FIGURE 2 In e ac ion e ec s.
|
1781
LEADER SUPPORT FOR RECOVERY
he di e en aspec s o leade suppo o eco e y, he same in e ac ion pa e n became e iden o
leade empa hy and espec o bounda ies, bu no o ole modelling.
Psychological de achmen om wo k p edic ed low exhaus ion and a high mo ning s a e o being
eco e ed, bo h a he pe son and day le el. The indi ec e ec o leade suppo o eco e y o ex-
haus ion ia psychological de achmen om wo k was signi ican when LMX was high bu no when
LMX was low. Simila ly, he indi ec e ec o leade suppo o eco e y o he mo ning s a e o being
eco e ed ia psychological de achmen and low exhaus ion was signi ican when LMX was high, bu
no when LMX was low.
Theo e ical and p ac ical implica ions
Ou esea ch o e s impo an heo e ical implica ions as i shows ha a high- quali y ela ionship be-
ween leade and ollowe (i.e., high LMX) is c ucial so ha leade suppo o eco e y can un old i s
posi i e po en ial o employee eco e y. A high le el o o e all leade suppo o eco e y alone was
un ela ed o employee de achmen om wo k du ing non- wo k ime. A high le el o LMX, howe e ,
p o ides he necessa y con ex wi hin which leade s' suppo o employee eco e y can become e ec-
i e. Wi hin high- quali y ela ionships, leade s' a i udes and communica ion abou eco e y a e mo e
likely o ac ually “ each” he employee and o be aken se iously, esul ing in highe le els o psychologi-
cal de achmen . Wi hin low- quali y ela ionships, howe e , employees discoun leade s' a i udes and
communica ion messages so ha leade s a e less in luen ial o employees' beha iou .
A close look a he h ee sub- dimensions o leade suppo o eco e y b ings mo e ligh in o he
ole o LMX o employee eco e y. We in oduced leade suppo o eco e y as a mul i- dimensional
concep comp ising empa hy o eco e y, espec o bounda ies, and ole modelling. Ou analyses
showed ha hese h ee sub- dimensions – al hough highly co ela ed – show dis inc esul pa e ns in
he p edic ion o employees' psychological de achmen . These dis inc pa e ns a e pa icula ly in o ma-
i e o unde s anding he ole o LMX o employee eco e y. Respec o bounda ies was a powe ul
p edic o o psychological de achmen , i espec i e o LMX. This inding implies ha all employees
bene i in e ms o men ally de aching om wo k when hei leade espec s bounda ies be ween wo k
and non- wo k ime and does no dis u b hei employees du ing non- wo k ime. Empa hy o eco -
e y, howe e , did no show a signi ican di ec associa ion wi h psychological de achmen om wo k,
implying ha – ac oss all employees – empa hy o eco e y does no help o de ach om wo k. The
signi ican in e ac ion e ec be ween LMX and empa hy o eco e y sugges s ha a leade 's empa hy
o eco e y is only e ec i e o employees who ha e a good ela ionship wi h hei leade . Only when
LMX is high, empa hy o eco e y can un old i s bene i s because employees eel emo ionally close o
hei leade s and can us leade s' good in en ions (Dulebohn e al., 2012). When LMX is low, howe e ,
empa hy o eco e y emains ine ec i e because when employees do no us hei leade s, employees
may ques ion i i is eally in hei own bes in e es o ake ime o eco e y and ully de ach om wo k.
Ano he possible explana ion o he in e ac ion e ec be ween LMX and o e all leade suppo o
eco e y can be de i ed om leade s' way o aming goals. Kakka (2019) sugges ed ha in low- quali y
LMX ela ionships, leade s ame goals in a manne ha p imes a p e en ion ocus among subo di-
na es, who hen ocus on hei obliga ions, expec a ions, and su i al in he wo kplace. This conce n by
employees in a low- quali y ela ionship may lead o o e - p eoccupa ion and o e - in es men in wo k,
which will esul in less de achmen and eco e y a e wo k hou s, e en when hei manage s e balize
suppo o eco e y. Recen ly, Volme e al. (2023) sugges ed ha high- quali y LMX will be associa ed
wi h s a e posi i e a ec a wo k ha in u n will be associa ed wi h highe le els o eco e y om wo k
du ing he e ening. This line o esea ch implies ha when manage s suppo eco e y, a high LMX
along wi h i s associa ed posi i e a ec may esul in highe le els o eco e y. This possibili y should
u he be explo ed in u u e s udies.
Wi h espec o he h ee sub- dimensions o leade suppo o eco e y, espec o bounda ies
ela ed di ec ly o psychological de achmen and showed he hypo hesized in e ac ion e ec wi h
1782
|
SONNENTAG e al.
LMX. As explained abo e, espec o bounda ies seems o be e ec i e o all employees, e en o
low- LMX employees. Respec o bounda ies migh be so impo an because no only leade s' objec-
i e ac ual in usions in o hei employees' non- wo k ime can ha m psychological de achmen , bu
jus employees expec ing ha leade s migh in ude impai psychological de achmen . The e o e,
e en e y a e ac ual in usions migh educe employees' psychological de achmen on e enings when
no in usion occu s.
Ou analysis u he showed ha empa hy o eco e y needs a high- LMX en i onmen o be e ec-
i e in s imula ing de achmen . In such an en i onmen , howe e , empa hy o eco e y is impo an
because i signals ha employees can ake he libe y o de ach om wo k. Finally, ole modelling
u ned ou o be ine ec i e o employee eco e y as i showed nei he a main e ec no a signi ican
in e ac ion e ec wi h LMX in p edic ing psychological de achmen . Al hough gene al ole modelling
o wo k- li e balance is a help ul leade ship beha iou wi h espec o wo k- amily con lic (Hamme
e al., 2009) and employee wo k- home segmen a ion beha iou (Koch & Binnewies, 2015), in ou e-
sea ch, ole modelling o leisu e and eco e y ac i i ies did no help wi h psychological de achmen ,
e en no when LMX was high. This inding sugges s ha a leade who shows ha hey ake ime o
leisu e and eco e y hemsel es does no help in employees' de achmen om wo k. Maybe, a leade 's
ole modelling encou ages employees' engagemen in leisu e ac i i ies, bu engagemen in leisu e ac i -
i ies may no ansla e in o psychological de achmen om wo k. Mo eo e , as he leade 's li e si ua ion
migh di e la gely om ha o hei employees (e.g., in e ms o inancial esou ces and leisu e in e -
es s), employees migh no iden i y su icien ly wi h hei leade wi h espec o eco e y p ocesses, and
his lack o iden i ica ion may make ole modelling ine ec i e (Mo gen o h e al., 2015).
Du ing he pas decade, esea che s inc easingly iden i ied app oaches ha leade s may use o os e
employee heal h and well- being, such as aluing heal h and being awa e o po en ial heal h p oblems
(F anke e al., 2014). Whe eas his line o esea ch has s a ed o examine how heal h- p omo ing lead-
e ship ela es o o he leade ship cons uc s (Kaluza e al., 2021), ou inding on he mode a o e ec
o LMX may help o be e unde s and he ci cums ances ha a e needed so ha heal h p omo ion by
leade s can be success ul. Ou indings can inspi e his b oade ield o heal h- p omo ing leade ship o
concep ualize speci ic heal h- p omo ing beha iou s, including leade suppo o eco e y.
We examined leade suppo o eco e y as a pe son- le el a iable. Acco dingly, ou indings imply
ha leade suppo o eco e y can un old i s bene i s – along wi h LMX – e en on days when leade s
and employees do no mee o when leade s do no exp ess hei suppo on a speci ic day. This esul is
impo an as i sugges s ha leade s do no need o show suppo o eco e y e e y day – he employees'
gene al pe cep ion ha hei leade is suppo i e o eco e y does su ice.
In e ms o p ac ical implica ions, ou esea ch showed ha employees' psychological de achmen
om wo k du ing non- wo k ime is mos likely when bo h o e all leade suppo o eco e y and LMX
a e high. Wi h espec o leade suppo o eco e y, pa icula ly espec o bounda ies is impo an .
Thus, leade s may inc ease he likelihood ha employees swi ch o om wo k du ing non- wo k ime
by s ic ly espec ing bounda ies be ween wo k and non- wo k li e. Fi s , leade s should e ain om
con ac ing employees du ing he e ening o on he weekend, o ins ance by a oiding phone calls o
e aining om ins an messaging. Second, leade s should alloca e asks and se deadlines in a way ha
employees do no need o spend a e - hou s on job asks. In some companies, espec ing bounda ies
may no be a he disc e ion o indi idual leade s bu migh equi e a change in company cul u e.
Acco dingly, op managemen should encou age manage s o espec bounda ies, implying ha also
leade s' bounda ies a e espec ed (Venz & Wöh mann, 2023).
Al hough leade empa hy o eco e y was no di ec ly ela ed o employees' psychological de ach-
men om wo k, i p edic ed psychological de achmen in in e ac ion wi h LMX. Thus, leade s should
be awa e o employees' eco e y needs and should be unde s anding when eco e y is needed. Simila
o gene al app oaches o heal h- o ien ed leade ship (F anke e al., 2014), leade s migh become mo e
awa e o employees' eco e y needs when hey a e mo e awa e o and esponsi e owa ds hei own
eco e y needs. Mo eo e , leade s migh be mo e empa he ic owa ds employees' eco e y needs when
hey cul i a e compassion wi hin hei wo k se ing (Du on e al., 2014).
|
1783
LEADER SUPPORT FOR RECOVERY
Ha ing a good ela ionship wi h one's leade (i.e., high LMX) is impo an o ake ad an age o
leade suppo o eco e y. Thus, bo h leade s and employees may wan o s i e o imp o e hei
exchange ela ionship, o ins ance by ocusing on simila i y (Dulebohn e al., 2012), using delega ion
(Baue & G een, 1996), and sha ing pe sonal expe iences (Ya e & Ka k, 2011). Mo eo e , o es ablish
a high- quali y LMX, leade s should p o ide especially new employees wi h guidance and ad ice su -
ounding hei jobs and hei oles wi hin he o ganiza ion (Sluss & Thompson, 2012).
In e ms o p ac ical use o measu es o assess leade suppo o eco e y, ou scales co e a b oade
cons uc space han does he Benne e al. (2016) measu e. Thus, when no only in e es ed in espec
o bounda ies bu also he leade 's emo ional unde pinning o beha iou s ha should p omo e em-
ployee eco e y, he assessmen o empa hy o eco e y is impo an – also because esul s o his as-
sessmen could in o m aining in e en ions o leade s. Such in e en ions migh no only emphasize
espec o bounda ies bu inco po a e a module ha a ge s leade empa hy. Rega ding he espec -
o - bounda ies dimension, he Benne e al. (2016) measu e emphasizes leade s' iews on highly pe -
o ming employees and associa ed expec a ions, whe eas ou measu e – in addi ion o i s assessmen
o empa hy and ole modelling – mo e di ec ly add esses leade s' beha iou o no in uding in o em-
ployees' pe sonal li es. The e o e, i could be seen as a measu e ha is concep ually close o leade
beha iou ha delibe a ely suppo s employee eco e y.
Limi a ions and di ec ions o u u e esea ch
Despi e he new insigh s i p o ides, ou esea ch has some limi a ions. Fi s , we assessed ou a iables
wi h sel - epo ed measu es p o ided by employees. Ob iously, his migh aise conce ns abou com-
mon me hod bias (Podsako e al., 2003). Al hough we canno ully ule ou common me hod bias, ou
speci ic s udy design, including bo h pe son- le el and day- le el measu es, educes i . We assessed leade
suppo o eco e y and LMX wi h pe son- le el measu es in he en ance su ey and all o he a i-
ables wi h day- le el measu es, which we agg ega ed o he pe son le el o pe son- le el analyses. This
app oach implies ha he measu es o leade suppo o eco e y and LMX on he one hand and meas-
u es o he emaining a iables on he o he hand we e assessed a di e en ime poin s and wi h di e -
en empo al e e en s, making an o e es ima ion o he empi ical associa ion less likely. Mo eo e , one
o ou co e indings e e s o he mode a o e ec o LMX – a esul pa e n ha is less suscep ible o
common me hod bias (Siemsen e al., 2010).
Second, ela ed o measu emen issues as well, we measu ed leade suppo o eco e y om he
employees' pe spec i e, omi ing he leade s' pe spec i e. As o en leade s and employees do no ag ee
abou leade ship beha iou (Lee & Ca pen e , 2018), i migh be ha leade s ha e a iew on hei sup-
po o eco e y ha di e s om employees' pe spec i e. Howe e , we included leade s in S udy 1 and
used leade s' epo s o c i ical eco e y si ua ions o de eloping ou i ems. In a nex s ep, i would be
in e es ing o use ou scales o assess leade s' iews on how hey suppo eco e y – al hough employees'
pe cep ion o leade suppo is p obably he mo e ele an ac o in d i ing employee eco e y.
Thi d, ou esea ch ocused on he di ec ela ion be ween leade suppo o eco e y and em-
ployee de achmen om wo k, neglec ing possible unde lying pa hways. Thus, i emains unclea how
an employee's pe cep ion ha hei leade suppo s eco e y ansla es in o psychological de achmen
om wo k. In addi ion o explici communica ion abou he need o eco e , a leade migh shape low
a ailabili y and esponse expec a ions o a e - wo k hou s ha make i easie o employees o o ge
abou wo k when being a home (Venz & Wöh mann, 2023). In addi ion, leade s who suppo eco e y
possibly ins il a eco e y clima e wi hin hei eams so ha no only he leade endo ses eco e y du ing
non- wo k ime, bu ha eam membe s mu ually ein o ce a posi i e iew on eco e y. Fu u e s udies
should add ess he mechanisms by which leade suppo o eco e y helps de achmen om wo k.
Al hough ole modelling o eco e y was no a signi ican p edic o o employee de achmen om
wo k in ou esea ch, u u e s udies may wan o con inue o add ess his in e es ing ace o leade
beha iou . Fo ins ance, i has been a gued ha leade s' engagemen in se ious leisu e in luences leade
1784
|
SONNENTAG e al.
pe o mance (Bunea e al., 2023). Knowing ha one's leade pu sues a se ious in e es beyond wo k
may inspi e employees o explo e a simila op ion o hemsel es, which, in u n, should help hem o
de ach om wo k. In addi ion, i migh be wo hwhile o explo e si ua ional and pe sonal ci cums ances
unde which leade s' eco e y ole modelling does play a ole o employees' psychological de achmen
om wo k.
We ocused on how leade suppo o eco e y ela es o employee psychological de achmen om
wo k and downs eam exhaus ion and s a e o being eco e ed. Fu u e esea ch migh wan o examine
how leade suppo o eco e y a ec s leade s' own eco e y. Al hough an emphasis on eco e y migh
also bene i leade s, less posi i e scena ios a e possible as well. Fo ins ance, by p o ec ing hei employ-
ees om o e wo k and ensu ing ha employees ge enough eco e y, no all asks migh ge comple ed
and, consequen ly, leade s migh co e up o any un inished asks, which, in u n, educes leade s' own
eco e y (Sy ek e al., 2017). Simila ly, when espec ing employees' bounda ies, wo k p ocesses migh
become mo e e o ul and cumbe some om he leade s' pe spec i e.
CONCLUSION
Wi h e e - inc easing demands on employees, i is impo an ha employees de ach om wo k and ge
enough eco e y du ing non- wo k ime so ha hey can s a hei wo kdays e eshed. Ou esea ch
shows ha leade s can suppo psychological de achmen om wo k, pa icula ly by showing empa hy
o eco e y and espec ing bounda ies. Impo an ly, leade suppo o eco e y wo ks bes in high
LMX ela ionships. O e all, his esea ch con ibu es o he accumula ing e idence ha leade s play a
co e ole in p o ec ing employee heal h and well- being.
AUTHOR CONTRIBUTIONS
Sabine Sonnen ag: Concep ualiza ion; unding acquisi ion; w i ing – o iginal d a ; o mal analysis;
da a cu a ion; me hodology. Roni Ka k: Concep ualiza ion; unding acquisi ion; w i ing – e iew and
edi ing; me hodology; o mal analysis; da a cu a ion; me hodology. Lau a Venz: In es iga ion; w i ing
– e iew and edi ing; da a cu a ion; me hodology.
ACKNOWLEDGEMENTS
This esea ch was suppo ed by a SHRM Founda ion g an o he i s wo au ho s. This g an is g a e-
ully acknowledged. We hank Tami Rubin, Ya den Keinan, and Ca a ina Schi ne o hei help in
collec ing and coding he da a. Open Access unding enabled and o ganized by P ojek DEAL.
FUNDING INFORMATION
This s udy was unded by a g an om he SHRM Founda ion. Howe e , he in e p e a ions, conclu-
sions and ecommenda ions a e hose o he au ho s and do no necessa ily ep esen he iews o he
SHRM Founda ion.
CONFLICT OF INTEREST STATEMENT
The au ho s do no ha e any con lic o in e es .
DATA AVAILABILITY STATEMENT
The da a ha suppo he indings o his s udy a e a ailable on eques om he co esponding au ho .
The da a a e no publicly a ailable due o p i acy o e hical es ic ions.
ORCID
Sabine Sonnen ag h ps://o cid.o g/0000-0002-9464-4653
Roni Ka k h ps://o cid.o g/0000-0003-0506-9683
Lau a Venz h ps://o cid.o g/0000-0003-0642-988X
|
1785
LEADER SUPPORT FOR RECOVERY
REFERENCES
Ande son, J. C., & Ge bing, D. W. (1991). P edic ing he pe o mance o measu es in a con i ma o y ac o analysis wi h a p e es as-
sessmen o hei subs an i e alidi ies. Jou nal o Applied Psycholog y, 76(5), 732–740. h ps:// doi. o g/ 10. 1037/ 0021- 9010. 76.5. 732
Anisman- Razin, M., Ka k, R., & Ash o h, B. E. (2023). “Doing dis ance”: The ole o manage s' enac men o psychological
dis ance in leade – ollowe ela ionships. G oup & O ganiza ion Managemen , 105960112311582. h ps:// doi. o g/ 10. 1177/
10596 01123 1158264
Bandu a, A. (1977). Social lea ning heo y. P en ice Hall.
Ba nes, C. M., Luciane i, L., Bha e, D. P., & Ch is ian, M. S. (2015). You wouldn' like me when I'm sleepy': Leade s' sleep,
daily abusi e supe ision, and wo k uni engagemen . Academy o Managemen Jou nal, 58(5), 1419–1437. h ps:// doi. o g/ 10.
5465/ amj. 2013. 1063
Baue , T. N., & G een, S. G. (1996). De elopmen o leade - membe exchange: A longi udinal es . Academy o Managemen Jou nal,
39(6), 1538–1567. h ps:// doi. o g/ 10. 5465/ 257068
Benne , A. A., Bakke , A. B., & Field, J. G. (2018). Reco e y om wo k- ela ed e o : A me a- analysis. Jou nal o O ganiza ional
Beha io , 39(3), 262–275. h ps:// doi. o g/ 10. 1002/ job. 2217
Benne , A. A., Gab iel, A. S., Calde wood, C., Dahling, J. J., & T ougakos, J. P. (2016). Be e oge he ? Examining
p o iles o employee eco e y expe iences. Jou nal o Applied Psychology, 101(12), 1635–1654. h ps:// doi. o g/ 10. 1037/
apl00 00157
Binnewies, C., Sonnen ag, S., & Mojza, E. J. (2009). Daily pe o mance a wo k: Feeling eco e ed in he mo ning as a p edic o
o day- le el job pe o mance. Jou nal o O ganiza ional Beha io , 30(1), 67–93. h ps:// doi. o g/ 10. 1002/ job. 541
B edeho s , J., K au e , K., Meu is, J., & Jachimowicz, J. M. (2023). The challenge o main aining passion o wo k o e ime: A daily
pe spec i e on passion and emo ional exhaus ion. O ganiza ion Science, 35, 364–386. h ps:// doi. o g/ 10. 1287/ o sc. 2023. 1673
B osscho , J. F., Ge in, W., & Thaye , J. F. (2006). The pe se e a i e cogni ion hypo hesis: A e iew o wo y, p olonged s ess-
ela ed ac i a ion, and heal h. Jou nal o Psychosoma ic Resea ch, 60(2), 113–124. h ps:// doi. o g/ 10. 1016/j. jpsyc ho es. 2005.
06. 074
B own, M. E., & T e iño, L. K. (2014). Do ole models ma e ? An in es iga ion o ole modeling as an an eceden o pe cei ed
e hical leade ship. Jou nal o Business E hics, 122(4), 587–598. h ps:// doi. o g/ 10. 1007/ s1055 1- 013- 1769- 0
Bunea, E., Ka k, R., & Hammond, M. (2023). “Leisu eship”: Impac o pu suing se ious leisu e on leade s' pe o mance. Human
Resou ce Managemen Re iew, 33(2), 100950. h ps:// doi. o g/ 10. 1016/j. h m . 2022. 100950
Bu ch, G. F., Benne , A. A., Humph ey, R. H., Ba chelo , J. H., & Cai o, A. H. (2016). Un a eling he complexi ies o empa hy
esea ch: A mul i- le el model o empa hy in o ganiza ions. In Emo ions and o ganiza ional go e nance (Vol. 12, pp. 169–189).
Eme ald G oup Publishing Limi ed.
Bu s, M. M., Becke , W. J., & Boswell, W. R. (2015). Ho bu ons and ime sinks: The e ec s o elec onic communica ion
du ing nonwo k ime on emo ions and wo k- nonwo k con lic . Academy o Managemen Jou nal, 58(3), 763–788. h ps:// doi .
o g/ 10. 5465/ amj. 2014. 0170
Cla k, M. A., Robe son, M. M., & Young, S. (2019). “I eel you pain”: A c i ical e iew o o ganiza ional esea ch on empa hy.
Jou nal o O ganiza ional Beha io , 40(2), 166–192. h ps:// doi. o g/ 10. 1002/ job. 2348
Clin on, M. E., Conway, N., & S u ges, J. (2017). I 's ough hanging- up a call': The ela ionships be ween calling and wo k hou s,
psychological de achmen , sleep quali y, and mo ning igo . Jou nal o Occupa ional Heal h Psycholog y, 22(1), 28–39. h ps://
doi. o g/ 10. 1037/ ocp00 00025
Cobb, S. (1976). Social suppo as a mode a o o li e s ess. Psychosoma ic Medicine, 38(5), 300–314. h ps:// doi. o g/ 10. 1097/ 00006
842- 19760 9000- 00003
Cohen, S., & Wills, T. A. (1985). S ess, social suppo , and he bu e ing hypo hesis. Psychological Bulle in, 98(2), 310–357. h ps://
doi. o g/ 10. 1037/ 0033- 2909. 98.2. 310
Colqui , J. A., Sabey, T. B., Rodell, J. B., & Hill, E. T. (2019). Con en alida ion guidelines: E alua ion c i e ia o de ini ional
co espondence and de ini ional dis inc i eness. Jou nal o Applied Psycholog y, 104(10), 1243–1265. h ps:// doi. o g/ 10. 1037/
apl00 00406
Di ks, K. T., & Fe in, D. L. (2002). T us in leade ship: Me a- analy ic indings and implica ions o esea ch and p ac ice.
Jou nal o Applied Psycholog y, 87(4), 611–628. h ps:// doi. o g/ 10. 1037// 0021- 9010. 87.4. 611
Dulebohn, J. H., Bomme , W. H., Liden, R. C., B oue , R. L., & Fe is, G. R. (2012). A me a- analysis o an eceden s and con-
sequences o leade - membe exchange: In eg a ing he pas wi h an eye owa d he u u e. Jou nal o Managemen , 38(6),
1715–1759. h ps:// doi. o g/ 10. 1177/ 01492 06311 415280
Du on, J. E., Wo kman, K. M., & Ha din, A. E. (2014). Compassion a wo k. The Annual Re iew o O ganiza ional Psycholog y and
O ganiza ional Beha io , 1(1), 277–304. h ps:// doi. o g/ 10. 1146/ annu e - o gps ych- 03141 3- 091221
Ende s, C. K., Du, H., & Kelle , B. T. (2020). A model- based impu a ion p ocedu e o mul ile el eg ession models wi h
andom coe icien s, in e ac ion e ec s, and nonlinea e ms. Psychological Me hods, 25(1), 88–112. h ps:// doi. o g/ 10. 1037/
me 00 00228
Flanagan, J. C. (1954). The c i ical inciden echnique. Psychological Bulle in, 51(4), 327–358. h ps:// doi. o g/ 10. 1037/ h0061470
F anke, F., Fel e, J., & Pund , A. (2014). The impac o heal h- o ien ed leade ship on ollowe heal h: De elopmen and es o
a new ins umen measu ing heal h- p omo ing leade ship. Ge man Jou nal o Human Resou ce Managemen , 28(1–2), 139–161.
h ps:// doi. o g/ 10. 1177/ 23970 02214 0280010
