Crafting for Health: A Longitudinal Study of Job and Off-Job Crafting Changes during the COVID-19 Pandemic

Abstract

We examined the association between changes in employee job and off-job crafting and their self-rated health during the COVID-19 pandemic. Further, we investigated how these associations differed across sample subgroups, contrasting differences in work location, living situation, and contractual changes (short-time work). We used four waves of a longitudinal dataset surveying NTotal = 783 German-speaking employees from Germany, Switzerland, and Austria from 2019 to 2021. We applied latent change score modeling and multigroup analyses to investigate our research questions. Results indicated that the mean job and off-job crafting and self-rated health trajectories remained relatively stable. However, we observed significant interindividual variance in job and off-job crafting changes. We found a consistent small positive relationship between crafting changes in both life domains over time, indicating that employees tended to change their crafting efforts similarly across domains. Additionally, job crafting increases between Waves 1 and 2 were linked to higher subsequent self-rated health at Wave 2, and similarly, off-job crafting increases between Waves 3 and 4 were linked to higher self-rated health at Wave 4. We observed only minor differences in this pattern across subgroups. Our results show how adaptive changes in crafting are linked to broader interindividual health differences and help identify groups who are not able to increase crafting during crises and thus could benefit from targeted support. Crafting can be an effective individual strategy for maintaining health, complementing organizational and public health measures. We encourage future research to incorporate temporal and contextual phenomena into crafting research.

Full text

Vol.:(0123456789)
Occupational Health Science
https://doi.org/10.1007/s41542-025-00222-5
ORIGINAL RESEARCH ARTICLE
Crafting forHealth: ALongitudinal Study ofJob andOff‑Job
Crafting Changes duringtheCOVID‑19 Pandemic
AnjaIsabelMorstatt1,2 · GeorgF.Bauer2 · JessicadeBloom3,4 ·
ZacharyJ.Roman5,6 · MartinTušl2 · PhilippKerksieck2
Received: 23 February 2024 / Revised: 22 January 2025 / Accepted: 24 January 2025
© The Author(s) 2025
Abstract
We examined the association between changes in employee job and off-job crafting
and their self-rated health during the COVID-19 pandemic. Further, we investigated
how these associations differed across sample subgroups, contrasting differences
in work location, living situation, and contractual changes (short-time work). We
used four waves of a longitudinal dataset surveying NTotal = 783 German-speaking
employees from Germany, Switzerland, and Austria from 2019 to 2021. We applied
latent change score modeling and multigroup analyses to investigate our research
questions. Results indicated that the mean job and off-job crafting and self-rated
health trajectories remained relatively stable. However, we observed significant
interindividual variance in job and off-job crafting changes. We found a consist-
ent small positive relationship between crafting changes in both life domains over
time, indicating that employees tended to change their crafting efforts similarly
across domains. Additionally, job crafting increases between Waves 1 and 2 were
linked to higher subsequent self-rated health at Wave 2, and similarly, off-job craft-
ing increases between Waves 3 and 4 were linked to higher self-rated health at Wave
4. We observed only minor differences in this pattern across subgroups. Our results
show how adaptive changes in crafting are linked to broader interindividual health
differences and help identify groups who are not able to increase crafting during
crises and thus could benefit from targeted support. Crafting can be an effective
individual strategy for maintaining health, complementing organizational and public
health measures. We encourage future research to incorporate temporal and contex-
tual phenomena into crafting research.
Keywords Latent change score models· Temporal patterns· Proactive strategies·
Individual job redesign· Leisure· DRAMMA needs
Extended author information available on the last page of the article
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Introduction
In March 2020, the World Health Organization (WHO) declared that the COVID-19
pandemic is global (WHO, 2021). The COVID-19 pandemic has since been termed the
“largest unplanned field experiments ever in the world of work” (Weigelt etal., 2021,
p. 181), and various public health measures were introduced, removed, and even rein-
troduced to contain the spreading of the virus (Hale et al., 2021). These public health
measures, such as forced teleworking, social isolation through contact restrictions, and
school and recreational site closures(OECD, 2020a, 2020b; Cotofan etal., 2021; Rudolph
etal., 2021), changed daily life in many ways. Some of the changes emerging during the
pandemic may even be permanent (Rudolph etal., 2021; Weigelt etal., 2021), possibly
expediting developments into the future of work (Ng etal., 2021; Sinclair etal., 2020).
For employees, these developments imply more flexibility, autonomy, and the need for
proactive actions, such as crafting, to navigate the complexity of their work/life situation
(de Bloom etal., 2020). By crafting within the job and outside the job domain, employees
proactively and on their own initiative address “certain [life domain characteristics] (…)
to better align their [experiences] with their skills, abilities and preferences” (Tims etal.,
2021). Within the job domain, job crafting refers to optimizing demands and resources as
outlined within the Job Demands-Resources (JD-R) model (Bakker & Demerouti, 2007;
Bakker etal., 2023). For example, employees might craft their job to reduce demands they
perceive as hindering, e.g., breaking down complex tasks or they might leverage social
resources by actively asking colleagues for feedback or structural resources by acquir-
ing new relevant skills (Tims etal., 2012). Outside the job, employees may be proactive
as well, which a recent stream of research describes as “needs-based off-job crafting”
(Kujanpää etal., 2022) based on the identity-based integrative needs model of crafting
(de Bloom etal., 2020). Off-job crafting refers to the proactive satisfaction of six psy-
chological needs according to the DRAMMA model (detachment, relaxation, autonomy,
mastery, meaning, and affiliation; D. B. Newman etal., 2014). An individual, therefore,
might craft off the job by seeking detachment and relaxation through reading a novel after
work, engaging in challenging hobbies to experience mastery, or increasing feelings of
affiliation by initiating quality time with loved ones. We highlight two conceptualizations
of crafting, noting that proactive efforts vary across life domains. However, we propose
examining them together, as both share the same core idea, and off-job crafting can also
be viewed through a JD-R perspective.1
1 While earlier research has investigated crafting outside the job domain, often framed within the JD-R
model resources (e.g., as home or leisure crafting; see: Demerouti etal., 2020; Haun etal., 2022; Petrou
etal., 2017; Petrou & Bakker, 2016), we focus on needs-based off-job crafting. Unlike job crafting, which
often involves concrete efforts like asking for feedback (Tims etal., 2012), off-job crafting efforts vary
widely, such as in different leisure activities. Despite these differences in crafting efforts, individuals may still
aim to satisfy underlying psychological needs (de Bloom etal., 2020; Kujanpää etal., 2022). Additionally,
from an approach-avoidance perspective (Bruning & Campion, 2018), needs-based off-job crafting aligns
with JD-R concepts: detachment efforts correspond to avoidance of demands, while crafting for mastery or
autonomy can be viewed as approach strategies aimed at increasing resources. This study combines needs-
based off-job crafting with the JD-R framework to explore crafting across life domains, drawing on the most
inclusive perspectives available. However, we acknowledge that this is still an emerging research field, with
initial efforts to increase the inclusivity of job crafting research through frameworks like the DRAMMA-
based conceptualization (Tušl etal., 2024). To further understand how crafting across life domains connects
from a theoretical perspective, more research that falls outside the scope of this study is needed.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Crafting strategies help employees stay healthy and well on and off the job
(Kujanpää etal., 2022; Rudolph etal., 2017). Job crafting (Tims & Bakker, 2010;
Wrzesniewski & Dutton, 2001) is associated with increased work engagement or
reduced job strain (Rudolph etal., 2017), and off-job crafting is linked with better
satisfaction of psychological needs, including recovery from work (Kujanpää etal.,
2021, 2022) and mental well-being (Tušl etal., 2022). However, with one exception
(Tušl etal., 2022), these studies predate the pandemic, and few examine changes in
crafting (Farrell, 2019; van Wingerden etal., 2017). Crafting changes likely became
more pronounced during the pandemic due to dramatic shifts in public health meas-
ures. These changes may have heightened the importance of situational factors over
stable behavioral tendencies (see Situational Strength Theory; Meyer etal., 2010,
2020),, prompting employees to adjust their crafting efforts to face new challenges
and prioritize their health across different life domains.
It remains largely unclear how crafting developed in the life domains throughout
the pandemic and how these changes are related to health, which has been a gen-
eral concern during the crisis. Most studies on job and off-job crafting conducted
during the pandemic focused on beneficial outcomes of crafting not on dynamic
changes in the crafting efforts themselves (Abdel Hadi etal., 2021; Behzadnia &
FatahModares, 2020; Brauchli etal., 2022; Pijpker etal., 2022). However, it is likely
that during different phases of the pandemic, individuals increased, decreased, or
retained their level of crafting in both or one life domain. We still lack an under-
standing of how crafting changed during the pandemic, but also whether certain
phases or contextual correlates have been associated with strong adaptive responses,
emerging as sharp increases or decreases in crafting, and/or with risks for impaired
health. We aim to bridge this gap by a) studying the changes in both job and off-job
crafting throughout the pandemic, b) reviewing how changes in the extent of craft-
ing are associated with employee health, and c) whether there are meaningful differ-
ences in these changes and associations with health between employee subgroups,
characterized e.g. by their work location during the pandemic.
Investigating crafting during the COVID-19 pandemic is important for four rea-
sons. First, this study expands the literature by examining crafting in both job and
off-job domains, offering a comprehensive view of proactive behaviors. In the job
domain, organizations helped implement public health measures like forced tel-
eworking and counterbalancing efforts, for instance, enabling colleagues to maintain
virtual contact (OECD, 2020b). Larger organizations may have been slow to iden-
tify and implement changes to help employees adapt, while smaller organizations
often lacked the resources to provide necessary support during physical distancing
(OECD, 2020b). With or without organizational support, employees likely engaged
in both job and off-job crafting their experiences when believing this necessary
and beneficial, supporting and maintaining their health. Understanding how craft-
ing evolved in these interconnected domains over time enhances our insight into its
impact on employee well-being.
Second, we know little about the changes in and patterns of crafting through-
out the pandemic, and we believe our study is the first to address this gap. Changes
in crafting might not have occurred linearly over the entire course of the pandemic
but rather episodic, responding to significant events like lockdowns. A change in
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
crafting might reflect an adaptive response of individuals to externally altered work
and living situations due to the pandemic (see also propositions stated by Demer-
outi and Bakker (2022)). Our methodological approach of latent change score mod-
els captures these changes between time points and allows us to examine overall
trends and variances (interpersonal differences), in our sample. Specifically, our
study, assesses the directions and magnitudes of changes in regard to the frequency
of crafting.
Third, we aim to investigate potential contextual correlates in addition to tempo-
ral patterns in crafting changes, based on various characteristics such as their living
situation or working location (office vs. home office). For instance, of those moving
to home offices at the beginning of the pandemic, earlier studies reported both posi-
tive and negative side effects of the pandemic (Kaltiainen & Hakanen, 2022; OECD,
2020a, 2020b; Tušl etal., 2021), e.g., more leisure time due to less commuting. Such
developments may have enabled employees to leverage positive aspects, for instance,
more leisure to invest in quality time with a partner/family. On the other hand, other
employees may have needed to focus crafting efforts on a more severely affected life
domain, for example, in case of increased workload. Either way, we suggest such
differences potentially emerge as different patterns in crafting changes throughout
the pandemic. This will advance our understanding of how a combination of tempo-
ral and contextual aspects are associated with crafting across life domains.
Lastly, we add to the research by investigating associations of changes in craft-
ing with employee health, extending the findings by assessing multiple time points
during the pandemic. By combining how intraindividual changes in crafting relate
to subsequent interindividual differences in health, our study provides insights into
how adaptive efforts during the pandemic might translate into broader differences
in health across individuals. Further, we address potential temporal and contextual
differences in the association between crafting and self-rated health. By this, we
advance our understanding of whether adaptive efforts have been beneficial dur-
ing certain periods of the pandemic and whether specific individuals benefitted
more (or less) from such adjustments during this crisis. In conclusion, we provide
insights for future guidelines and interventions to promote crafting and to ascertain
which employee groups likely will experience less severe disruptions in future crisis
situations.
Background
Changes inCrafting During thePandemic
During the pandemic, employees needed to balance different aspects of their lives
to ensure positive experiences. Challenges during the first and second lockdowns
differed; the second lockdown let people apply proven strategies. A study in Austria
found that during the second lockdown, people were better able to focus on work or
daily activities, feeling less busy than during the first lockdown (Łaszewska etal.,
2021). However, they also reported feeling less connected to others and experienc-
ing more negative impacts on family life, leisure, and education (Łaszewska etal.,
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
2021), hence a need to craft their off-job experiences. Therefore, one aim of this
study is to investigate how employees’ job and off-job crafting changed considering
the different phases of the pandemic.
Moreover, we investigate whether employees changed their crafting in both life
domains similarly or focused – potentially temporarily – on one specific life domain.
Employees may compensate for the lack of crafting opportunities in one domain by
emphasizing crafting in the other (Petrou etal., 2017), but may also craft equally
in both (Demerouti etal., 2020). Verelst etal. (2023) revealed associated potential
risks, including depleted energy levels. We examine how crafting in the job and off-
job domains changes over time based on contextual situations, possibly explaining
when employees allocate energy evenly between two domains or focus on one. A
(temporal) focus on one domain in crafting might, therefore, emerge as an increase
in crafting during a certain period, potentially even accompanied by a decrease in
crafting in the other life domain during the same period. Given the study’s time-
frame from June 2019 to December 2021, we can observe shifts in these foci, as
declines in one domain may be followed by increases in another (and vice versa).
However, crafting might also evolve jointly across life domains, following ideas
often coined as congruency or spillover2 across life domains (Guest, 2001; Petrou
& Bakker, 2016; Snir & Harpaz, 2002). As it remains unclear how crafting evolved
during the pandemic, we combine these ideas into our first research question:
RQ1: How did job and off-job crafting levels, and their interrelation, change over
the course of the pandemic?
Crafting Changes andHealth During thePandemic
The pandemic significantly altered employees’ lives and their perceptions of infec-
tion risk, both for themselves and others (Jimenez etal., 2023). Health and eco-
nomic anxiety rose for many (LeNoble etal., 2023). A study found that those who
perceived the pandemic impacts negatively reported poorer mental well-being
(Tušl etal., 2021), while another study noted minor declines in occupational well-
being, particularly among younger employees and those living alone (Kaltiainen &
Hakanen, 2022).
We focus on self-rated health, thereby capturing holistically but parsimoniously
how employees’ assessments of their health changed over time. Self-rated health
combines individual assessments of body and mind within a single rating, and meta-
analyses have shown its predictive value for mortality (DeSalvo etal., 2006; Idler
& Benyamini, 1997; Jylhä, 2009). Individuals report poorer self-rated health when
2 We note, however, that our study design would not allow us to examine actual spillover processes in
crafting between the life domains, as we focus on cross-sectional interrelations between crafting changes.
Other study designs, for example, cross-lagged panel designs (Zyphur etal., 2020) or continuous time
modelling (Ryan etal., 2018; van Montfort etal., 2018) would be more appropriate to uncover spillo-
ver processes and examine temporal precedencies in crafting. This study allows us to investigate only
whether crafting efforts changed congruently across life domains from a cross-sectional perspective.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
facing chronic conditions, including mental health problems, or low social support
(Amstadter etal., 2010; French etal., 2012; Van Lente etal., 2012). Overall, self-
rated health seems a relatively stable concept, as demonstrated before the pandemic
(Perruccio et al., 2010) and when comparing pre-pandemic with pandemic health
(Peters etal., 2020; Recchi etal., 2020; Szwarcwald etal., 2021; van de Weijer etal.,
2022). However, self-rated health remains responsive to varying health conditions,
e.g., during convalescence from complex surgeries (Perruccio etal., 2010). Based on
pre-pandemic studies linking crafting to health-promoting characteristics (e.g., work
engagement, Rudolph etal., 2017), we suggest that crafting during the COVID-19
pandemic will be positively associated with self-rated health.
Although not necessarily implicated in assessing the risks of physical infections,
crafting is pivotal in other mental health perceptions, e.g., of social support systems
(Holman etal., 2023; Tims etal., 2013). Pre-pandemic research indicates that job
crafting interventions enhance self-efficacy (van den Heuvel etal., 2015), which is
vital for health. Additionally, off-job crafting is beneficial for increasing sense of
meaning in life (Kujanpää etal., 2022; Petrou etal., 2017) and has been associated
with improved mental well-being during the pandemic (Tušl etal., 2022).
Studies during the pandemic have reiterated the protective role of crafting. For
instance, a simple intervention study utilizing WhatsApp messages prompted Ira-
nian students to increase their needs satisfaction, thereby reducing perceived stress
during this critical time (Behzadnia & FatahModares, 2020). In Germany, research-
ers found that employees proactively crafting their leisure time experienced less
emotional exhaustion, underscoring the importance of such proactive measures dur-
ing initial strict public health measures (Abdel Hadi etal., 2021). Additionally, stud-
ies investigating changes from pre-pandemic to pandemic states have demonstrated
the benefits of high levels of crafting for health. Employee groups identified as High
Job Crafters and High Off-job Crafters reported positive changes in self-rated health
before and after the pandemic (Brauchli etal., 2022). Similarly, those actively craft-
ing their off-job domain before the pandemic reported less burnout during those
challenging times (Pijpker etal., 2022). However, although emphasizing the positive
impacts of crafting, these studies usually refer to crafting levels and interpersonal
differences but fail to consider how adaptive efforts, e.g., increasing crafting, might
also be related to subsequent higher levels of health.
Our study aims to elaborate on the findings related to crafting changes during the
pandemic and their connection to self-rated health. In line with previous research
findings, we expect positive associations between a crafting change in one life
domain and the subsequent level of self-rated health. Job or off-job crafting increases
may enhance personal or social resources, leading to better health outcomes com-
pared to individuals who maintain or reduce their crafting efforts. Our study is the
first to examine crafting in both life domains jointly and during the pandemic. It
remains unclear whether crafting changes in both life domains are associated with
subsequent health to a similar extent or whether crafting changes focused on one life
domain played a more substantial role for maintaining health. Additionally, there
might have been changes to these patterns throughout the pandemic, such as shifts in
domain relevance for health from one pandemic phase to another. To allow for such
an open investigation, we pose the following research question:
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
RQ2: How are changes in job and off-job crafting associated with subsequent
levels of self-rated health?
Interindividual Differences inCrafting Changes During thePandemic
We explore interindividual differences by taking into account prior studies that
suggested certain demographic groups to have faced adverse outcomes to a vary-
ing extent during the pandemic. For instance, younger individuals living alone,
employees with reduced leisure time, or those with more caregiving duties report-
edly perceived a negative impact (Tušl etal., 2021), while job uncertainty increased
in the financial crisis following the pandemic (OECD, 2020a, 2020b). Positive side
effects emerged concurrently, including perceived improvements in both work and
private life (Tušl etal., 2021). Positive impacts on employees included increased lei-
sure time, time spent with partners and family, and reduced commuting (Tušl etal.,
2021). Moreover, in the early stages, teleworking employees experienced heightened
work engagement – a notable positive side effect of the transformed work landscape
(Kaltiainen & Hakanen, 2022).
Three major themes emerged and were further investigated: Firstly, employees’
physical work locations played a pivotal role, with widely differing experiences.
Those transitioning to remote work faced unique challenges and opportunities com-
pared to those continuing in physical workspaces. Secondly, living situations sig-
nificantly influenced the pandemic’s impact. Individuals living alone might have
faced reduced social interactions and their implications. Conversely, those living in
families found solace in strengthened familial bonds but also incurred increased car-
egiving duties. Lastly, contractual changes introduced further complexity. Employ-
ees with changing job structures, e.g., a forced contractual reduction of the working
hours, faced potential uncertainty in the job domain. In Germany, over eight million
employees were in a short-time work scheme in April 2020, the highest since 2009
(Bundesagentur für Arbeit, 2024).
Further, the contextual variables mentioned above may well have occasioned
changes in employees’ crafting throughout the pandemic to respond to specific
needs and opportunities. Certain potentially more vulnerable groups may have inten-
sified their crafting efforts (temporarily) to mitigate negative impacts. For instance,
employees transitioning to home offices possibly increased efforts to connect with
others virtually to combat social isolation, thereby increasing their job crafting in
early phases of the pandemic. Some may have also found benefits through increas-
ing crafting in either life domain, like mastering remote work or greater virtual
closeness with loved ones. Additionally, specific groups may have focused on one
domain, particularly those facing job insecurity, intensifying job crafting to secure
enjoyment and stability amidst uncertainties. We formulate our third research ques-
tion accordingly:
RQ3: How are subgroup differences in work location, living situation, and con-
tractual changes related to changes in job and off-job crafting and their interrela-
tion?
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Finally, we assume in general that increases in crafting should be positively asso-
ciated with subsequent levels of self-rated health. However, as outlined referring to
RQ2, temporal or domain-specific deviations might form noteworthy patterns in the
associations between crafting changes and health. Expanding this idea, we might
also observe noteworthy patterns linked with contextual factors. For example, it is
possible that certain groups’ crafting changes were not beneficial in ensuring high
levels of health during the pandemic. Identifying such groups might advance our
understanding of potential risk groups in crises that cannot support their own health
through increasing their crafting in the job and/or off-job domain. We investigate
this along with our fourth research question:
RQ4: How are subgroup differences in work location, living situation, and con-
tractual changes related to the associations between crafting changes and subse-
quent self-rated health?
Methods
Study Design, Sample, andProcedure
We used four survey waves from a larger longitudinal online panel study inviting
German-speaking employees from Germany, Switzerland, and Austria to participate
repeatedly (panel provider: bilendi, formerly respondi). We chose these four survey
waves to divide the pandemic into four phases: pre-pandemic (Wave 1, June/July
2019), post-onset and first lockdown (Wave 2, April 2020), second lockdown (Wave
3, December 2020), and incipient normalization phase (Wave 4, November/Decem-
ber 2021). To plot the data of the stringency index (scoring closure and containment
measures) against our survey waves in Fig.1, we referred to data from the Oxford
COVID-19 Government Response Tracker (Hale etal., 2021).
The extensive longitudinal panel study, in which ours is embedded, includes two prior
study waves in 2018 and 2019, not considered here. Generally, participants were invited
to participate in the following survey waves. However, to expand the set, we invited all
participants from previous waves to participate in our second survey wave. Since some
participants decided to rejoin the panel study, our total N is highest for Wave 2 (N = 783);
573 participants completed our survey in Wave 1, 500 in Wave 3, and 399 in Wave 4. For
all waves, we included participants currently working over 20h per week and excluded
self-employed individuals. We cleaned the data after collection to remove careless
responses, mainly speeders and straightliners (Meade & Craig, 2012).
We present demographics from Wave 2, the key survey wave. Table1 presents
all demographics for the full sample and all group comparisons.
We investigated attrition in our sample, comparing those with complete par-
ticipation (n = 300) against those with incomplete participation (n = 483, missing
at least one wave) regarding demographics (age, gender, caregiving duties, work-
ing hours) and the variables in our model (job crafting, off-job crafting, self-rated
health), all at Wave 2. We found mean differences for age MCompletes = 49.84years
vs. MIncompletes = 48.02 years (t(697) = 2.592, p = 0.009), gender MCompletes = 1.42
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
vs. MIncompletes = 1.49 (1 = “male”, 2 = “female”, 3 = “other”; t(643.89) = −1.986,
p = 0.047), and caregiving duties (number of persons cared for) MCompletes = 1.05
vs. MIncompletes = 1.30 years (t(681.37) = −2.582, p = 0.01). Effect sizes for these
comparisons remained below small effects (Age: d = 0.20, Gender: d = 0.16, Car-
egiving duties: d = 0.20).
Measures
For each wave, participants completed an online survey including the following
measures. We assessed all variables in German, anchored using a 5-point Likert
Fig. 1 Average stringency index for Germany, Austria, and Switzerland based on the Oxford COVID-19
Government Response Tracker. Note. We plot the average stringency index (0–100) based on the Oxford
COVID-19 Government Response Tracker (Hale etal., 2021). The stringency index indicates the sever-
ity of political closure and containment measures during the COVID-19 pandemic. After the COVID-19
pandemic was declared as such in March 2020 (WHO, 2021), governments all over the world placed
first and stern measures to contain the virus spreading, e.g., nationwide lockdowns paired with school
closures (Hale etal., 2021; Rudolph etal., 2021; Weigelt etal., 2021). Towards the summer of 2020,
incidence rates of COVID-19 infections decreased (Bundesministerium für Gesundheit, n.d.) and many
containment measures were then relaxed. However, in fall 2020, new virus variants increased the risks of
infection and mortality again (RKI, 2023; WHO, 2023), so most governments again placed strict contain-
ment measures (Hale etal., 2021). Since then, and with the development of effective vaccinations against
the virus (Mathieu etal., 2021), most political measures have been relaxed or even removed (Hale etal.,
2021). Additionally, we plot our survey waves (Wave 1: June/July 2019, Wave 2: April 2020, Wave 3:
December 2020, Wave 4: October/November 2021) within the graph to contextualize each survey time-
frame. The starting dates of each survey wave are indicated by black dots at the bottom of each graph and
a vertical gray line
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Table 1 Demographics of full sample and subgroups
Full sample Work location Living situation Contractual changes
(N = 783) FOW
(n = 407) HO exp
(n = 187) HO new
(n = 162) Alone
(n = 213) P/F
(n = 540) CC
(n = 187) NC
(n = 588)
Characteristic M SE M SE M SE M SE M SE M SE M SE M SE
Age 48.72 9.90 50.14 9.38 47.07 10.38 47.17 10.11 48.50 10.16 48.99 9.72 48.86 9.82 48.79 9.89
Gender (% female) 46.23% 51.35% 41.71% 37.04% 51.64% 43.34% 50.27% 44.39%
Living situation
Alone 27.20% 28.78% 23.50% 29.63% 100% 0% 24.06% 28.62%
With partner or family 68.97% 67.99% 74.86% 67.90% 0% 100% 72.19% 68.79%
With roommates 2.81% 3.22% 1.64% 2.47% 0% 0% 3.74% 2.59%
No. of persons needing
caregiving
1.20 1.33 1.18 1.28 1.4 1.52 1.04 1.21 0.38 0.55 1.54 1.4 1.39 1.33 1.15 1.33
Of that: No. of children
under 11years
0.19 0.51 0.14 0.44 0.25 0.57 0.27 0.60 0.02 0.17 0.26 0.57 0.25 0.55 0.17 0.49
Working hours per
weeka38.14 7.02 37.13 7.44 39.26 6.67 39.07 6.11 39.19 6.33 37.74 7.21 37.39 7.2 38.34 6.97
Job function
Apprentice, student,
intern
0.38% 0.50% 0.54% 0% 0% 0.37% 0% 0.51%
Employee without
supervising function
70.37% 77.92% 56.99% 69.75% 77.14% 68.40% 65.95% 72.14%
Employee with super-
vising function
24.14% 18.11% 36.02% 24.69% 20.95% 26.02% 28.65% 23.08%
Employee in C-Level
position
4.47% 3.47% 6.45% 5.56% 1.90% 5.20% 5.41% 4.27%
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Demographics were assessed at Wave 2, except for COVID-19-specific items, which were assessed only at Wave 3 as a voluntary additional survey at the end of the main
survey. We report the available sample sizes in this section. In Germany and Switzerland, parents with children under 12years enjoyed special protection during the
COVID-19 pandemic (CMS Law Tax Future, 2022; Deutscher Gewerkschaftsbund, 2022). For example, parents were entitled to additional sick days if needed (CMS Law
Tax Future, 2022; Deutscher Gewerkschaftsbund, 2022). In our data collection, we asked the age of the person needing caregiving offering categories, with “7–10years”
being closest to the threshold presented by political support for parents. Therefore, we set the cut-off to define the special caregiving duties for children at 11years.
FOW = Full office workers, HO exp. = Home office experienced, HO new = Home office new, P/F = Living with partner or family, CC = Contractual changes, NC = No con-
tractual changes
a Weekly working hours were assessed using categories from 0–9 to 49 + . We report the mean via the midpoints of the categories and 54.5 as the last category midpoint
Table 1 (continued)
Full sample Work location Living situation Contractual changes
(N = 783) FOW
(n = 407) HO exp
(n = 187) HO new
(n = 162) Alone
(n = 213) P/F
(n = 540) CC
(n = 187) NC
(n = 588)
Characteristic M SE M SE M SE M SE M SE M SE M SE M SE
Percentage home office
Before pandemic 8.50% 21.32 0 0 32.67 29.66 0 0 6.60 18.80 9.29 22.39 8.18 21.07 8.53 21.21
Since pandemic 34.28% 42.98 0 0 73.69 33.55 78.33 27.69 34.62 43.87 34.42 42.77 22.73 35.75 38.12 44.43
Temporary contractual
work reduction
23.88% 27.52% 20.86% 16.66% 21.13% 25% 100% 0%
COVID-19 specific
items (Wave 3) (n = 470) (n = 233) (n = 122) (n = 98) (n = 131) (n = 321) (n = 106) (n = 363)
Belonging to high-risk
group
28.51% 27.47% 25.41% 33.67% 22.90% 30.22% 28.30% 28.37%
Quarantined (Self) 5.74% 4.72% 5.74% 8.16% 6.87% 5.61% 5.66% 5.79%
Quarantined (Family) 14.26% 14.59% 11.48% 17.35% 5.34% 18.69% 15.09% 14.05%
Diagnosed COVID-19
(Self)
1.70% 2.15% 0% 3.06% 0.76% 2.18% 1.89% 1.65%
Diagnosed COVID-19
(Family)
6.60% 5.58% 6.56% 7.14% 2.29% 8.41% 8.49% 6.06%
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
scale from 1 = “strongly disagree” to 5 = “strongly agree” unless otherwise stated.
We present means, standard deviations, internal consistencies using Cronbach’s
alpha, and intercorrelations in Table 2. All demographic variables (age, gender,
weekly working hours, job function, sector, living situation, caregiving duties, con-
tract changes, and percentage working in home office) presented in Table 1 and
partly used to segment our sample for later group comparisons were used from the
assessment at wave 2.
We assessed four facets of job crafting (increasing structural job resources,
increasing social job resources, increasing challenging job demands, and reducing
hindering job demands) using a total of 17 items (five, five, three, and four items for
the facets) adapted from the Job Crafting Scale by Tims etal. (2012) and a version
used by Petrou etal. (2012). We referred to the items by Petrou etal. (2012) for the
facet “increasing challenging job demands” to address concerns regarding the com-
plex wording of the version by Tims etal. (2012) (Nielsen & Abildgaard, 2012). Par-
ticipants rated the items using a scale from 1 = “never” to 5 = “very often”. Example
items for each facet are: “I try to develop my capabilities” (increasing structural job
resources); “ I ask others for feedback on my job performance.” (increasing social
job resources); “I make sure that my work is mentally less intense.” (reducing hin-
dering job demands); “I ask for more tasks when I have finished my work.” (increas-
ing challenging job demands).
The six facets of off-job crafting3 (detachment, relaxation, autonomy, mastery,
meaning, and affiliation) were assessed using 18 items (three items per facet) from
the Needs-based Off-job Crafting Scale developed by Kujanpää etal. (2022). Partic-
ipants rated the items using a scale from 1 = “never” to 5 = “very often”. In the early
stages of the data collection (Waves 1 & 2), an additional residual option, “I don’t
know,” was presented to participants. We coded this residual option as a missing
value. We introduced each item with the prefix “Over the past month, …”. Exam-
ple items for each facet are: “I’ve made sure to detach from work-related thoughts
during off-job time.” (Detachment); “I’ve made sure to experience relaxation of my
3 We thank an anonymous reviewer for pointing out existing ambiguities regarding the distinction
between off-job crafting and recovery experiences. Recovery experiences are part of the six psychologi-
cal needs that relate to subjective well-being (D. B. Newman etal., 2014). While the measurement for
recovery experiences targets whether individuals experienced certain states, e.g., not having thought
about work, off-job crafting captures the proactive and intentional actions by individuals toward satis-
fying specific recovery needs (Kujanpää & Olafsen, 2024; Kujanpää etal., 2022; Sonnentag & Fritz,
2007). Consequently, there is a notable difference in how these constructs are typically assessed. For
example, experience of detachment is usually assessed as “I don’t think about work at all.” or “I distance
myself from my work.” (Sonnentag & Fritz, 2007), whereas off-job crafting for detachment is assessed as
“I’ve made sure to detach from work-related thoughts during off-job time.” or “I’ve arranged my off-job
time so that I distance myself from work-related tasks.” (Kujanpää etal., 2022). Within recovery experi-
ences, active aspects might also be integrated, as distancing oneself from work is an active form of effort.
At the time of conducting this research and writing this study, the presented measure for off-job crafting
is – in our opinion – the best available, as it most inclusively captures proactive efforts within the off-job
domain compared to earlier conceptualizations closely mirroring the JD-R model (e.g., home or leisure
crafting; Demerouti etal., 2020; Petrou etal., 2017). However, we conclude that further disentangling
recovery experiences and off-job crafting, is a fruitful research field that will aid our understanding of
recovery from work.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Table 2 Means, standard deviations, correlations, and internal consistencies for multiple indicator scales
We include gender and age assessed at W2. M = Mean, SD = Standard deviation, JC = Job crafting, OJC = Off-job crafting, SRH = Self-rated health, BRG = Belonging to
risk group. Means and standard deviations were obtained using mean scores for the measures. Cronbach’s alpha is reported in italics along the diagonal for constructs with
multiple indicators. * = p < 0.05, ** = p < 0.01, *** = p < 0.001
a 1 = male, 2 = female
Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1. Gendera1.46 0.50 –
2. Age 48.72 9.90 .02 –
3. JC W1 2.99 0.48 .00 -.17*** .84
4. JC W2 2.97 0.48 -.01 -.20*** .74*** .85
5. JC W3 2.94 0.51 -.01 -.15*** .74*** .79*** .86
6. JC W4 2.81 0.50 .04 -.17*** .73*** .77*** .77*** .86
7. OJC W1 3.78 0.60 .06 .11** .24*** .19*** .17*** .16** .91
8. OJC W2 4.07 5.93 .03 -.00 -.08* -.05 -.07 -.14** -.03 .92
9. OJC W3 3.57 0.63 -.02 .03 .23*** .27*** .33*** .30*** .59*** .01 .93
10. OJC W4 3.61 0.63 .03 .09 .22*** .20*** .24*** .24*** .59*** -.02 .66*** .93
11. SRH W1 3.59 0.79 .08 -.20*** .15*** .12** .17** .13* .21*** .03 .25*** .23*** –
12. SRH W2 3.67 0.77 .01 -.17*** .17*** .15*** .22*** .14** .18*** .05 .32*** .21*** .73*** –
13. SRH W3 3.71 0.76 .02 -.12** .08 .09* .20*** .12* .25*** -.00 .34*** .21*** .69*** .71*** –
14. SRH W4 3.67 0.78 .02 -.05 .07 .05 .12* .05 .26*** -.00 .31*** .33*** .65*** .67*** .68***
15. BRG W3 0.29 0.45 -.03 .40** -.04 -.03 -.10* -.04 .08 -.02 -.01 .09 -.30** -.30** -.30** -.20**
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
body and mind during off-job time.” (Relaxation); “I’ve organized my off-job activi-
ties so that I determine my own course of action.” (Autonomy); “I’ve organized my
off-job activities so that I put my skills, knowledge or abilities into action.” (Mas-
tery); “I’ve organized my off-job activities so that I achieve a sense of purpose in
what I am doing.” (Meaning); “I’ve made sure to experience close connections to
the people around me during off-job time.” (Affiliation).
Self-rated health was measured using a single item as suggested by the WHO
(1996). The item was “How would you rate your health in general?” with response
options from 1 = “very bad” to 5 = “very good”.
Lastly, we controlled for belonging to a high-risk group for severe COVID-19
infections. At Wave 6, we asked participants: “Do you belong to a high-risk group
for the coronavirus due to your age or an illness?” (0 = “no”, 1 = “yes”). We decided
to include this measure as a control variable, as both COVID-19 infection rates (at
Wave 6: 1.702%) and having been in quarantine (at Wave 6: 5.745%) only occurred
with a very low base rate in our sample.
Data Analysis Strategy
For data analysis, R (R Core Team, 2020) and the following packages were used:
tidyverse (Wickham etal., 2019), readr (Wickham & Hester, 2021), and lubridate
(Grolemund & Wickham, 2011) for data handling, ggplot2 (Wickham, n.d.) and
apaTables (Stanley, 2021) for plot and table creation, and psych (Revelle, 2021) and
lavaan (Rosseel, 2012) for data analysis. We openly provide additional material in
the Electronic Supplementary Materials 1–5 (e.g., overview of model fits and com-
parison tests for measurement invariance and latent growth curve testing, full model
results of final models, R syntax regarding additional analyses regarding multigroup
analyses based on caregiving duties).
First, we added categorical variables indicating our subgroups of interest to assess
the third research question to our dataset.4 We added three categorical variables rep-
resenting employees’ work location, living situation, and contractual changes based
4 We initially planned to review differences in job and off-job crafting changes for subgroups also refer-
ring to reported caregiving duties (Groups: No caregiving duties (n = 342), Caregiving duties for chil-
dren under 11 (n = 110), Caregiving duties for older children & adults (n = 331)). However, we could
not establish (metric) measurement invariance for job crafting for all three groups and only partial met-
ric measurement invariance when excluding the subgroup without caregiving duties. Possibly due to the
resulting reduced sample size, a group comparison with only two subgroups having caregiving duties
did not work, as the model did not converge properly due to Heywood cases. We explored other ways
of splitting the sample regarding caregiving duties, but the group without caregiving duties continued
to prove problematic. We suggest that the issue of measurement variance for job crafting should be fur-
ther investigated in crafting research. Job crafting may be understood differentially by employees depend-
ing on their other life obligations as seen here with caregiving duties. We note, however, that our study
focuses on the pandemic situation. Therefore, issues of measurement invariance for job crafting emerging
outside the pandemic context also merit investigation (see, e.g., Vignoli etal., 2023).
We decided to treat caregiving duties as a demographic distal to the other groups and report them in
Table1 with other demographics. For transparency, we included the respective R code for analysis in
ESM 5 (available online).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
on information collected at the beginning of the COVID-19 pandemic (Wave 2). The
first set of subgroup comparisons covered the work location. We defined three sub-
groups: Full office workers (N = 407), home office workers experienced (N = 187),
and home office workers inexperienced (N = 162). The second set of subgroup
comparisons covered the living situation. We formed two subgroups: Living alone
(N = 213) and living with partner/family (N = 540). The third set of subgroup com-
parisons covered the contractual changes. We formed two subgroups: no contrac-
tual changes (N = 588) and contractual changes with temporary work reduction up to
zero percent (N = 187).
Next, we noted missing data in our dataset due to participation dropout and the
selection of residual options (in off-job crafting). While missing values are common,
appropriate handling is required to prevent biased parameter estimates. We follow
the suggestions by Newman (2014) and employ full information maximum likeli-
hood methods for our analyses to handle both full missing survey participations and
partial missings on selected scales.
Before seeking answers to our research questions, we assessed important assump-
tions. When exploring differences over time and across groups, the main assump-
tion is measurement invariance (Putnick & Bornstein, 2016; Steinmetz etal., 2009;
Vandenberg & Lance, 2000). Measurement invariance refers to the stability of con-
structs examined regarding their structure, factor loadings, and intercepts across
time points and/or groups. With measurement invariance established, researchers
can assume that the understanding of a construct remains unchanged between time
points or groups, and comparisons between time points and groups become mean-
ingful. Therefore, we started our analyses with elaborate measurement invariance
testing for our multiple indicator constructs job and off-job crafting using item par-
cels along their respective facets. We referred to item parcels for job and off-job
crafting to reduce the parameters in the models (Little etal., 2002; Orcan, 2013).
We first assessed longitudinal measurement invariance for the full sample and all
individual subgroups for later comparison (e.g., workers staying in their offices dur-
ing the pandemic vs. workers moving to home offices for the first time). Next, we
assessed intergroup measurement invariance referring to Wave 2. For all tests, meas-
urement models using confirmatory factor analyses (CFA) were specified for the
different stages of measurement invariance (configural, metric, scalar). To evaluate
the models, we referred to recommended cut-off criteria: Root Mean Square Error
of Approximation (RMSEA < 0.06), Comparative Fit-Index (CFI close to 0.95),
Tucker-Lewis-Index (TLI close to 0.95), Standardized Root Mean Square Residual
(SRMR < 0.08) (Hu & Bentler, 1999). Additionally, model comparison tests were
performed to find the best-fitting model.
We specified a series of bivariate latent change score models (LCSM) to answer
our first and second research questions. First, we specified an LCSM for the full
sample, including self-rated health as an outcome of crafting changes. We followed,
adapted, and extended the modeling approach for the LCSM by Geiser (2020) and
Wiedemann etal. (2022).
To answer our third and fourth research questions, we again specified a series
of LCSMs, but to compare subgroups, we nested those within three multigroup
analyses (MGA). Per group comparisons of interest, we specified one MGA
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
incorporating the same LCSM for each subgroup. For all models, we controlled
for belonging to a high-risk group as a predictor of self-rated health. Addition-
ally, we compared models with and without equality constraints (e.g., restraining
covariances of job and off-job crafting changes to be equal over time and restrain-
ing change score intercepts to be equal across groups). We retained the equality
constraint if the model comparison test indicated that the data could be equally
well described by a more parsimonious (restrained) model. We next present the
final models. Overviews of model fits and comparison tests are given in ESM 1
and ESM 2.
Results
Measurement Invariance, Equality Constraints, andModel Fit Indices
For the full sample and for the multigroup analyses splitting the sample by work
location and living situation, we established partial scalar measurement invariance
(invariant structure, same factor loadings, partially same item intercept) both over
time and across groups for both job and off-job crafting. For the multigroup analy-
sis splitting the sample by contractual changes, we could only establish partial met-
ric and scalar measurement invariance for job crafting and full metric measurement
invariance for off-job crafting. This suggests the job crafting measure is not exactly
the same when comparing the group with contractual changes and the group with-
out contractual changes at the beginning of the COVID-19 pandemic, as indicated
by the need to release one factor loading from the parcel of “increasing social job
resources” to the overall job crafting factor. In response, we fixed as much of the
construct between groups as possible to ensure comparable results. However, allow-
ing for partial measurement invariance yields only an approximation and will be dis-
cussed later (Putnick & Bornstein, 2016).
We implemented and tested several equality constraints within our models. We
applied an equality constraint on the job and off-job crafting change score covar-
iances across time for the full sample. Given the unequal time intervals between
the measurements in our data, most parameters, e.g., regression parameters, are
impacted by this. Therefore, an equality constraint on the covariances across time
was the only comparison test that could be tested for the full sample. We applied
equality constraints across groups for the multigroup analyses to ensure we inter-
pret meaningful and not only spurious differences in parameters. Across groups, we
tested for the equality of the auto-regressive effects of self-rated health, the regres-
sion from our control variable to self-rated health, the regressions from job and off-
job crafting changes to self-rated health, the covariances between job and off-job
crafting change scores, and the change score intercept of both job and off-job craft-
ing. Our data indicated that the more parsimonious models with equality constraints
could, in most cases, be retained. We present the final model fit indices, all of which
indicate good model fit, in Table3. We provide an overview of model fits and com-
parison tests for measurement invariance and equality constraints in our supplemen-
tary material ESM 1-4.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Latent Change Score Models
To answer our RQs 1–4, we next present the results from our LCSM analyses for
the full sample and the multigroup analyses (Comparison by work location: HO
new = new in home office, FOW = Full office workers, HO exp = experienced in
home office; comparison by living situation: alone, P/F = Living with partner/family;
comparison by contractual changes: NC = No contractual changes, CC = Contractual
changes). RQ1 referred to the crafting changes over time and their interrelations;
RQ2 considered temporal differences in the (positive) associations between craft-
ing changes and subsequent levels of self-rated health. As RQ3 and RQ4 can be
considered enhancements of RQ1 and RQ2 that explore differences in the patterns
depending on contextual variables, we will present these results in combination with
RQ1 and RQ2.
For an overview, we present the estimated latent trajectories of job and off-job
crafting as well as self-rated health in Fig.2, a conceptual summary of the latent
change score model associations in Fig.3, and a summary of the most relevant
model estimates and conclusions in Table4.5 As our analyses are inherently explora-
tory, we highlight several features of the model results related to our research ques-
tions (Latent score intercepts, variances, and covariances, as well as the latent
regression coefficients).
RQ1 andRQ3: Changes inJob andOff‑Job Crafting andTheir Interrelations
We examine changes in job and off-job crafting across time by referring to the
means of latent change scores. Significant latent change score means (results see
Table4) represent a common trend of change for the sample and are visible as a star
in between two time points in Fig.2. For the full sample, we observe a decreasing
trend between Waves 1 and 2 both for job crafting and off-job crafting. Between
Waves 2 and 3, job crafting increased in the full sample, but off-job crafting fur-
ther decreased. Finally, between Waves 3 and 4, we find a further decrease for job
crafting, but no significant mean trend for off-job crafting. Further, latent change
score variances indicate the extent of interindividual variance in crafting changes.
Our results show overall significant variance in changes between all time points,
given that the variances of job crafting range between ϕ JC CS2 = 0.02, SE = 0.004,
p < 0.001 (Waves 2 to 3), and ϕJC CS3 = 0.029, SE = 0.006, p < 0.001 (Waves 3 to
4), and variances of off-job crafting range between ϕOJC CS3 = 0.137, SE = 0.017,
p < 0.001 (Waves 3 to 4), and ϕOJC CS1 = 0.174, SE = 0.019, p < 0.001 (Waves 1 to 2).
In our multigroup analyses, most equality constraints on the latent change score
means (results in Table4) could be retained, indicating that the groups differ little
from each other in their pattern of change. Job crafting showed the same pattern for
all subgroups depending on the work locations. For off-job crafting, the pattern of
5 Full model results are available in the supplementary materials ESM 3. Please note, that we overall
adhere to the SEM coefficient notation as described by Bollen (1989). In ESM 3, we included the nota-
tion for all reported coefficients.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Table 3 Model fit indices for latent change score models
LCSM Latent change score model, SRH Self-rated health. For the full sample model, we retained the equality constraint on the covariances between job and off-job craft-
ing change scores across time. For the model investigating the differences in work location, we only rejected equality constraints on the job crafting change score intercepts
at Wave 4 and the off-job crafting change score intercepts at Waves 1, 2, and 4. For the model investigating the differences in living situation, we only rejected equality
constraints on the regression from our control variable to self-rated health at Wave 4, the regression from off-job crafting change score to self-rated health at Wave 4, the
job crafting change score intercepts at Wave 4, and the off-job crafting change score intercepts at Waves 3 and 4. For the model investigating the differences of contract
changes, we only rejected equality constraints on the covariances of job and off-job crafting changes between Waves 1 and 2, the regression from off-job crafting change
score to self-rated health at Wave 3, the job crafting change score intercepts at Wave 4, and the off-job crafting change score intercepts at Waves 3 and 4. * p < .05 **
p < .01 *** p < .001
Modell χ2df CFI TLI RMSEA SRMR
Full sample analyses
Baseline bivariate LCSM with SRH 2004.200*** 872 0.928 0.918 0.041 0.076
Multigroup analyses
Bivariate LCSM with SRH for group comparison of work location 4895.312*** 2698 0.864 0.850 0.057 0.093
Bivariate LCSM with SRH for group comparison of living situation 3357.756*** 1784 0.900 0.889 0.048 0.086
Bivariate LCSM with SRH for group comparison of contract changes 3226.126*** 1764 0.909 0.898 0.046 0.085
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
change differed between the subgroups with different work locations. For those new
to home office, we observed a decrease in off-job crafting between Waves 1 and 2
and no further mean trends at later time intervals. For those full-time in the office,
we observed decreases in off-job crafting between Waves 1 and 2 and between
Waves 2 and 3, but no trend between Waves 3 and 4. For those experienced in home
office, we observed no mean trends between Waves 1 and 2 and between Waves 3
and 4, but we find a decrease of off-job crafting between Waves 2 and 3. We again
observed significant variance of changes between all time points for all subgroups.
For job crafting, the variances ranged ϕJC CS2 = 0.016, SE = 0.006, p = 0.006 (Waves
2 to 3, HO exp), and ϕJC CS3 = 0.036, SE = 0.008, p < 0.001 (Waves 3 to 4, FOW).
For off-job crafting, variances ranged ϕOJC CS3 = 0.078, SE = 0.018, p < 0.001 (Waves
3 to 4, HO exp), and ϕOJC CS1 = 0.233, SE = 0.041, p < 0.001 (Waves 1 to 2, HO new).
For those living alone or with partner/family, we observed the same patterns
of change for job crafting and off-job crafting. Further, we again observed signifi-
cant variance of changes between all time points for all subgroups. For job craft-
ing, the variances ranged ϕJC CS2 = 0.02, SE = 0.006, p = 0.001 (Waves 2 to 3, alone),
Fig. 2 Job and off-job crafting and self-rated health trajectories. Note. HO new = new in home office,
FOW = Full office workers, HO exp = experienced in home office, P/F = Living with partner/family,
NC = No contractual changes, CC = Contractual changes. The sample was split according to reported
demographics at Wave 2. Sample sizes: N HO new/FOW/HO exp/HO new = 162 / 407 / 187, N P/F/alone = 540 / 213,
N NC/CC = 588 / 187. Mean trajectories across time for job and off-job crafting as well as self-rated health.
The values are generated from the latent change score models. Stars between two survey time points
represent a significant mean change as indicated by a significant intercept of the respective change score.
Self-rated health was modeled as an autoregressive model; no information about significant changes is
available
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
and ϕJC CS3 = 0.032, SE = 0.007, p < 0.001 (Waves 3 to 4, P/F). For off-job crafting,
variances ranged ϕOJC CS3 = 0.101, SE = 0.016, p < 0.001 (Waves 3 to 4, P/F), and
ϕOJC CS2 = 0.235, SE = 0.038, p < 0.001 (Waves 2 to 3, alone).
Comparing those with and without contractual changes, we observed the same
change patterns for job crafting and off-job crafting. Further, we again observed
significant variance of changes between all time points for all subgroups. For job
crafting, the variances ranged ϕJC CS2 = 0.079, SE = 0.014, p < 0.001 (Waves 2 to 3,
NC), and ϕJC CS3 = 0.17, SE = 0.04, p < 0.001 (Waves 3 to 4, CC). For off-job craft-
ing, variances ranged ϕOJC CS3 = 0.132, SE = 0.018, p < 0.001 (Waves 3 to 4, NC),
and ϕOJC CS2 = 0.219, SE = 0.038, p < 0.001 (Waves 2 to 3, CC).
Interestingly, we observed decreasing mean trends for both job and off-job craft-
ing, except for the slight increase in job crafting between Waves 2 and 3 for the full
sample. This pattern did not re-emerge in any of our subgroups. However, between
Waves 2 and 3, mean trends for job crafting in the subgroups do not reach the signif-
icance level of α = 0.05. All variances in job crafting change scores were significant
within this time interval, pointing to heterogeneity in crafting trends, which remains
unexplained by the demographic variables. Lastly, we note that visually (see Fig.2),
all job and off-job crafting trajectories remain relatively flat during the pandemic as
the magnitudes of crafting changes were relatively small.
Lastly, within RQ1 and RQ3, we also considered the associations between job
and off-job crafting changes over time. To answer these research questions, we
referred to the cross-sectional covariances of job and off-job crafting changes and
further tested whether equality constraints across time would hold as an indicator of
Fig. 3 Graphical Summary of LCSM results. Note. W = Wave. We present a graphical summary of the
model results focusing on the crafting changes and their relations with health over time. Crafting changes
were modeled using latent change score modelling (Geiser, 2020; Wiedemann etal., 2022); however the
underlying latent variables for crafting per time point are omitted in this figure to reduce complexity.
This summary considers the full sample model as well as the group comparisons to highlight similarities
and differences. We omit numerical model estimates and focus on the conceptual level. Non-dotted and
non-bolded lines represent significant paths within our models without differences across group compari-
sons. Dotted lines represent non-significant paths within our models. Bolded lines represent differences
in this model estimate in different group comparisons. A detailed outline of these changes is presented in
the results sections and in the electronic supplementary materials
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Table 4 Summary of research questions, related model estimates, and conclusions
Full sample Work location Living situation Contractual changes
HO new FOW HO exp P/F Alone NC CC
Research
question Parameter EST (SE) EST (SE) EST (SE) EST (SE) EST (SE) EST (SE) EST (SE) EST (SE)
RQ1 & RQ3aMean trends of change: Latent change score intercepts
JC W1-2 −0.044(0.01)*** −0.046(0.01)*** −0.043(0.01)*** −0.032(0.025)
JC W2-3 0.022(0.01)* 0.011(0.009) 0.016(0.01) −0.002(0.026)
JC W3-4 −0.075(0.013)*** −0.065(0.013)*** −0.064(0.012)*** −0.147(0.03)***
OJC W1-2 −0.047(0.019)* −0.147(0.048)** −0.074(0.025)** 0.05(0.037) −0.06(0.019)** −0.066(0.019)***
OJC W2-3 −0.095(0.02)*** 0.002(0.04) −0.097(0.026)*** −0.141(0.036)*** −0.094(0.02)*** −0.092(0.02)***
OJC W3-4 0.028(0.021) 0.036(0.021) 0.037(0.02) 0.03(0.021)
Interrelations: Concurrent covariances of latent change scores
W1-2 0.008(0.003)** 0.01(0.003)** 0.007(0.003)* 0.025(0.006)*** −0.018(0.015)
W2-3 0.008(0.003)** 0.01(0.003)** 0.007(0.003)* 0.025(0.006)***
W3-4 0.008(0.003)** 0.01(0.003)** 0.007(0.003)* 0.025(0.006)***
Conclusion We observed small decreases across the four waves in both JC and OJC, with one exception: JC between Waves 2 and 3, which showed little evidence of change. There were almost no dif-
ferences between groups. During the pandemic, JC and OJC tended to increase or decrease at the same time, except for employees with contractual changes (CC) at the beginning of the
pandemic (W1-2)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
a In the results section, we also present the model estimates for latent change score variances but omit these here to keep the summary concise
HO new new in home office, FOW Full office workers, HO exp experienced in home office, P/F Living with partner/family, NC No contractual changes, CC contractual
changes, JC Job crafting, OJC Off-job crafting, SRH Self-rated health, BRG Belonging to risk group. In case of no differences between groups in the group comparisons,
we only report one estimate per block. For the full model results, we kindly refer interested readers to ESM 3. * p < .05 ** p < .01 *** p < .001
Table 4 (continued)
Full sample Work location Living situation Contractual changes
HO new FOW HO exp P/F Alone NC CC
RQ2 & RQ4 Structural regression estimates crafting change scores to subsequent levels of self-rated health
Pre-pandemic to pandemic onset
JC W1-2 to
SRH W2 0.497(0.203)* 0.419(0.202)* 0.452(0.207)* 0.245(0.095)**
OJC W1-2 to
SRH W2 0.087(0.06) 0.107(0.062) 0.115(0.06) 0.095(0.061)
BRG W3 to
SRH W2 −0.049(0.068) −0.056(0.07) −0.058(0.067) −0.05(0.069)
Pandemic onset to second lockdown
JC W2-3 to
SRH W3 0.468(0.251) 0.666(0.276)* 0.411(0.254) 0.197(0.117)
OJC W2-3 to
SRH W3 0.114(0.062) 0.102(0.064) 0.113(0.06) 0.161(0.074)* −0.002(0.1)
BRG W3 to
SRH W3 0(0.064) 0.029(0.066) −0.023(0.065) 0.006(0.065)
Second lockdown to normalization phase
JC W3-4 to
SRH W4 −0.026(0.237) −0.099(0.232) 0.098(0.241) 0.001(0.108)
OJC W3-4 to
SRH W4 0.236(0.088)** 0.238(0.093)* 0.507(0.132)*** 0.054(0.144) 0.209(0.087)*
BRG W3 to
SRH W4 0.123(0.075) 0.113(0.077) 0.012(0.083) 0.413(0.14)** 0.103(0.074)
Conclusion At the start of the pandemic (W1-2), increases in JC were linked to better subsequent self-rated health. Toward the end of the pandemic (W3-4), such positive associations emerged instead
between increases in OJC and subsequent self-rated health. For individuals living alone, however, increases in OJC between W3-4 were not associated with subsequent self-rated health
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
a non-changing relationship between the changes. For the full sample and all sub-
groups, we could restrict covariances of crafting change scores to be equal across
time and groups (exception: Waves 1 to 2 for the group comparison of contractual
changes). The raw estimates of the covariance are included in Table4. The only
difference observed was for the group with contractual changes between Waves 1
and 2. In this case, the covariance between job and off-job crafting changes was
insignificant (ϕJC CS1, OJC CS1 = −0.018, SE = 0.015, p = 0.229). Standardizing these
estimates lets us examine the strength of these associations, as they then rep-
resent correlations. Given that we allowed the variances to be estimated freely,
the standardized estimates of covariances varied between all time points and
groups. However, we note that no standardized estimate was greater than 0.30
(range from rJC CS1, OJC CS1 = 0.103, CI = [0.018; 0.189] (Waves 1 to 2, alone) to
rJC CS2, OJC CS2 = 0.218, CI = [0.105; 0.331] (Waves 2 to 3, NC)), which would be
the threshold to consider an effect of medium size (Rosenthal, 1996). Therefore, all
covariances indicated a weak positive association between job and off-job crafting
changes.
RQ2 andRQ4: Associations Between Crafting Changes andSelf‑Rated Health
Lastly, we reviewed the associations between crafting changes and self-rated health.
To answer these research questions, we refer to the regression estimates from craft-
ing changes to subsequent self-rated health (results in Table4). A summary of these
associations and patterns are graphically presented in Fig. 3. In our models, we
controlled for a potential association between self-reported belonging to a COVID-
19 high-risk group (assessed only at Wave 3) and self-rated health (not included
in Fig.3). Interestingly, this relationship only reached significance for those living
alone at Wave 4.
For the full sample, we observed that positive job crafting changes between
Waves 1 and 2 positively predicted self-rated health at Wave 2, but at no other time
point. For off-job crafting, we observed the inverse temporal pattern. Positive off-job
crafting changes between Waves 1 and 2 and Waves 2 to 3 were not associated with
self-rated health at Waves 2 and 3, respectively. However, positive changes in off-job
crafting between Waves 3 and 4 positively predicted self-rated health at Wave 4.
Throughout the multigroup analyses, we observed similar patterns. First, compar-
ing groups by work location revealed no differences in associations between crafting
changes or the control variable with self-rated health. We observed that job crafting
changes between Waves 1 and 2 and Waves 2 and 3 positively predicted subsequent
self-rated health. Further, we observed that off-job crafting changes between Waves
3 and 4 positively predicted self-rated health at Wave 4.
Comparing those living alone to those living with partner/family, we observed in
both groups that job crafting changes between Waves 1 and 2 positively predicted
self-rated health at Wave 2. However, the pattern regarding off-job crafting changes
found in the full sample only re-emerged for those living with partner/family. Here,
we observed that off-job crafting changes between Waves 3 and 4 positively pre-
dicted self-rated health at Wave 4, but we found no such link for those living alone.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Finally, comparing subgroups with and without contractual changes, the overall
pattern between job and off-job crafting with self-rated health re-emerged, with only
a minor difference between the groups. For both groups, we observed that job craft-
ing changes between Waves 1 and 2 and off-job crafting changes between Waves 3
and 4 positively predicted subsequent self-rated health. For those without contrac-
tual changes, we additionally observed that off-job crafting changes between Waves
2 and 3 positively predicted self-rated health at Wave 3, but not for those with con-
tractual changes.
Overall, the standardized estimates for all reported significant relation-
ships remained below 0.30 (range: β JC CS1–SRH2 = 0.093 (Wave 2, HO new) to β
OJC CS3–SRH4 = 0.242 (Wave 4, P/F)), indicating that we only found small positive pre-
dictions of crafting changes regarding self-rated health.
Discussion
Changes inJob andOff‑Job Crafting: Mean Trends andVariances ofChanges,
andtheRelationship Between Job andOff‑Job Crafting Over Time
In our study, we sought to consider non-linear changes in crafting within certain
phases of the pandemic, such as from a pre-pandemic phase to the first lockdown
(RQ1). We investigated whether strong adaptive responses, e.g. sharp increases or
decreases in crafting, might be linked with certain temporal or contextual charac-
teristics (RQ3). Our results show that crafting remained on average relatively stable
in both domains – for the full sample and all subgroups. The trajectories of both job
and off-job crafting were almost flat and mean trends in change at different time
points were negative yet also close to zero. This suggests that employees, on aver-
age, have reduced their crafting, which is surprising given that one would expect
an increase in crafting to adapt to pandemic challenges. Employees were likely
absorbed with managing day-to-day life, leaving no time for crafting. Syrek etal.
(2022) found that participants described vividly how managing different roles and
increased work-nonwork conflicts was found challenging and stressful, likely leav-
ing little opportunity to address one’s own needs and proactively shape daily life
experiences in both domains to support wellbeing. Interestingly, Syrek etal. (2022)
found that work-nonwork balance improved towards the end of their study period,
which they interpret as an indication that employees quickly developed suitable
strategies to handle the pandemic situation. In contrast, our study, covering a longer
timeframe, indicates that employees, even in the long run, seemed to have directed
their energies less toward proactive crafting and potentially more towards reactively
managing daily pandemic life.
However, since we measured crafting on a frequency scale, our study findings
may also be interpreted in such that employees retained their crafting frequency
or the time spent on crafting over the time span covered in this research (over two
years from June 2019 to December 2021), despite the stress of the pandemic (Syrek
etal., 2022). individuals only decreased their crafting time to a small extent. How-
ever, individuals may have altered how and what they crafted during this period. We
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
found partial metric and scalar measurement invariance in most models, indicating
potential changes in prominent crafting strategies or overall strategy usage. The pan-
demic may have also led to new crafting strategies not captured in current research.
Future studies should explore this further.
Further, considering the decreasing trends in crafting (frequency) that we observe
in this study, it also leads to the question of whether new or more prominent craft-
ing strategies applied during the pandemic simply require less time or need to be
exhibited less often to be deemed fruitful and satisfy the crafter’s needs. For exam-
ple, before the pandemic, employees might have crafted for social resources through
regular interactions, i.e. by joining work breaks. With social contact limited during
the pandemic, they may have adapted by meeting less often, likely in virtual set-
tings, and focusing more on the quality of these interactions. This shift may reflect
an overall decrease in crafting activity, despite changes in specific strategies. Future
research should explore these crafting changes qualitatively for deeper insights.
Two findings highlight the need to closely examine how and why crafting
changed during the pandemic. Firstly, there was a slight increase in job crafting
from Waves 2 to 3 for the full sample, but not for subgroups (RQ3), which showed
no mean trend. Secondly, while overall job and off-job crafting trajectories appeared
flat or slightly decreasing, significant variances in change scores indicated individual
differences. This suggests that crafting patterns varied even if group-level trends did
not show notable differences over time. There is apparently variation in the job and
off-job crafting patterns that our analyses did not capture, e.g., by comparing the
demographic subgroups in our sample.
Therefore, although we examined group differences for employee subgroups that
have been previously discussed as differentially affected by the pandemic (Kalti-
ainen & Hakanen, 2022; OECD, 2020a, 2020b; Tušl etal., 2021), our results suggest
no notable differences in the crafting patterns change over time. Employees’ percep-
tions of the pandemic and associated life disruptions might better explain fluctua-
tions in their crafting behaviors than demographic factors. While the pandemic itself
might be considered a worthy “reason to” engage in crafting, according to the model
of proactive behavior (Parker etal., 2010), it may be that employees perceived fewer
opportunities to craft due to restrictions (fitting the “can do” motivational state) or
lower energy to engage in behavior to promote their own health (fitting the “ener-
gized to” motivational state). Future qualitative research could utilize this frame-
work to capture specific perceived motivations and barriers to explain the crafting
changes observed in this study. Future research should moreover consider person-
centered approaches, e.g., clustering, to identify groups with more similar patterns
in crafting over time and relate this to other characteristics, e.g., personality traits.
Lastly, we focus on the connection between job and off-job crafting changes.
Our results indicate a consistent, small but positive association between job and off-
job crafting changes throughout the pandemic. Therefore, employees decreased or
increased their crafting similarly in both domains. This finding supports spillover/
congruency theories for crafting (Demerouti et al., 2020), apparently regardless of
the pandemic context. However, we found no connection between job and off-job
crafting changes between Waves 1 and 2 among those employees experiencing con-
tractual changes (RQ3). Conceivably, this group focused crafting efforts more on one
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
domain temporarily, e.g., by explicitly crafting the job domain to secure a more finan-
cial future despite the forced contractual changes. Our results, however, do not sup-
port such assumptions of a consistent, temporary compensatory effect between job
and off-job crafting. Instead, it seems that affected employees differed greatly in their
crafting allocation in this phase, as seen in the non-significant covariance. It remains
unclear what identified those crafting to compensate vs. those crafting to a similar
extent in this specific phase of the pandemic. Future research could utilize qualitative
methods (e.g., interviews) with affected employees to gain insights into the motives
and drivers of different strategies, e.g., the pressure of precarious employment and
financial issues (OECD, 2020a, 2020b) or positive aspects, such as reported increases
in leisure time and time spent with partner/family (Tušl etal., 2021).
To sum up, although the trajectories of job and off-job crafting remained flat for
all groups, we observed interindividual changes, and for most groups, the associa-
tion between changes was weak and positive throughout the pandemic. Therefore,
our findings seem to support earlier research suggesting a congruency in crafting
across life domains (Demerouti etal., 2020), even when now referring to crafting
changes.
Changes inCrafting andSelf‑Rated Health
In RQ2, we posed the question of whether we might observe temporal differences in
the (positive) associations between crafting changes and subsequent levels of self-
rated health during the pandemic. In our study, we investigated how intraindividual
changes in crafting during the pandemic may have led to interindividual differences
in health outcomes, providing insights into the potential benefits of changing craft-
ing efforts across life domains. We also examined temporal and contextual variations
in the relationship between crafting and self-rated health to understand how different
individuals have been affected by adaptive adjustments during the pandemic (RQ4).
However, before scrutinizing evidence answering RQ2 and RQ4, we review
the development of self-rated health during the pandemic. Our results indicate
that self-rated health remained stable during the pandemic for the full sample
and all subgroups. Self-rated health is reportedly responsive to changing health-
related circumstances (Perruccio et al., 2010). While some individuals affected
by COVID-19 reported lower self-rated health (Peters etal., 2020), other studies
support our findings of overall stability (Peters etal., 2020; Recchi etal., 2020;
Szwarcwald etal., 2021; van de Weijer etal., 2022). Additionally, mental health
indicators remained stable (Van Tilburg etal., 2021), and a Finnish study found
only minor changes in occupational well-being across demographic groups Finland
(Kaltiainen & Hakanen, 2022). Notably, our sample was not significantly impacted
by COVID-19, with only 1.70% reporting infections at Wave 3 (see Table1).
Even though self-rated health remained largely stable, positive – yet weak—asso-
ciations were discernible between crafting changes and self-rated health (RQ2),
that emerged in a noteworthy pattern. Early in the pandemic, we observed positive
associations between job crafting increases and self-rated health, whereas later in
the pandemic, we observed positive associations between off-job crafting increases
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
and self-rated health. Changes in crafting occurred at all time points, but apparently
only those changes in the job domain at the pandemic onset related to subsequent
higher levels of self-rated health. Only those changes in the off-job domain towards
a normalization phase related to subsequent higher levels of self-rated health, even
though both domains were similarly affected by the pandemic (Tušl etal., 2021).
First, this indicates that individual increases in crafting in the job and off-job
domain relate to later interindividual differences in health with a temporal shift
during the pandemic. Increasing crafting in the job domain in the beginning of the
pandemic relates to health-related advantages compared to those employees who
retained their level of crafting or decreased it. Similarly, increasing crafting outside
the job relates to health-related advantages, compared to employees who retained or
decreased their level of crafting. However, this beneficial process seemed to have
occurred rather at the end of the pandemic.
It may be that the job domain was more salient for individuals at the beginning
of the pandemic. Employment means a source of income, structure, identity, social
connections, and development opportunities (Witte, 1999). When employment and
pay are threatened, individuals’ quality of life is impaired (Winefield & Tiggemann,
1990). Further, job insecurity increased at the beginning of the pandemic but later
decreased again (El Khawli etal., 2022). Therefore, understandably, our results indi-
cate that when the pandemic broke out, increasing crafting in the job domain was
related to advantages regarding self-rated health, whereas less so increasing crafting
in the off-job domain. However, with this study, we can only reveal this notable pat-
tern, but not potential causes. Future research should first aim to corroborate such a
domain shift in this association in other changing circumstances, e.g., when employ-
ees face major organizational restructuring or more individual health crises.
This pattern of a domain shift in the association between crafting changes and
self-rated health was seen in all subgroup analyses, except for employees living
alone (RQ4). For these, we found no association between off-job crafting changes
and self-rated health. However, we observed a positive association between self-
reported belonging to a high-risk group and self-rated health. Potentially, for this
group, perceived high risks of severe complications with a COVID-19 infection
overshadowed individual crafting efforts when predicting self-rated health. Future
research should investigate barriers between individual off-job crafting and self-
rated health for those with major salient health risks.
Strengths andLimitations
To the best of our knowledge, our study is the first to examine the changes in crafting
in different life domains during the COVID-19 pandemic and how crafting changes
relate to self-rated health. Our results contribute to the knowledge of crafting as a
protective factor for health during the past pandemic. In future crises, this individual
strategy could be promoted to equip individuals with knowledge, skills, and oppor-
tunities to maintain their own health. Further, we note our relatively large and repre-
sentative sample from Germany, Austria, and Switzerland, which we have followed
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
for several years. This rich dataset, let us examine changes over a significant part of
the pandemic and also include pre-pandemic and more stable phases.
However, our study also has several limitations. First, we encountered issues in
establishing metric measurement invariance (equal factor loadings) for job crafting.
These made it impossible to examine differences in the dynamic association of job
and off-job crafting and also with self-rated health depending on individual caregiv-
ing duties for others (further information: see methods). For this set of subgroups,
we could not establish even partial metric measurement for job crafting for sub-
groups with caregiving duties, a relevant criterion to correctly interpret longitudinal
associations and differences in latent constructs (Putnick & Bornstein, 2016). Job
crafting may mean something slightly different to an individual depending on the
life stage (e.g., young parent vs. older parent) or life circumstances (e.g., children
vs. no children). Differences in the understanding of job crafting by specific demo-
graphic groups should be considered and investigated more thoroughly in future
research (e.g., Vignoli etal., 2023). For transparency and replicability, we provide
our analyses in ESM 5.
Second, we were overall only able to establish partial scalar measurement invari-
ance (equal item/item parcel intercepts) for job and off-job crafting in all analyses,
and above that, we were only able to establish partial metric measurement invari-
ance for job crafting for our group comparison based on contractual changes. Given
the complexity of our analyses with four time points over a long period and several
multigroup analyses, it is unsurprising that rigorous assumptions for measurement
invariance cannot be met. Both job and off-job are constituted by four respective
six facets, and it is likely that the overall composition of crafting changes over time.
Such a change would indicate a shift regarding the main crafting strategy within
a domain. Moreover, the overall composition of job crafting strategies may dif-
fer among employees with and without contractual changes. In our analyses, we
released the factor loading from “increasing social job resources” to the overall job
crafting factor. Those with contractual changes likely had less need or opportunity
to contact colleagues and supervisors, therefore relying less on this job crafting
strategy. Future research should consider whether the composition of job crafting
strategies also changes in other similar circumstances, e.g., during organizational
changes.
In general, our analyses focused on overall job and off-job crafting trends, not
specific facets. Therefore, we argue that combining job and off-job crafting into
single factors and allowing for partial measurement invariance is an appropri-
ate strategy to handle complexities. Further, we suggest that the differences in
the measurement model emerging due to partial measurement invariance should
be relatively minor, as differences, e.g., in item intercepts were close to zero
between a fixed and freely estimated model, and results are still interpretable. We
compiled model fit indices and comparison tests regarding measurement invari-
ance in ESM 1.
Lastly, our study focused specifically on the pandemic, and our results cannot be
generalized to other crises (e.g., financial crises, natural disasters, war). Moreover,
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
our sample consists solely of participants from Germany, Austria, and Switzer-
land, so generalizations to other socioeconomic and political settings are not advis-
able. Future research could investigate whether intraindividual changes can also be
observed over time outside the pandemic context and which other larger changes,
e.g., major organizational transformations, may also involve changes in crafting of a
whole population.
Practical Implications
Political and organizational institutions have a dual responsibility during cri-
ses—implementing measures to contain the spread of the virus and protecting
individuals while also supporting their health. Our study reveals that individuals
can benefit from increasing their crafting efforts in terms of self-rated health,
mainly at the beginning and end of the pandemic. This has three important
implications for practice. First, increasing crafting might be encouraged during
times of crisis. Previous research has demonstrated the use and effectiveness of
crafting interventions (Petrou & de Vries, 2023; Roczniewska etal., 2023; van
den Heuvel etal., 2015; van Wingerden etal., 2017). Institutions, organizations,
and political bodies might incorporate crafting interventions in future crisis
mitigation plans to support employees’ health. Second, to minimize health dif-
ferences between individuals emerging during crises, those individuals who did
not increase their crafting might be considered a risk group needing further sup-
port. Within this study, we found no links to demographic characteristics to help
identify such individuals. We encourage organizations and leaders to engage in
conversations with their employees specifically to understand additional needs
for support, induced e.g. by increased care duties or reduced contact with cow-
orkers, and to offer targeted support, e.g., by increasing social support or help
identifying further institutional offers such as additional childcare (for further
suggestions, see, e.g.: Bernhardt etal., 2023; Greer etal., 2023; Pichler etal.,
2023).
Third, we found that for those living alone, changes in off-job crafting at later
pandemic phases did not predict self-rated health, and self-reported belonging to
a high-risk group potentially overshadowed individual efforts to meet one’s needs.
These employees may need support to stay healthy in a crisis like the COVID-19
pandemic. These employees risked severe isolation from usual crafting opportu-
nities outside the job, as they adhered strictly to political social distancing meas-
ures to minimize the health risks. Employers and main leisure sites, e.g., sports
clubs, could offer targeted consultations in form of crafting interventions to help
these individuals overcome perceived barriers to their needs satisfaction. Employ-
ers could consider how affected individuals could achieve detachment and relaxa-
tion by exploring new activities, while main leisure sites might need to consider
ways of allowing such individuals to participate safely in relevant activities, e.g.,
by continuing outdoor sports activities.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Conclusion
Although the mean trajectories of change in crafting and self-rated health during
the pandemic remained flat, individuals did change their crafting during the pan-
demic. Further, their crafting changes were positively associated with later self-
reported health, highlighting how increasing crafting relates to broader interindi-
vidual health differences over time. We note a temporal amplification of specific
life domains in explaining associations between crafting changes and self-rated
health. First, job crafting changes – but not off-job crafting changes – predict
self-rated health, while later in the pandemic, the relationship is reversed. How-
ever, it remains unclear which employees increased their crafting, and why others
decreased it. We call for future research to investigate perceived barriers more
closely to better support employees in maintaining their health in times of crisis.
Overall, we suggest that as an individual strategy for maintaining health during
crises, crafting complements organizational and public health measures designed
to support individuals in challenging times. Employees might benefit from craft-
ing interventions providing them with the knowledge and skills to promote their
own health, making the workforce more resilient in future crises.
Supplementary Information The online version contains supplementary material available at https:// doi.
org/ 10. 1007/ s41542- 025- 00222-5.
Author Contributions AM: Conceptualization, Data curation, Formal analysis, Methodology, Project
administration, Visualization, Writing – original draft, Writing – review and editing. GB: Conceptual-
ization, Funding acquisition, Supervision, Writing – review and editing. JdB: Conceptualization, Fund-
ing acquisition, Supervision, Writing – review and editing. ZR: Formal analysis, Methodology, Writ-
ing – review and editing. MT: Conceptualization, Formal analysis, Methodology, Visualization, Writing
– review and editing. PK: Conceptualization, Funding acquisition, Supervision, Writing – review and
editing.
Funding Open access funding provided by University of Zurich. The author(s) disclose receipt of the
following financial support for the research, authorship, and/or publication of this article: This work was
supported by the Swiss National Science Foundation (SNSF) [grant number 100019M_201113]. Addi-
tionally, JdB was supported by the Academy of Finland [grant number: 308718].
Data Availability The data that support the findings of this study are available from the authors upon
reasonable request. Study participants were asked to give consent to use the data for research and within
research publication, but not for open public access. We provide additional material, including analysis
syntax, in the electronic supplementary materials ESM 1–5.
Declarations
Ethical Approval For this study, according to the regulations of the author’s institute’s and in accordance
with Switzerland’s Federal Act on Research Involving Human Beings (Human Research Act, HRA) no
ethical approval was required (further information is available: https:// www. uzh. ch/ en/ resea rchin novat ion/
ethics/ human resea rch. html). This observational study collected anonymized data through a panel provider
and did not assess any health-related data. Further, the data collection did not cause any psychological
distress to participants or use any form of experimental manipulation. We obtained informed consent
from all individual participants included in the study. Only participants over 18years were eligible for
participation.
Conflicts of Interest All authors declare no conflicts of interest.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended
use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permis-
sion directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/
licenses/by/4.0/.
References
Abdel Hadi, S., Bakker, A. B., & Häusser, J. A. (2021). The role of leisure crafting for emotional exhaus-
tion in telework during the COVID-19 pandemic. Anxiety, Stress, & Coping, 34(5), 530–544.
https:// doi. org/ 10. 1080/ 10615 806. 2021. 19034 47
Amstadter, A. B., Moreland Begle, A., Cisler, J. M., Hernandez, M. A., Muzzy, W., & Acierno, R.
(2010). Prevalence and correlates of poor self-rated health in the United States: The national elder
mistreatment study. The American Journal of Geriatric Psychiatry, 18(7), 615–623.
Bakker, A. B., & Demerouti, E. (2007). The job demands-resources model: State of the art. Journal of
Managerial Psychology, 22(3), 309–328. https:// doi. org/ 10. 1108/ 02683 94071 07331 15
Bakker, A. B., Demerouti, E., & Sanz-Vergel, A. (2023). Job Demands-Resources Theory: Ten Years
Later. https:// doi. org/ 10. 1146/ annur ev- orgps ych- 120920
Behzadnia, B., & FatahModares, S. (2020). Basic psychological need-satisfying activities during the
COVID-19 outbreak. Applied Psychology: Health and Well-Being, 12(4), 1115–1139. https:// doi.
org/ 10. 1111/ aphw. 12228
Bernhardt, J., Recksiedler, C., & Linberg, A. (2023). Work from home and parenting: Examining the
role of work-family conflict and gender during the COVID-19 pandemic. Journal of Social Issues,
79(3), 935–970. https:// doi. org/ 10. 1111/ josi. 12509
Bollen, K. A. (1989). Structural equations with latent variables. Wiley. https:// doi. org/ 10. 1002/ 97811
18619 179
Brauchli, R., Kerksieck, P., Tusl, M., & Bauer, G. F. (2022). Staying healthy during COVID-19 crisis:
Well-being and salutogenic crafting among German and Swiss working population. Health Promo-
tion International, 1–13. https:// doi. org/ 10. 1093/ heapro/ daac0 04
Bruning, P. F., & Campion, M. A. (2018). A role-resource approach-avoidance model of job crafting: A
multimethod integration and extension of job crafting theory. Academy of Management Journal,
61(2), 499–522. https:// doi. org/ 10. 5465/ amj. 2015. 0604
Bundesagentur für Arbeit. (2024). Kurzarbeit (Statistiken). Retrieved January 24, 2024, from https:// stati
stik. arbei tsage ntur. de/ DE/ Navig ation/ Stati stiken/ Inter aktive- Stati stiken/ Kurza rbeit ergeld/ Kurza
rbeit ergeld- Nav. html
Bundesministerium für Gesundheit. (n.d.). Die Situation der Corona-Pandemie in Deutschland.
RetrievedAugust 9, 2023, from https:// corona- pande miera dar. de/ inzid enz
CMS Law Tax Future. (2022). FAQ – Ansprüche von Eltern bei pandemiebedingter Kinderbetreuung.
Retrieved March 08, 2023, from https:// cms. law/ de/ deu/ publi cation/ faq- anspr ueche- von- eltern- bei-
pande miebe dingt er- kinde rbetr euung# Was- besagt- der- neue- 45- Abs-2- a- SGB-V
Cotofan, M., De Neve, J.-E., Golin, M., Kaats, M., & Ward, G. (2021). World happiness report, Chap-
ter7:Work and well-being during COVID-19: Impact, inequalities, resilience, and the future of
work. https:// world happi ness. report/ ed/ 2021/ work- and- well- being- during- covid- 19- impact- inequ
aliti esres ilien ce- and- the- future- of- work/
de Bloom, J., Vaziri, H., Tay, L., & Kujanpää, M. (2020). An identity-based integrative needs model of
crafting: Crafting within and across life domains. Journal of Applied Psychology, 105(12), 1423–
1446. https:// doi. org/ 10. 1037/ apl00 00495
Demerouti, E., Hewett, R., Haun, V., De Gieter, S., Rodríguez-Sánchez, A., & Skakon, J. (2020). From
job crafting to home crafting: A daily diary study among six European countries. Human Rela-
tions, 73(7), 1010–1035. https:// doi. org/ 10. 1177/ 00187 26719 848809
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Demerouti, E., & Bakker, A. B. (2022). Job demands-resources theory in times of crises: New proposi-
tions. Organizational Psychology Review, 204138662211350. https:// doi. org/ 10. 1177/ 20413 86622
11350 22
DeSalvo, K. B., Bloser, N., Reynolds, K., He, J., & Muntner, P. (2006). Mortality prediction with a single
general self-rated health question. Journal of General Internal Medicine, 21(3), 267–275. https://
doi. org/ 10. 1111/j. 1525- 1497. 2005. 00291.x
Deutscher Gewerkschaftsbund. (2022). Corona und Kinderbetreuung: Gibt es Entschädigung oder Fre-
istellung für Eltern? Retrieved March 08, 2023, from https:// www. dgb. de/ theme n/++ co++ 18c1d
a2a- 69d0- 11ea- ad58- 52540 088ca da
El Khawli, E., Keller, A. C., Agostini, M., Gützkow, B., Kreienkamp, J., Leander, N. P., & Scheibe, S.
(2022). The rise and fall of job insecurity during a pandemic: The role of habitual coping. Journal
of Vocational Behavior, 139, 103792. https:// doi. org/ 10. 1016/j. jvb. 2022. 103792
Farrell, S. (2019). The impact of individual motivation profiles on job crafting behaviours over time:
A four wave, within-person study among low-skilled workers. Dublin City University Business
School.
French, D. J., Browning, C., Kendig, H., Luszcz, M. A., Saito, Y., Sargent-Cox, K., & Anstey, K. J.
(2012). A simple measure with complex determinants: Investigation of the correlates of self-rated
health in older men and women from three continents. BMC Public Health, 12(1), 649. https:// doi.
org/ 10. 1186/ 1471- 2458- 12- 649
Geiser, C. (2020). Longitudinal structural equation modeling with Mplus: A latent state-trait perspective.
Guilford Publications.
Greer, T. W., Payne, S. C., & Thompson, R. J. (2023). Pandemic-induced telework challenges and strate-
gies. Occupational Health Science. https:// doi. org/ 10. 1007/ s41542- 023- 00151-1
Grolemund, G., & Wickham, H. (2011). Dates and times made easy with lubridate. Journal of Statistical
Software, 40(3), 1–25. https:// doi. org/ 10. 18637/ jss. v040. i03
Guest, D. E. (2001). Perspectives on the study of work-life balance. Social Science Information, 41(2),
255–279.
Hale, T., Angrist, N., Goldszmidt, R., Kira, B., Petherick, A., Phillips, T., Webster, S., Cameron-Blake,
E., Hallas, L., Majumdar, S., & Tatlow, H. (2021). A global panel database of pandemic policies
(Oxford COVID-19 Government Response Tracker). Nature Human Behaviour, 5(4), 529–538.
https:// doi. org/ 10. 1038/ s41562- 021- 01079-8
Haun, V. C., Mülder, L. M., & Schmitz, A. (2022). From job crafting to home crafting and back again: A
shortitudinal study on the spillover between job and home crafting. Journal of Occupational and
Organizational Psychology. https:// doi. org/ 10. 1111/ joop. 12417
Holman, D., Escaffi-Schwarz, M., Vasquez, C. A., Irmer, J. P., & Zapf, D. (2023). Does job crafting affect
employee outcomes via job characteristics? A meta-analytic test of a key job crafting mechanism.
Journal of Occupational and Organizational Psychology. https:// doi. org/ 10. 1111/ joop. 12450
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conven-
tional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. https:// doi. org/
10. 1080/ 10705 51990 95401 18
Idler, E. L., & Benyamini, Y. (1997). Self-rated health and mortality: A review of twenty-seven commu-
nity studies. Journal of Health and Social Behavior, 38(1), 21–37.
Jimenez, W. P., Katz, I. M., & Liguori, E. A. (2023). Fear and trembling while working in a pandemic:
An exploratory meta-analysis of workers’ COVID-19 distress. Occupational Health Science, 7(1),
39–69. https:// doi. org/ 10. 1007/ s41542- 022- 00131-x
Jylhä, M. (2009). What is self-rated health and why does it predict mortality? Towards a unified con-
ceptual model. Social Science and Medicine, 69(3), 307–316. https:// doi. org/ 10. 1016/j. socsc imed.
2009. 05. 013
Kaltiainen, J., & Hakanen, J. (2022). Changes in occupational well-being during COVID-19: The impact
of age, gender, education, living alone, and telework in a Finnish four-wave population sample.
Scandinavian Journal of Work, Environment and Health, 48(6), 457–467. https:// doi. org/ 10. 5271/
sjweh. 4033
Kujanpää, M., & Olafsen, A. H. (2024). Take action, recover well? The role of daily proactive recovery
strategies for recovery, stress, affect, and next-day performance. Journal of Business and Psychol-
ogy. https:// doi. org/ 10. 1007/ s10869- 024- 09978-z
Kujanpää, M., Syrek, C., Lehr, D., Kinnunen, U., Reins, J. A., & de Bloom, J. (2021). Need satisfaction
and optimal functioning at leisure and work: A longitudinal validation study of the DRAMMA
model. Journal of Happiness Studies, 22(2), 681–707. https:// doi. org/ 10. 1007/ s10902- 020- 00247-3
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Kujanpää, M., Syrek, C., Tay, L., Kinnunen, U., Mäkikangas, A., Shimazu, A., Wiese, C. W., Brauchli,
R., Bauer, G. F., Kerksieck, P., Toyama, H., & de Bloom, J. (2022). Needs-based off-job crafting
across different life domains and contexts: Testing a novel conceptual and measurement approach.
Frontiers in Psychology, 13, 959296. https:// doi. org/ 10. 3389/ fpsyg. 2022. 959296
Łaszewska, A., Helter, T., & Simon, J. (2021). Perceptions of COVID-19 lockdowns and related public
health measures in Austria: A longitudinal online survey. BMC Public Health, 21(1), 1502. https://
doi. org/ 10. 1186/ s12889- 021- 11476-3
LeNoble, C., Naranjo, A., Shoss, M., & Horan, K. (2023). Navigating a context of severe uncertainty:
The effect of industry unsafety signals on employee well-being during the COVID-19 crisis. Occu-
pational Health Science. https:// doi. org/ 10. 1007/ s41542- 023- 00155-x
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel:
Exploring the question, weighing the merits. Structural Equation Modeling: A Multidisciplinary
Journal, 9(2), 151–173. https:// doi. org/ 10. 1207/ S1532 8007S EM0902_1
Mathieu, E., Ritchie, H., Ortiz-Ospina, E., Roser, M., Hasell, J., Appel, C., Giattino, C., & Rodés-Guirao,
L. (2021). A global database of COVID-19 vaccinations. Nature Human Behaviour, 5(7), 947–953.
https:// doi. org/ 10. 1038/ s41562- 021- 01122-8
Meade, A. W., & Craig, S. B. (2012). Identifying careless responses in survey data. Psychological Meth-
ods, 17(3), 437–455. https:// doi. org/ 10. 1037/ a0028 085
Meyer, R. D., Dalal, R. S., & Hermida, R. (2010). A Review and synthesis of situational strength in the
organizational sciences. Journal of Management, 36(1), 121–140. https:// doi. org/ 10. 1177/ 01492
06309 349309
Meyer, R. D., Kelly, E. D., & Bowling, N. A. (2020). Situational Strength Theory. In The Oxford Hand-
book of Psychological Situations (pp. 78–95). Oxford University Press. https:// doi. org/ 10. 1093/
oxfor dhb/ 97801 90263 348. 013.7
Newman, D. A. (2014). Missing Data: Five practical guidelines. Organizational Research Methods,
17(4), 372–411. https:// doi. org/ 10. 1177/ 10944 28114 548590
Newman, D. B., Tay, L., & Diener, E. (2014). Leisure and subjective well-being: A model of psychologi-
cal mechanisms as mediating factors. Journal of Happiness Studies, 15(3), 555–578. https:// doi.
org/ 10. 1007/ s10902- 013- 9435-x
Ng, M. A., Naranjo, A., Schlotzhauer, A. E., Shoss, M. K., Kartvelishvili, N., Bartek, M., Ingraham, K.,
Rodriguez, A., Schneider, S. K., Silverlieb-Seltzer, L., & Silva, C. (2021). Has the COVID-19 pan-
demic accelerated the future of work or changed its course? Implications for research and practice.
International Journal of Environmental Research and Public Health, 18(19), 10199. https:// doi.
org/ 10. 3390/ ijerp h1819 10199
Nielsen, K., & Abildgaard, J. S. (2012). The development and validation of a job crafting measure for use
with blue-collar workers. Work & Stress, 26(4), 365–384. https:// doi. org/ 10. 1080/ 02678 373. 2012.
733543
OECD. (2020a). Employment outlook 2020. Retrieved October 11, 2022 from https:// www. oecd. org/
emplo yment- outlo ok/ 2020/
OECD. (2020b). Supporting people and companies to deal with the COVID-19 virus: Options for an
immediate employment and social-policy response. Retrieved October 11, 2022 from http:// oe. cd/
covid 19bri efsoc ial
Orcan, F. (2013). Use of item parceling in structural equation modeling with missing data [Florida State
University]. Retrieved March 08, 2023 from https:// digin ole. lib. fsu. edu/ islan dora/ object/ fsu%
3A185 150
Parker, S. K., Bindl, U. K., & Strauss, K. (2010). Making things happen: A model of proactive motiva-
tion. Journal of Management, 36(4), 827–856. https:// doi. org/ 10. 1177/ 01492 06310 363732
Perruccio, A. V., Badley, E. M., Hogg-Johnson, S., & Davis, A. M. (2010). Characterizing self-rated
health during a period of changing health status. Social Science and Medicine, 71(9), 1636–1643.
https:// doi. org/ 10. 1016/j. socsc imed. 2010. 07. 042
Peters, A., Rospleszcz, S., Greiser, K. H., Dallavalle, M., & Berger, K. (2020). The impact of the COVID-
19 pandemic on self-reported health. Deutsches Ärzteblatt International, 117(50), 861–867. https://
doi. org/ 10. 3238/ arzte bl. 2020. 0861
Petrou, P., & Bakker, A. B. (2016). Crafting one’s leisure time in response to high job strain. Human
Relations, 69(2), 507–529. https:// doi. org/ 10. 1177/ 00187 26715 590453
Petrou, P., & de Vries, J. (2023). Context-free and work-related benefits of a leisure crafting intervention:
A randomized controlled trial. Journal of Leisure Research. https:// doi. org/ 10. 1080/ 00222 216.
2023. 22449 53
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Petrou, P., Demerouti, E., Peeters, M. C. W., Schaufeli, W. B., & Hetland, J. (2012). Crafting a job on
a daily basis: Contextual correlates and the link to work engagement. Journal of Organizational
Behavior, 33, 1120–1141. https:// doi. org/ 10. 1002/ job
Petrou, P., Bakker, A. B., & van den Heuvel, M. (2017). Weekly job crafting and leisure crafting: Impli-
cations for meaning-making and work engagement. Journal of Occupational and Organizational
Psychology, 90(2), 129–152. https:// doi. org/ 10. 1111/ joop. 12160
Pichler, S., Casper, W. J., Fletcher, L., & Babu, N. (2023). Adaptation in work and family roles link sup-
port to mental health during a pandemic. Journal of Occupational and Organizational Psychology.
https:// doi. org/ 10. 1111/ joop. 12452
Pijpker, R., Kerksieck, P., Tušl, M., de Bloom, J., Brauchli, R., & Bauer, G. F. (2022). The role of off-job
crafting in burnout prevention during COVID-19 crisis: A longitudinal study. International Jour-
nal of Environmental Research and Public Health, 19(4), 2146. https:// doi. org/ 10. 3390/ ijerp h1904
2146
Putnick, D. L., & Bornstein, M. H. (2016). Measurement invariance conventions and reporting: The
state of the art and future directions for psychological research. Developmental Review, 41, 71–90.
https:// doi. org/ 10. 1016/j. dr. 2016. 06. 004
R Core Team. (2020). R: A language and environment for statistical computing. R Foundation for Statis-
tical Computing.
Recchi, E., Ferragina, E., Helmeid, E., Pauly, S., Safi, M., Sauger, N., & Schradie, J. (2020). The “eye
of the hurricane” paradox: An unexpected and unequal rise of well-being during the COVID-19
lockdown in France. Research in Social Stratification and Mobility, 68. https:// doi. org/ 10. 1016/j.
rssm. 2020. 100508
Revelle, W. (2021). psych: Procedures for personality and psychological research, Northwestern Univer-
sity.Retrieved August 16, 2021 from https:// CRAN.R- proje ct. org/ packa ge= psych Versi oN=2. 1.9
RKI. (2023). SARS-CoV-2: Virologische Basisdaten sowie Virusvarianten. Retrieved August 09, 2023
from https:// www. rki. de/ DE/ Conte nt/ InfAZ/N/ Neuar tiges_ Coron avirus/ Virol ogisc he_ Basis daten.
html
Roczniewska, M., Rogala, A., Marszałek, M., Hasson, H., Bakker, A. B., & von Thiele Schwarz, U.
(2023). Job crafting interventions: What works, for whom, why, and in which contexts? Research
protocol for a systematic review with coincidence analysis. Systematic Reviews, 12(1), 10. https://
doi. org/ 10. 1186/ s13643- 023- 02170-z
Rosenthal, J. A. (1996). Qualitative descriptors of strength of association and effect size. Journal of
Social Service Research, 21(4), 37–59. https:// doi. org/ 10. 1300/ J079v 21n04_ 02
Rosseel, Y. (2012). lavaan: An R package for structural equation modeling. Journal of Statistical Soft-
ware, 48(2), 1–36. https:// doi. org/ 10. 18637/ jss. v048. i02
Rudolph, C. W., Katz, I. M., Lavigne, K. N., & Zacher, H. (2017). Job crafting: A meta-analysis of rela-
tionships with individual differences, job characteristics, and work outcomes. Journal of Voca-
tional Behavior, 102(May), 112–138. https:// doi. org/ 10. 1016/j. jvb. 2017. 05. 008
Rudolph, C. W., Allan, B., Clark, M., Hertel, G., Hirschi, A., Kunze, F., Shockley, K., Shoss, M., Son-
nentag, S., & Zacher, H. (2021). Pandemics: Implications for research and practice in industrial
and organizational psychology. Industrial and Organizational Psychology, 14(1–2), 1–35. https://
doi. org/ 10. 1017/ iop. 2020. 48
Ryan, O., Kuiper, R. M., & Hamaker, E. L. (2018). A continuous-time approach to intensive longitudinal
data: What, why, and how? In Continuous Time Modeling in the Behavioral and Related Sciences
(pp. 27–54). Springer International Publishing. https:// doi. org/ 10. 1007/ 978-3- 319- 77219-6_2
Sinclair, R. R., Allen, T., Barber, L., Bergman, M., Britt, T., Butler, A., Ford, M., Hammer, L., Kath, L.,
Probst, T., & Yuan, Z. (2020). Occupational health science in the time of COVID-19: Now more
than ever. Occupational Health Science, 4(1–2), 1–22. https:// doi. org/ 10. 1007/ s41542- 020- 00064-3
Snir, R., & Harpaz, I. (2002). Work-leisure relations: Leisure orientation and the meaning of work. Jour-
nal of Leisure Research, 34(2), 178–203. https:// doi. org/ 10. 1080/ 00222 216. 2002. 11949 968
Sonnentag, S., & Fritz, C. (2007). The recovery experience questionnaire: Development and validation of
a measure for assessing recuperation and unwinding from work. Journal of Occupational Health
Psychology, 12(3), 204–221. https:// doi. org/ 10. 1037/ 1076- 8998. 12.3. 204
Stanley, D. (2021). apaTables: Create american psychological association (APA) style tables. R package
version 2.0.8. Retrieved August 16, 2021 from https:// CRAN.R- proje ct. org/ packa ge= apaTa bles
Steinmetz, H., Schmidt, P., Tina-Booh, A., Wieczorek, S., & Schwartz, S. H. (2009). Testing measure-
ment invariance using multigroup CFA: Differences between educational groups in human values
measurement. Quality & Quantity, 43(4), 599–616. https:// doi. org/ 10. 1007/ s11135- 007- 9143-x
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Syrek, C., Kühnel, J., Vahle-Hinz, T., & de Bloom, J. (2022). Being an accountant, cook, entertainer and
teacher—all at the same time: Changes in employees’ work and work-related well-being during the
coronavirus (COVID-19) pandemic. International Journal of Psychology, 57(1), 20–32. https:// doi.
org/ 10. 1002/ ijop. 12761
Szwarcwald, C. L., Damacena, G. N., De Azevedo Barros, M. B., Malta, D. C., De Souza Júnior, P. R.
B., Azevedo, L. O., Machado, Í. E., Lima, M. G., Romero, D., Gomes, C. S., Werneck, A. O.,
Da Silva, D. R. P., Gracie, R., Fátima, De., & de Pina, M. (2021). Factors affecting Brazilians’
self-rated health during the COVID-19 pandemic. Cadernos de Saude Publica, 37(3), e00182720.
https:// doi. org/ 10. 1590/ 0102- 311X0 01827 20
Tims, M., & Bakker, A. B. (2010). Job crafting: Towards a new model of individual job redesign. SA
Journal of Industrial Psychology, 36(2), 1–9. https:// doi. org/ 10. 4102/ sajip. v36i2. 841
Tims, M., Bakker, A. B., & Derks, D. (2012). Development and validation of the job crafting scale. Jour-
nal of Vocational Behavior, 80(1), 173–186. https:// doi. org/ 10. 1016/j. jvb. 2011. 05. 009
Tims, M., Bakker, A. B., & Derks, D. (2013). The impact of job crafting on job demands, job resources,
and well-being. Journal of Occupational Health Psychology, 18(2), 230–240. https:// doi. org/ 10.
1037/ a0032 141
Tims, M., Twemlow, M., & Fong, C. Y. M. (2021). A state-of-the-art overview of job-crafting research:
Current trends and future research directions. Career Development International, ahead-of-
p(ahead-of-print). https:// doi. org/ 10. 1108/ CDI- 08- 2021- 0216
Tušl, M., Brauchli, R., Kerksieck, P., & Bauer, G. F. (2021). Impact of the COVID-19 crisis on work and
private life, mental well-being and self-rated health in German and Swiss employees: A cross-sec-
tional online survey. BMC Public Health, 21(1), 741. https:// doi. org/ 10. 1186/ s12889- 021- 10788-8
Tušl, M., de Bloom, J., & Bauer, G. F. (2022). Sense of coherence, off-job crafting, and mental well-
being: A path of positive health development. Health Promotion International, 37(6), 1–11. https://
doi. org/ 10. 1093/ heapro/ daac1 59
Tušl, M., Bauer, G. F., Kujanpää, M., Toyama, H., Shimazu, A., & De Bloom, J. (2024). Needs-based job
crafting: Validation of a new scale based on psychological needs. Journal of Occupational Health
Psychology. https:// doi. org/ 10. 1037/ ocp00 00372
van de Weijer, M. P., de Vries, L. P., Pelt, D. H. M., Ligthart, L., Willemsen, G., Boomsma, D. I., de
Geus, E., & Bartels, M. (2022). Self-rated health when population health is challenged by the
COVID-19 pandemic: A longitudinal study. Social Science and Medicine, 306, 115156. https:// doi.
org/ 10. 1016/j. socsc imed. 2022. 115156
van den Heuvel, M., Demerouti, E., & Peeters, M. C. W. (2015). The job crafting intervention: Effects on
job resources, self-efficacy, and affective well-being. Journal of Occupational and Organizational
Psychology, 88(3), 511–532. https:// doi. org/ 10. 1111/ joop. 12128
Van Lente, E., Barry, M. M., Molcho, M., Morgan, K., Watson, D., Harrington, J., & McGee, H. (2012).
Measuring population mental health and social well-being. International Journal of Public Health,
57(2), 421–430. https:// doi. org/ 10. 1007/ s00038- 011- 0317-x
Van Tilburg, T. G., Steinmetz, S., Stolte, E., Van Der Roest, H., & De Vries, D. H. (2021). Loneliness
and mental health during the COVID-19 pandemic: A study among Dutch older adults. Journals of
Gerontology - Series B Psychological Sciences and Social Sciences, 76(7), E249–E255. https:// doi.
org/ 10. 1093/ geronb/ gbaa1 11
van Montfort, K., Oud, J. H. L., & Voelkle, M. C. (2018). Continuous Time Modeling in the Behavio-
ral and Related Sciences. In Continuous Time Modeling in the Behavioral and Related Sciences.
Springer International Publishing. https:// doi. org/ 10. 1007/ 978-3- 319- 77219-6
van Wingerden, J., Bakker, A. B., & Derks, D. (2017). The longitudinal impact of a job crafting interven-
tion. European Journal of Work and Organizational Psychology, 26(1), 107–119. https:// doi. org/
10. 1080/ 13594 32X. 2016. 12242 33
Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance litera-
ture: Suggestions, practices, and recommendations for organizational research. https:// doi. org/ 10.
1177/ 10944 28100 31002
Verelst, L., De Cooman, R., & Verbruggen, M. (2023). Crafting when teleworking: A daily diary study
on the combinations of job and home crafting and their relationship with energy depletion. Journal
of Vocational Behavior, 143, 103880. https:// doi. org/ 10. 1016/j. jvb. 2023. 103880
Vignoli, M., Perinelli, E., Demerouti, E., & Truxillo, D. M. (2023). An analysis of the multidimensional
structure of job crafting for older workers with a managerial role. Work, Aging and Retirement,
9(1), 136–150. https:// doi. org/ 10. 1093/ workar/ waab0 31
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Occupational Health Science
Authors and Aliations
AnjaIsabelMorstatt1,2 · GeorgF.Bauer2 · JessicadeBloom3,4 ·
ZacharyJ.Roman5,6 · MartinTušl2 · PhilippKerksieck2
* Anja Isabel Morstatt
anja-isabel.morstatt@tu-braunschweig.de
1 Industrial/Organizational andSocial Psychology, Technische Universität Braunschweig,
Brunswick, Germany
2 Public andOrganizational Health / Center ofSalutogenesis, Institute ofEpidemiology,
Biostatistics, andPrevention, University ofZurich, Zurich, Switzerland
3 Faculty ofSocial Sciences (Psychology), Tampere University, Tampere, Finland
4 Department ofHRM&OB, Faculty ofEconomics andBusiness, University ofGroningen,
Groningen, Netherlands
5 Department ofInformatics, Social Computing Group, University ofZurich, Zurich, Switzerland
6 Department ofPsychology, Psychological Methods, Evaluation, andStatistics, University
ofZurich, Zurich, Switzerland
Weigelt, O., Marcus, B., Felfe, J., Kluge, A., & Ontrup, G. (2021). Working under pandemic conditions:
Crisis as an opportunity for change and some thoughts on the future of work. In Zeitschrift fur
Arbeits- und Organisationspsychologie (Vol. 65, Issue 4, pp. 181–187). Hogrefe Verlag GmbH &
Co. KG. https:// doi. org/ 10. 1026/ 0932- 4089/ a0003 76
WHO. (2021). Pandemie der Coronavirus-Krankheit (COVID-19). Retrieved February 05, 2025 from
https:// www. who. int/ europe/ emerg encies/ situa tions/ covid- 19
WHO. (2023). Tracking SARS-CoV-2 variants. Retrieved March 09, 2022 from https:// www. who. int/ en/
activ ities/ track ing- SARS- CoV-2- varia nts/
Wickham, H., & Hester, J. (2021). readr: Read rectangular text data. R package version 2.1.0. Retrieved
August 16, 2021 from https:// CRAN.R- proje ct. org/ packa ge= readr
Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L., François, R., Grolemund, G., Hayes, A.,
Henry, L., Hester, J., Kuhn, M., Pedersen, T., Miller, E., Bache, S., Müller, K., Ooms, J., Robinson,
D., Seidel, D., Spinu, V., … Yutani, H. (2019). Welcome to the Tidyverse. Journal of Open Source
Software, 4(43), 1686. https:// doi. org/ 10. 21105/ joss. 01686
Wickham, H. (n.d.). ggplot2: Elegant graphics for data analysis. Springer.
Wiedemann, M., Thew, G., Košir, U., & Ehlers, A. (2022). lcsm: An R package and tutorial on latent
change score modelling. Wellcome Open Research, 7, 149. https:// doi. org/ 10. 12688/ wellc omeop
enres. 17536.1
Winefield, A. H., & Tiggemann, M. (1990). Employment status and psychological well-being: A longitu-
dinal study. Journal of Applied Psychology, 75(4), 455–459.
Witte, H. D. (1999). Job insecurity and psychological well-being: Review of the literature and explora-
tion of some unresolved issues. European Journal of Work and Organizational Psychology, 8(2),
155–177. https:// doi. org/ 10. 1080/ 13594 32993 98302
World Health Organization (WHO). (1996). Health interview surveys: Towards international harmoniza-
tion of methods and instruments. WHO Office for Europe.
Wrzesniewski, A., & Dutton, J. E. (2001). Crafting a job: Revisioning employees as active crafters of
their work. Academy of Management Review, 26(2), 179–201.
Zyphur, M. J., Allison, P. D., Tay, L., Voelkle, M. C., Preacher, K. J., Zhang, Z., Hamaker, E. L., Sham-
sollahi, A., Pierides, D. C., Koval, P., & Diener, E. (2020). From data to causes I: Building a
general cross-lagged panel model (GCLM). Organizational Research Methods, 23(4), 651–687.
https:// doi. org/ 10. 1177/ 10944 28119 847278
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center
GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers
and authorised users (“Users”), for small-scale personal, non-commercial use provided that all
copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of
use (“Terms”). For these purposes, Springer Nature considers academic use (by researchers and
students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and
conditions, a relevant site licence or a personal subscription. These Terms will prevail over any
conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of
the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may
also use these personal data internally within ResearchGate and Springer Nature and as agreed share
it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not otherwise
disclose your personal data outside the ResearchGate or the Springer Nature group of companies
unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial
use, it is important to note that Users may not:
use such content for the purpose of providing other users with access on a regular or large scale
basis or as a means to circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any
jurisdiction, or gives rise to civil liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association
unless explicitly agreed to by Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a
systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a
product or service that creates revenue, royalties, rent or income from our content or its inclusion as
part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large
scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not
obligated to publish any information or content on this website and may remove it or features or
functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke
this licence to you at any time and remove access to any copies of the Springer Nature journal content
which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or
guarantees to Users, either express or implied with respect to the Springer nature journal content and
all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published
by Springer Nature that may be licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a
regular basis or in any other manner not expressly permitted by these Terms, please contact Springer
Nature at
onlineservice@springernature.com