Leader–Member Exchange (LMX) and Adjustment to the Work Mode as Protective Factors to Counteract Exhaustion and Turnover Intention: A Chain Mediation Model

Abstract

In the context of the New Normal and the VUCA (volatile, uncertain, complex, ambiguous) era, organisations face adjustments to the changes brought by the COVID-19 pandemic, especially the shifting to new work modes and configurations with their consequences on employees’ wellbeing, in terms of exhaustion, resignation, and quitting tendencies. This study, rooted in the psychology of sustainability and sustainable development, employs a primary prevention perspective to examine a relational factor, the leader–member exchange (LMX), which might shield employees from exhaustion and turnover intention. Specifically, we propose a double-chained mediation model to investigate how high-quality LMX fosters a positive adjustment to employees’ specific work modes, in-person or hybrid work, subsequently reducing employees’ feelings of exhaustion and their intentions to leave the organisation. A convenience sample of 257 Italian employees participated in this study by completing an online self-report survey. Hypotheses were tested using the PROCESS macro in SPSS 25.0 (Model 6). The results indicate that positive LMX and adjustment to the work mode reduce exhaustion and turnover intention; furthermore, they highlight the existence of a more complex dynamic linking LMX to turnover intention through a double-chained mediation of adjustment to the work mode and exhaustion. Indeed, higher LMX quality favours adjustment to the work mode, decreasing employees’ feelings of exhaustion and, in turn, their turnover intention. Specifically, the findings of this study add a novel contribution to the literature on the psychology of sustainability and sustainable development by emphasising the significance of positive LMX in becoming a factor of wellbeing and sustainability in the workplace through the promotion of the adjustment to both in-person and hybrid work modes. Organisations may benefit of this approach to LMX that, through the consideration of employees’ needs, may favour their adjustment to different work modes, thus becoming a sustainable LMX, and a promoter of employees’ wellbeing and retention.

1. Introduction

The COVID-19 pandemic has disrupted daily life and has brought many changes in all spheres of society. These changes have also touched working life and work contexts, impacting processes and dynamics in organisations, thus endangering their sustainability [1]. Ever since, the world of work has known a period of adjustment and settlement until it reached the so-called New Normal, a new era characterised by new arrangements and work configurations [2] and by volatility, uncertainty, complexity, and ambiguity, known under the acronym VUCA [3]. Nowadays, the work context is characterised by multiple work configurations and modes, some of which are more flexible thanks to the use of technologies, such as teleworking, and remote and hybrid work, that have added to the most renowned traditional working modes, e. g. working “in-person”, “at the office”, or “onsite”. All these old, new, modified, and changing work environments, intended as the physical and psychological spaces and places in which people work, have an impact on workers, their work identity, the meaning they attribute to their work and their wellbeing [4,5].

Research on the impact of technological changes and new ways of working on workers’ and organisational wellbeing is characterised by some trends, including the study of factors that can facilitate technological change in the workplace, such as the leadership style [6], the research on workers’ adjustment to technological changes [7,8,9], and the investigation of the outcomes of these changes in terms of individual and organisational wellbeing, such as exhaustion [10] and turnover intention [11].

In the matter of leadership, those approaches that deal with leadership in technological, remote, changing, and uncertain work environments [12], where we find issues such as e-leadership [6], digital leadership [13] or AI-generated leadership, predominate [14,15]. In particular, the quality of interactions between leaders and members stands out as an important aspect fostering wellbeing, adjustment, and performance in work environments, impacted by technological change [16].

Furthermore, some recent studies conducted in the context of new ways of working have focused on the dimension of work adjustment, namely a correspondence between a worker and their work environment, considering its potential in predicting job satisfaction and lowering turnover intention [17]. Several authors have investigated the factors of adjustment to remote working during the pandemic [7,9]. However, some authors consider that the study of work adjustment should not be limited only to the remote mode but extended to different work modes [8] since the change is not just represented by the adoption of a particular way of working but also by the new awareness people developed during and after the pandemic concerning the existence of different ways of working, of which were less considered previously.

Finally, several previous studies have highlighted that technology-assisted work promotes flexibility, but also a culture of being always on that makes the boundaries between work and personal domain more blurred [18]. In the presence of adequate and sufficient resources, workers are able to manage this risk and prefer flexible work modes [8], possibly until the point of intending to leave the workplace when they perceive that this need is not met by their organisation. However, when remote work modes permit work demands to go beyond the boundaries of the work domain, summing up with personal demands, people need to intensify their efforts to manage all these demands, which may lead to resource depletion and to the risk of feeling more isolated, stressed, and exhausted [10] to the point of intending to leave the workplace.

All these premises considered, this study aims to investigate whether leader–member exchange favours people’s adjustment to different work modes, starting a potential positive chain that reduces exhaustion and turnover intention. This study is the first that investigates this topic and it does so by adopting the framework of psychology of sustainability and sustainable development.

The psychology of sustainability and sustainable development refers to the promotion of the wellbeing of people [19] and adopts a positive psychology approach to wellbeing based on the identification and promotion of personal and organisational resources, in a perspective of primary prevention at work [20] that aims at preventing risks before they occur. It can be mobilised in order to understand which aspects affect the health of people in organisations more by focusing on those psychological individual, relational, and collective dimensions at work that may turn out to be protective factors for health and wellbeing [21]. Indeed, this perspective seems to us particularly suitable to tackle the issue of how workers adjust to the changes and choices made by their organisations in terms of work modes [8] since they permit us to focus on those factors that can facilitate workers’ adjustment to their new or old work modes. In addition, the identification of these factors could help us to provide directions to implement sustainability and a culture of wellbeing in today’s organisations [21].

The present study purposes to offer several contributions. Firstly, our study adopts the framework of the psychology of sustainability and sustainable development and the perspective of primary prevention and, unlike many of the studies already conducted [22,23], it focuses on two protective factors for wellbeing: leader–member exchange (LMX) [24] and adjustment to a work mode [8], a dimension mostly studied in specific contexts, e.g., remote work, and less investigated in broader contexts. Secondly, this study is the only one that investigates LMX as a relational factor favouring individual adjustment to different work modes, namely onsite and hybrid work, and that verifies the dynamic through which these two dimensions act together, creating a preventive chain against exhaustion and turnover intention. Thirdly, our study focuses on exhaustion and turnover intention, which, due to the costs entailed for organisations, for instance, in terms of reduced productivity, the management of absenteeism, the hiring processes and work rearrangement, represent two of the major concerns for organisational sustainability [25,26]. In sum, our study aims to represent an original contribution to the psychology of sustainability and sustainable development to the extent that it questions the role of LMX and adjustment to different work modes as protective factors for wellbeing and sustainable development. Indeed, in the current work configurations, leadership and one’s way of working represent the dimensions of work environment, which may have the potential of corresponding to individual’s needs and, thus, of being factors reducing illbeing and motivating boosts for employees’ retention.

The following paragraphs are divided into several key sections. The first section will examine the concept of sustainable leader–member exchange (LMX), highlighting the essential role of leadership in promoting individual wellbeing and sustainability. The section after addresses the adjustment to the work mode, discussing the evolving physical and psychological work environments in the New Normal and VUCA era. Subsequently, the relationship between sustainable LMX and the adjustment to a work mode will be discussed as protective factors against exhaustion and turnover intention.

Finally, the hypothesised double-chained model, which illustrates the interconnected relationships among LMX, adjustment to a work mode, exhaustion, and turnover intention, will be presented.

2. Literature Review

2.1. Sustainable LMX

Leadership is considered a crucial aspect and fundamental ingredient for wellbeing, sustainability, and the sustainable development of organisations [27]. In the literature, many different approaches to leadership, which can be identified as trait, behavioural, contingency, contemporary, and new approaches, have been theorised [28]. Among the current contemporary perspectives to leadership, some positive leadership theories stand out, such as transformational leadership [29], LMX (leader–member exchange) [24], servant leadership [30,31] and the authentic leadership [32] theory.

More recently, the concept of sustainable leadership has emerged in the background of the awareness that organisations cannot survive in the long run unless they care about workers’ wellbeing, and about the physical, social, and psychological environment in which they are embedded [33]. Some authors define socially sustainable leadership as “influencing followers to accomplish sustainable objectives while treating followers with dignity and respect” [34] (pag. 155). Since this definition encompasses the importance of positive leader–member exchanges [35], in this study, we adopt the leader–member exchange (LMX) theory [24], which aims to define the quality of the relationship between a leader and a member of an organisation or a team, assuming that this relationship is dyadic and of mutual exchange [36,37]. The LMX theory has already been considered by previous research on the impact of technological introduction, influence, and change at work, and on work dynamics [6]. Furthermore, by presenting certain features, the LMX can be a form of socially sustainable leadership. According to Carbo and colleagues “socially sustainable LMX leaders must be developers/improvers. It is the socially sustainable leader’s responsibility to care for every follower as they develop and maximize the quality of each leader-follower relationship. In addition, they must provide the resources that enable their followers to develop themselves. Finally, they must strive to improve working conditions, organizational culture, and other situational factors on which LMX depends” [34] (pag. 161).

The LMX theory assumes that the higher the quality of the leader–member exchange, the greater the levels of trust, support, and mutual influence. All of these constructs could foster the correspondence between a worker and their work environment. Thinking of the different work modalities, where the physical and psychological workspace changes, alternates, and modifies, to be sustainable, the LMX should have the characteristics that correspond to the new needs of the worker. For example, during remote work, there must be a continuous exchange that allows the worker to feel supported, but not controlled; this allows them to feel that there is a bond, albeit at a distance, which makes them feel less of a sense of isolation and as though they still belong to the company [38].

2.2. Adjustment to the Work Mode

Considering that in the New Normal and VUCA era workers who are confronted with different work modes experience new physical and psychological environments, it is even more important, not only to study the quality of the relationship one has with one’s leader, but also the context in which this LMX is established and consolidated over time by measuring the individual perception of one’s work environment. The work environment, if intended as the physical space in which people perform their work, can no longer be identified as “the office”, “the organisation”, or “the workplace”; it seems to take the form of new work spaces, physical and virtual, often defined through flexible or non-traditional ways of working [39] and characterised by the possibility to work anywhere and anytime.

Take, for example, the blue-collar worker who, by definition, works onsite, but frequently participates in remote meetings with their manager, or the white-collar worker who responds to e-mails during off-work hours from home or anywhere else, thanks to technologies. In the new environments/spaces correspondent to different work modes, notably those of which we focus on in our study, namely working in-person and the alternation of remote and in-person work known as hybrid work, workers develop new perceptions, new representations, and new needs emerge, including those of satisfaction, commitment, productivity, and the ability to balance work and non-work demands [9]. In fact, in the New Normal and VUCA era, the worker needs to find a way to better adjust to new and different work modes without running into new risks, such as over-connection, work intensification, and workaholism [40]. The challenge for a worker is to find the best adjustment to these new work modes, which were either imposed on them by the organisation or, at best, negotiated with their manager, considering that, often, the range of ways of working, officially defined or informally implemented, is wider compared to the pre-pandemic era.

The dimensions of worker’s needs, correspondence, and work environment are part of the Work Adjustment Theory [17,41], which is based on the concept of correspondence between the individual and their work environment, implying a harmonious and complementary relationship between them, in which the individual sees in that environment a good fit and a good correspondence with their needs and, likewise, the work environment responds with specific rewards (salary, prestige, interpersonal relationships between bosses and colleagues, flexibility, autonomy) that fit the individual’s commitment to achieving specific work requirements [42,43].

In this study, we refer to the adjustment to one’s work environment as a state of adaptation to the mode in which people perform their work, expressed by satisfaction, self-efficacy, performance, and a work–life balance [7,8,9].

2.3. Sustainable LMX and Adjustment to the Work Mode as Protective Factors Against Exhaustion and Turnover Intention

Despite leaders establishing different quality relationships with their employees due to the factors of personality and personal compatibility, and the competence and reliability that the subordinate demonstrates [44], some studies highlight that, when presenting positive and sustainable characteristics, the exchange between leaders and members can become a protective factor to exhaustion and turnover intention [45].

The literature has shown the importance of the leader making themselves feel present and forging a good quality relationship with their employees. A number of studies [46,47], in fact, have shown that employees who perceive a high quality LMX relationship with their leader experience greater empowerment, a greater sense of connectedness to their work, and, as a result, work harder and achieve more desirable organisational results [48]. Furthermore, a high and continuous exchange between a leader and an organisation member is associated with higher job performance [49] and job satisfaction [50], high organisational commitment [51], and frequent organisational citizenship behaviour [52], as well as lower employee turnover intention [53,54].

Positive LMX is also considered a protective factor for wellbeing [55]. Several studies have underlined that the higher the quality of the leader–member exchange, the lower the employee’s exhaustion [56,57]. Exhaustion, in turn, is positively related to turnover intention [58]. As highlighted by Bakker and de Vries, becoming increasingly exhausted at work leads to “job burnout [that] is an enduring psychological condition of ill-being signaling that employees are no longer able and no longer willing to invest effort in their work” [59] (pag. 3). In this regard, after a long exposure to high work demands, the worker may become chronically exhausted by their work and psychologically distance themselves from it. This leads the worker to be more inclined to leave the company [59].

Therefore, we assume the following:

H1. 

LMX is negatively related to exhaustion and turnover intention.

Additionally, the adjustment to one’s job or mode can also be considered an individual resource and protective factor against both exhaustion [60] and turnover intention [43,61]. In fact, if, as described above, an “adjusted” worker is also a worker who is satisfied, committed, productive, and has a good work–life balance, they will be more “equipped” with resources [62] to counteract exhaustion and will be less inclined to leave a job and an organisation that responds to their needs [61,63,64].

Therefore, we assume that:

H2. 

Adjustment to a work mode is negatively related to exhaustion and turnover intention.

2.4. The Hypothesised Double-Chained Mediation Model

In this section, all the constructs presented above will be related: LMX, adjustment to a work mode, exhaustion, and turnover intention.

The model that we present below is conceived from a primary prevention perspective within the framework of the psychology of sustainability and sustainable development, i.e., it aims to show which of the protective factors are possible in the work environment and which are the risk factors to be taken into account in order to move in the direction of preventing a phenomenon of malaise before it develops, with the aim of promoting the wellbeing in organisations.

After an analysis of the scientific literature, how LMX and adjustment to a work mode can act as protective factors in the workplace against exhaustion and turnover intention emerges. But, there is more: within the psychology of sustainability and sustainable development, several authors have suggested that positive relationships with the leader lead to wellbeing outcomes and can act as a factor of adjustment to the work environment [27]. Specifically, previous studies have shown that having a high-quality interaction with one’s leader, a good communicative exchange, and satisfying rewards for achieving goals and expectations, consolidates the exchange relationship and is a mutual adjustment resource [65,66].

Therefore, it is the quality of the exchange between a leader and a member (LMX) that allows workers to perceive that there is a mutual and functional interaction between themselves and their environment [67] and, consequently, increases their level of person–environment fit [66,68]. This will lead to a greater adjustment to the individual’s work mode on the part of the worker who will perceive an achieved but continuous correspondence between themselves and their work environment [69,70].

Thus, we assume the following:

H3. 

LMX is positively related to adjustment to the work mode.

Furthermore, according to previous studies, LMX seems to create a fertile or sterile ground for the worker to find their own work adjustment: if there is a high relational quality exchange between a leader and an organisation member, work engagement levels will increase, thus limiting exhaustion and turnover intention [65,71].

Therefore, a mechanism that links the organisational resource, represented by the LMX, to the individual resource, namely the adjustment to the work mode, that not only acts as isolated protective factors, but that are also sequentially connected in reducing exhaustion and turnover intention, seems to exist. Consequently, the objective of our study is to verify whether leader–member exchange (LMX) contributes to employees’ adjustment to different work modes and, through this process, reduces exhaustion and turnover intention. To sum up, we expect that the more a worker has a quality exchange with their leader, the more they will feel adjusted to their work mode which will lead to a decrease in exhaustion [60], and, in turn, to a decrease in turnover intention [58,59].

Our final hypothesis is the following:

H4. 

The relationship between LMX and turnover intention is sequentially mediated by adjustment to the work mode and exhaustion.

3. Research Method

3.1. Participants and Procedure

This study was approved in June 2023 by the Ethical Committee of the Catholic University of the Sacred Heart in Milan with the protocol number 65–23 and was conducted by a research team in Work and Organisational Psychology from the Catholic University of the Sacred Heart in Milan and the University of Paris Nanterre. Concerning the procedure, data collection took place between June and July 2023 using convenience sampling, which is a non-probabilistic sampling method commonly used in quantitative research due to its simplicity and the limited effort, time, and cost it requires to recruit participants [72]. The research team acquired respondents by disseminating a message of invitation to participate in the research through social and professional networks (e.g., LinkedIn and Facebook). The inclusion criteria were being aged between 18 and 67 years old and being employed; the exclusion criterion was the refusal to give informed consent. The participants who met the inclusion criteria were invited on a voluntary basis to fill out an online survey on LimeSurvey. This survey was divided into three sections: the introduction to the study, including information about the research objectives; instructions for filling out the questionnaire; a reminder of voluntary participation and the respect of anonymity and the informed consent question; a section asking for socio-professional information; and a final section containing questions related to the variables of the study. It took approximately 20 min to complete; the data collection lasted for one and a half months. Out of the 311 Italian voluntary participants, a final sample of 257 participants was retained for analysis, a sample comprising those who entirely completed the questionnaire (mean age = 46.09, SD = 12.46, range = 22–67 years old; 42.8% female).

Table 1. Respondents’ socio-demographic profiles.

Demographics	Gender		Total
	Female	Male	N	%
Age category
<25	4	2	6	2.4%
25 to 34	34	27	61	23.8%
35 to 44	23	22	45	17.6%
45 to 54	24	35	59	22.8%
55 to 64	22	56	78	30.2%
65+	3	5	8	3.1%
Total	110	147	257	100%
Education
High school	38	60	98	38.1%
Bachelor	14	18	32	12.15%
Master	37	52	89	34.6%
Specialised master	16	15	31	12.1%
PhD	5	2	7	2.7%
Total	110	147	257	100%
Occupational sector
Industry	21	31	52	20.2%
Commerce	20	24	44	17.1%
Education and research	20	10	30	11.7%
Health	13	12	25	9.7%
Finance/Bank/Insurance	10	14	24	9.3%
Administration	6	17	23	8.9%
IT	5	13	18	7%
Other sectors	15	26	41	16%
Total	110	147	257	100%
Employment sector
Private	78	109	187	72.8%
Public	32	38	70	27.2%
Total	110	147	257	100%
Employment status
Full time	86	138	224	87.2%
Part time	24	9	33	12.8%
Total	110	147	257	100%
Contract type
Open-ended	73	119	192	74.7%
Fixed term	34	25	59	22.9%
Other	3	3	6	2.3%
Total	110	147	257	100%
Work mode
Onsite	73	101	174	67.7%
Hybrid	37	46	83	32.3%
Total	110	147	257	100%
Remote days a week
1	4	8	12	4.7%
2	18	18	36	14%
3+	15	20	35	13.7%
Total	37	46	83	32.3%

presents the socio-demographic profiles of the final sample.

3.2. Measures

The leader–member exchange (LMX) was assessed through a five-item scale from Graen and Uhl-Bien’s measure [24], which had already been used in previous studies in Italy [73]. An example of an LMX item was “My manager tells me if he/she is satisfied with my work”. The items were measured with a five-point frequency scale ranging from “Never” (1) to “Always” (5).

The adjustment to the work mode was considered from the angle of adjustment to the specific work mode experienced by the worker. It was assessed through an adaptation of the five-item scale from van Zoonen et al. [9], used to measure adjustment to remote working. The term “remote” was replaced with “my work mode”, which could be onsite or hybrid [6]. An example of an item was “My work mode allows me to do my job better than I could have done in any other mode”. The items were measured with a five-point scale ranging from “Strongly disagree” (1) to “Strongly agree” (5).

Exhaustion was assessed through the five-item scale from the Maslach Burnout Inventory General Survey [74,75] which has already been used in previous Italian studies [76]. An example of an item was “I feel emotionally drained from my work”. The scale of job exhaustion was composed of five items on a seven-point frequency scale ranging from “Never” (1) to “Always” (7).

Turnover intention was assessed through the five-item scale from Bertrand’s measure [77]. An example of an item was “If nothing would prevent me from leaving my position, I would do so.” The items were measured with a four-point frequency scale ranging from “Strongly disagree” (1) to “Strongly agree” (5).

Socio-professional characteristics such as gender (0 = man, 1 = woman), age, and work mode (0 = fully onsite, 1 = hybrid) were considered as control variables.

Items from the original scales that had never been used in Italian, were adapted and translated into Italian following the translation/back translation technique [78]. The whole list of items is provided in

Table 2. Item factor loadings (pattern matrix coefficients) for confirmatory factor analysis (CFA).

Construct	1	2	3	4
Leader–member Exchange
1. Il mio manager mi dice se è soddisfatto/a del mio lavoro.	0.80
2. Il mio manager è interessato ai miei problemi e ai miei desideri in relazione al mio lavoro.	0.86
3. Mi sento apprezzato/a dal mio responsabile.	0.87
4. Il mio manager interviene per aiutarmi a risolvere i miei problemi professionali.	0.79
5. Il mio manager è amichevole e disponibile con me.	0.76
Adjustment to the Work Mode
1. Nel complesso, sono soddisfatto/a della mia modalità di lavoro.		0.86
2. La mia modalità di lavoro mi permette di fare il mio lavoro meglio di quanto avrei potuto fare in qualsiasi altra modalità.		0.79
3. Se avessi la possibilità di cambiare modalità di lavoro, sarebbe poco probabile che lo faccia.		0.68
4. Da quando lavoro in questa modalità, riesco a conciliare maggiormente la mia vita lavorativa con quella privata		0.71
5. Da quando ho iniziato a lavorare con questa modalità, la mia produttività è aumentata.		0.71
Exhaustion
1. Mi sento emotivamente svuotato/a dal mio lavoro.			0.82
2. Mi sento stanco/a quando mi alzo al mattino e devo affrontare un’altra giornata di lavoro.			0.84
3. Mi sento frustrato/a dal mio lavoro.			0.88
4. Sento di lavorare “troppo” nel mio lavoro.			0.71
5. Mi sento al limite della sopportazione.			0.91
Turnover Intention
1. A parità di impatto finanziario, deciderei di lasciare il mio lavoro.				0.78
2. Se nulla mi impedisse di lasciare la mia posizione, lo farei.				0.85
3. Il mio intento è quello di lasciare la mia organizzazione.				0.88
4. Non si può trovare di peggio rispetto al lavoro che ho adesso				0.64
5. Penso spesso a lasciare l’organizzazione in cui lavoro.				0.89

with their respective factor loadings.

3.3. Data Analyses

Data analyses were performed in several steps using MPLUS 8 [79], SPSS 25.0, and PROCESS macro 4.2 (Model 6). Firstly, we evaluated the measurement model by performing a confirmatory factor analysis (CFA). The goodness-of-fit for the model was tested using the χ² value, the comparative fit index (CFI), the Tucker–Lewis index (TLI), the root means square error of approximation (RMSEA), and the standardised root mean square residual (SRMR). In order to verify the model that was best fitted to the data, the measurement model was compared with alternative models, including a single-factor model, in order to address the common method variance issue through Harman’s single-factor test [80]. In general, the models with fit indices of >0.92 and an RMSEA of <0.08 indicated a fair fit between the model and the data [81]. Secondly, the construct validity, reliabilities, and convergent validity were evaluated considering the factor loadings greater than the threshold of 0.50 [82], the values of Cronbach’s alpha and McDonald’s omega greater than 0.70 [83], the values of composite reliability (CR) higher than 0.70 [84], and average variance extracted (AVE) that was higher than 0.50 [84]. Successively, descriptive statistics (mean, standard deviation) and the correlations between variables (Pearson’s r) were computed and the discriminant validity was assessed by comparing the root mean of the AVE with the correlation coefficients. Finally, the double mediation was tested using the PROCESS macro for SPSS that we chose because of its simplicity and ease of use [85], its less restricted requirements regarding sample size, and its recognised capability to obtain similar results to SEM for mediation analyses [86].

4. Results

4.1. Measurement Model

The measurement model was assessed through a confirmatory factor analysis. The results of the CFA and models’ comparisons are presented in

Table 3. Goodness-of-fit of the measurement model and comparison with alternative models.

Models	χ2	Δχ2	RMSEA	CFI	TLI	SRMR
Model 1	χ2 (164) = 341.92 ***		0.07	0.94	0.93	0.05
Model 2	χ2 (170) = 1394.76 ***	Compared to Model 1 Δχ2 (6) = 1052.84 ***	0.17	0.59	0.54	0.15
Model 3	χ2 (167) = 682.08 ***	Compared to Model 1 Δχ2 (3) = 340.16 ***	0.11	0.83	0.80	0.09
Model 4	χ2 (167) = 481.51 ***	Compared to Model 1 Δχ2 (3) = 139.59 ***	0.09	0.89	0.88	0.06

Note: *** p < 0.001. Chi-square (χ²). Chi-square difference (Δχ²). Degrees of freedom between parentheses. Comparative fit index (CFI). Tucker–Lewis index (TLI). Root means square error of approximation (RMSEA). Standardised root mean square residual (SRMR).

. Model 1, a four-latent-factor model, presented good fit indices: (χ² (164) = 341.92, p < 0.001, RMSEA = 0.07, CFI = 0.94, TLI = 0.93, SRMR = 0.05). This measurement model was firstly compared with a one-factor model (Model 2) in which all the items had loaded on a single latent variable in order to verify the non-presence of common method bias. The results indicated that one single factor could not account for the variance in the data (χ² (170) = 1394.76, p < 0.001, RMSEA = 0.16, CFI = 0.59, TLI = 0.54, SRMR = 0.15) and, therefore, the threat of common method bias was unlikely. Subsequently, Model 1 was compared with a three-factor model (Model 3) in which the items from LMX and adjustment to the work mode loaded on the same latent variable and with a four-factor model (Model 4) in which the items from exhaustion and turnover intention loaded on the same latent variable. Globally, we found a significant diminution of the fit and the difference between the theoretical model and alternative models was found to be significant.

The results of the CFA and models’ comparison showed that the hypothesised model (Model 1) was the best fitted to the data, with the items loading to their respective factor ranging from 0.64 to 0.91 (Table 2).

Furthermore, the measurement model was evaluated in terms of reliabilities and convergent validity. The results are reported in

Table 4. Cronbach’s alpha (α), McDonald’s omega (ω), composite reliability (CR), and average variance extracted (AVE).

Construct	α	ω	CR	AVE
Leader–member exchange	0.91	0.91	0.91	0.67
Adjustment to the work mode	0.86	0.87	0.87	0.57
Exhaustion	0.92	0.92	0.92	0.70
Turnover intention	0.91	0.91	0.91	0.66

. Firstly, all the scales presented good reliability, with the values of Cronbach’s α, McDonalds’s ω, and composite reliability (CR) ranging between 0.86 and 0.87, and 0.92. Moreover, to examine convergent validity, we verified the average variance extracted (AVE), which indicates the shared average variance between the factor and its indicators, requiring a threshold of 0.50; namely more than 50% of the variance of a construct should be due to its indicators. Since all the constructs of the study showed an AVE ranging between 0.57 and 0.70, convergent validity was met.

4.2. Descriptive Statistics, Correlations, and Discriminant Validity

Descriptive statistics and correlations are reported in

Table 5. Means, standard deviations, correlations, and square root of the average variance extracted (AVE).

	Mean	SD	1	2	3	4	5	6	7
1. Gender (1 = women)	-	-	-
2. Age	46.09	12.46	−0.21 ***	-
3. Work mode (1 = hybrid)	-	-	0.03	−0.16 *	-
4. LMX	3.32	0.90	−0.01	0.07	0.19 **	(0.82)
5. ADJ	3.54	0.87	−0.08	0.06	0.28 ***	0.45 ***	(0.75)
6. Exhaustion	3.38	1.50	0.08	−0.08	−0.05	−0.38 ***	−0.36 ***	(0.84)
7. Turnover Intention	2.11	0.77	0.14 *	−0.05	−0.07	−0.41 ***	−0.40 ***	0.77 ***	(0.81)

Note. N = 247; * p < 0.05; ** p < 0.01; *** p < 0.001. M = mean; SD = standard deviation; LMX = leader–member exchange; ADJ = adjustment to the work mode; values in brackets are square root of average variance extracted (AVE).

. To assess the discriminant validity, we followed the criterion of Fornell and Larcker [87] which suggests that discriminant validity is met when the square root of the AVE of each construct is higher than the correlations the same construct presents with the other study variables. As it is shown in Table 5, all the values of the square root of the AVE (in brackets on the diagonal) are superior to the correlations of the correspondent dimension with the other variables. Therefore, discriminant validity was met for all the dimensions of our study.

Concerning correlations between the control variables and the variables of the study, gender showed a slightly positive correlation with turnover intention, thus indicating that women reported a higher intention to quit their job than men. The work mode (onsite or hybrid) was positively associated with the LMX and the adjustment to the work mode; thus, the people working in hybrid modes reported higher levels of positive leader–member exchange and work adjustment.

Regarding the correlations between the constructs, the results showed that LMX presented a positive relationship with adjustment to the work mode (r = 0.45, p < 0.001) and a negative relationship with exhaustion (r = −0.38, p < 0.001) and turnover intention (r = −0.41, p < 0.001). Adjustment to the work mode was negatively correlated with exhaustion (r = −0.36, p < 0.001) and turnover intention (r = −0.40, p < 0.001). Furthermore, exhaustion presented a high positive relationship with turnover intention (r = 0.77, p < 0.001).

4.3. Hypotheses Testing

We tested the hypotheses using Model 6 in PROCESS for SPSS, which is indicated for serial mediation analysis. The results are reported in Figure 1. Specifically, LMX had a positive impact on adjustment to the work mode (β = 0.45, p < 0.001), thus increasing it, and a negative impact on exhaustion (β = −0.27, p < 0.001) and turnover intention (β = −0.10, p < 0.05), thus reducing them. Adjustment to the work mode had a direct negative impact on exhaustion (β = −0.24, p < 0.001) and turnover intention (β = −0.11, p < 0.05), thus decreasing them. Furthermore, exhaustion had a direct positive impact on turnover intention (β = 0.69, p < 0.001).

The results of regression with specified direct effects, the variance explained, and the effect sizes, are presented in

Table 6. Estimated direct effects.

	to ADJ	to Exhaustion	to Turnover Intention
LMX	0.45 ***	−0.27 ***	−0.10 *
ADJ		−0.24 **	−0.11 *
Exhaustion			0.69 ***
	R2 = 0.20	R2 = 0.19	R2 = 0.61
	F = 64.47	F = 29.70	F = 134.10

Note. N = 247; * p < 0.05; ** p < 0.01; *** p < 0.001. LMX = leader–member exchange; ADJ = adjustment to the work mode; EX = exhaustion; TI = turnover intention; R² = R squared; F = Cohen’s F statistic.

. The R squared values for adjustment to the work mode, exhaustion, and turnover intention corresponded, respectively, to 20%, 19%, and 61% of the variance explained. Therefore, taken collectively, the LMX adjustment to the work mode and exhaustion explained a good amount of variation in the turnover intention. The lower levels of variance explained and the effect size for adjustment to the work mode and exhaustion, as dependent variables, may have been due to the fact that the variation in these variables was due to some other variables not included in the model.

Finally, the estimated indirect effects are presented in

Table 7. Estimated indirect effects.

Effects	Coefficients	SE	95% CI
			LL	UL
LMX→ADJ→TI	−0.05	0.02	−0.10	−0.00
LMX→EX→TI	−0.19	0.05	−0.27	−0.03
LMX→ADJ→EX→TI	−0.07	0.02	−0.12	−0.03
Total Indirect Effect	−0.31	0.05	−0.40	−0.20

Note. N = 247; LMX = leader–member exchange; ADJ = adjustment to the work mode; EX = exhaustion; TI = turnover intention.

. The significance of the indirect effects was assessed using 5000 bootstrap samples and a 95% bias-corrected confident interval (CI). Following Preacher and Hayes [88] recommendations, we considered the indirect effect significant when the CI did not contain zero. As we can observe, the LMX decreased the turnover intention through adjustment to the work mode; furthermore, the LMX decreased the turnover intention by decreasing the exhaustion; finally, the LMX decreased the turnover intention through the serial mediation of both the adjustment to the work mode and exhaustion.

5. Discussion

Thinking and projecting sustainability in current work environments, characterised by the existence or awareness of the possibility of different and more flexible work modalities, means taking into consideration, in our research, the variety of such work modes, and reflecting on the workers’ need to find, in these new configurations of physical and psychological spaces, new adjustments. We see all this without missing that, in their sustainability efforts, researchers and practitioners should help organisations in identifying the factors that could contribute, not just to enhancing individual adjustment, but to reducing the levels of exhaustion and turnover intentions, which represent threats to organisational sustainability [25,26]. In this sense, our study was developed with the intent to contribute in an innovative way to the psychology of sustainability and sustainable development, considering how a high-quality leader–member exchange may represent a sustainable resource, fostering the adjustment to in-person and hybrid work, at the base of a resourceful dynamic that triggers wellbeing in contemporary work environments.

Based on a literature review, by adopting the perspective of the psychology of sustainability and sustainable development, we formulated different hypotheses comprised in a model composed of the direct relationships of two work protective factors, namely the social factor of the LMX and the individual factor of the adjustment to the work mode with exhaustion and turnover intention (hypothesis 1—H1 and hypothesis 2—H2, respectively). But, we also wanted to explore more finely the mechanism that links all the variables considered by investigating whether there is also a relationship between LMX and adjustment to the work mode (hypothesis 3—H3). In fact, the concatenation between these two protective factors, one a facilitator of the other, may better explain the dynamic of the reduction in exhaustion and turnover intention. Therefore, our research question was whether leader–member exchange (LMX) contributes to employees’ adjustment to different work modes and, through this process, reduces exhaustion and turnover intention (hypothesis 4—H4).

The results confirmed the ensemble of our model and all the hypothesised relationships.

Concerning the relationships between the protective factors and our dependent variables, we found that both LMX and adjustment to the work mode are negatively related to both exhaustion and turnover intention, confirming H1 and H2.

Firstly, regarding the role of LMX, our results are in line with previous studies showing that when workers perceive a low quality of LMX, they will likely attribute negative characteristics to their work, feeling unrecognised or undervalued, and reporting higher levels of exhaustion [89]. Conversely, the level and quality of exchange they manage to establish between themselves and their leader can act as ‘affective forces’ [90,91], a protective factor that decreases exhaustion [56]. Furthermore, previous research has shown how a positive exchange relationship with one’s manager makes the worker perceive that they are also supported in terms of organisational support [92,93]. This leads to an increased commitment to the job and to the organisation they belong to and, consequently, a lower intention to leave the company, since it is perceived as being attentive to their needs [94]. According to comparative recent studies, the quality of the exchange between a leader and organisation members acts as a protective factor to exhaustion [95,96] and turnover intention [97].

Secondly, regarding the role of the adjustment to the work mode, according to recent studies, our findings indicate that a better adjustment to the work mode, to the extent to which it allows the worker to be satisfied, productive, and to manage the balance between work and life, acts as protective factor against exhaustion and turnover intention [98]. This converges with multiple studies that, in the last few years, have shown that a worker who feels a better adjustment, hence an achieved correspondence, between themselves and their work environment, will report higher levels of job satisfaction, seeing a reciprocity of expectations, needs, and rewards [99,100]. A worker that is adjusted to their work mode, thus being fulfilled in their needs, and being satisfied, productive, and balanced [9], will feel more secure and self-efficacious, and they will have sufficient resources to feel able to cope with the demands of a work environment potentially stressful, without feeling exhausted. This will consequently encourage them to remain in a work environment that gives awards, and that enhances and involves them positively [65].

Finally, our results are in favour of the existence of a more complex dynamic that links the variables of the study. It emerged that the quality exchange with one’s leader increases the degree of adjustment between the worker and their work mode, thus confirming H3. This emphasises the importance of creating a good working climate and satisfying relationships with one’s leader in order to better adjust to work [65]. This will lead the worker to feel that they have established a trusting relationship with their leader and that they can “move” in the work environment with a certain autonomy [56], which will lead to a better correspondence between their expectations and needs, and their work environment [68]. Consequently, a worker who has found such a correspondence between themselves and the work mode in which their organisation allows them to work will be more resilient and resourceful [8], and thus more protected from exhaustion [57] and less inclined to leave the workplace [58,100]. Therefore, our results underline that one’s relationship with their leader acts as a real ‘antidote’ to work stress and exhaustion [52] via the adjustment to their work mode, because this exchange becomes social and emotional support for coping with a stressful or particularly demanding work environment in terms of work demands, also linked to the need to adapt to new ways of working [39].

To conclude, the findings of the study met the purpose of demonstrating that a positive and sustainable LMX contributes to the worker’s adjustment to their specific work mode and this, in turn, leads to lower exhaustion and lower turnover intention [101,102], confirming H4. Our findings represent a new contribution to the psychology of sustainability and sustainable development which puts at its core people’s wellbeing in different environments [103]. Indeed, this study highlights the sustainable role played by LMX as a facilitator of workers’ adjustment to their specific work modes. Furthermore, by focusing on the construct of adjustment to the work mode, referring back to the initial theorisation of work adjustment [17,41], but set in the New Normal and VUCA era, our study constitutes a novelty compared to previous research on this topic, since it investigates the adjustment to different work modes in a post-crisis period, a broader context studied with multiple and new ways of working [104]. Our findings echo the prior literature that associates work environment fit with positive wellbeing outcomes [65,99,100]. However, this study contributes an additional layer by showing that the adjustment to the work mode buffers against exhaustion in addition to against turnover, a finding that emphasises the resilience-building potential of work mode fit.

This study has several limitations. Firstly, its cross-sectional design limits the possibility to infer the direction of the relationships tested. Additional studies, including those with temporal designs, longitudinal or diary, could be useful to further verify our results. Secondly, the convenience sampling method could generate sample bias since the participants were not chosen for their randomness and representativeness of the population; therefore, this limits the possibility to generalise our results. Future studies using convenience sampling should include the possibility to compare data from a sample to the known data from a population using, for instance, an analysis of the percentage and average variability in order to assess the possible bias and to determine whether a sample represents a population in a research setting. Alternatively, they should use other methods, like probability sampling.

Furthermore, the data collection was self-reported which implies the risk of common method bias. Even if we conducted a Harman’s test, future studies should further reduce the risk of this bias by using, for example, Structural Equation Modelling with method latent variables. Furthermore, future studies could benefit from the inclusion of objective types of measurement, like data collected from other sources in the organisations, for instance, from colleagues or supervisors that could evaluate the employees’ attitudes and performance.

Additionally, since this study investigates work adjustment in the New Normal and VUCA era, it should be noted that different ways of working (onsite, remote, hybrid mode) could correspond to different needs and, therefore, different perception levels of the variables of the study. Our results have shown some differences in LMX and adjustment to the work mode in favour of hybrid workers, highlighting how the hybrid mode better meets the mutual needs between workers and their work environments [105] to the extent that it enables people to work in a more flexible and autonomous way, and thus be more satisfied, productive, and able to balance work and non-work demands [106]. Compared to hybrid workers, the onsite workers of our sample perceived a lower quality of relationship with their leader and reported being less adjusted. This can be explained by the fact that, since the pandemic, all the attention has been put on new ways of working, neglecting the more conventional ways.

However, the findings of this study suggest that LMX and work adjustment are transversal protective factors for wellbeing that act overall positively regardless of the work mode. Albeit, this does not mean that they could not be even more protective against negative wellbeing outcomes depending on the work mode; it seems important to develop additional studies that can explore further possible differences across different groups according to the work mode, testing comparative models using, for instance, multigroup analysis.

Moreover it is worth noting that to official work modes, formally defined, often correspond undeclared work modes, implemented but not officially recognized, which might significantly affect the level of work adjustment. This aspect, not yet addressed in this study, opens up an interesting avenue for future research.

Despite its limitations, this study has important implications for practitioners and organisations.

First of all, organisations could leverage these findings by providing training for leaders on how to become more sustainable by cultivating good-quality relationships with their team members, and fostering a social environment characterised by support and trust. Furthermore, companies should establish regular feedback mechanisms, such as employee surveys or focus groups, to assess their perceptions of wellbeing and identify specific areas for improvement.

Furthermore, as the New Normal has brought with it new ways of working, it is necessary for organisations to be concerned about them and focus on the impact they have on workers and their wellbeing [8]. In fact, some modalities may be more or less suitable for the worker, who may find better or lesser work adjustment in them. Worried by the trends of resignation and quiet quitting, and by the implications of the physical and psychological distance of remote work modes, organisations that adopt flexible works modes during and after the pandemic tend to step backward by depriving people of the possibility to work remotely by cutting down on the number of remote working days allowed or by imposing a number of days of presence at work. However, our study highlights that workers adjust better to flexible work modes. Therefore, in order to facilitate work adjustment, organisations should consider not imposing a work mode, but implementing a dialogue with their workers through the facilitation of exchanges between the managers and employees with the purpose of evaluating, according to tasks and objectives, the work modality that better meets the mutual needs of the work environment and the individuals together.

Considering that employee wellbeing is at the core of concerns of the psychology of sustainability and sustainable development, the prevention of exhaustion and turnover remains a critical issue in the creation of healthy organisations [35]. It is necessary, therefore, for companies to not only be aware of and monitor all the possible causes of illbeing and drop-out behaviour in organisations, but to also identify the resources of the work context that can be activated to create a culture of wellbeing in organisations [107]. Putting attention on the quality of the relationship with one’s employees, therefore, can be a decisive factor for organisational sustainability since it can lead a person to decide whether or not to leave one’s company, and this makes one realise how important it is for managers to also focus on relational aspects, and not only on productivity and performance [37]. Understanding and taking into account the needs of their employees will enable them to ‘fit in’ better in their working environment, allowing them to develop a greater fit with their organisation and to be less likely to leave the company [108].

In this respect, companies could invest in effective prevention and monitoring systems, e.g., by taking concrete care of their employees’ health, by providing medical care, and by supporting training on psychophysical wellbeing and health management in the company [109]. The leader, in addition to being a support figure, should become a sustainable leader by engaging in creating frequent communication moments, not only to monitor the performance and commitment level of their employees, but also to discuss possible problems and dissatisfactions, and try to find solutions [110]. Regular one-on-one meetings between a leader and their team members can provide a platform for open dialogue and problem-solving, fostering a culture of wellbeing and continuous improvement and support—in other words, a culture of sustainability.

6. Conclusions

This study aimed to test the relationships between LMX, adjustment to the work mode, exhaustion, and turnover intention with the objective of verifying a double-chained mediation model linking all these dimensions sequentially. Our results confirmed all the hypothesised relationships, adding new insights to previous findings concerning the role of positive LMX in contributing to the adjustment to the work mode, independent from the specific in-person or hybrid work modality, and in creating a resourceful dynamic that counteracts employees’ exhaustion and turnover intentions [101,102,108].

Building on these findings, organisations can adopt actionable strategies to favour positive leader–member exchanges and to enhance employees’ adjustment to different work modes in order to protect them from exhaustion and dissatisfaction, and to enhance their motivation to stay, such as involving employees in dialogue on the choice of the most fitted work mode based on their tasks and organisational goals; establishing regular systems to monitor employees’ needs and levels of adjustment to their work mode, such as surveys or regular interviews; investing in leadership training in order to support managers in establishing objectives that could be clear, adjusted, and coherent to the employees’ work modes; and encouraging frequent one-on-one meetings between the leaders and their team members to create opportunities to address their needs and concerns.

To conclude, organisations should take advantage of this post-crisis period of settlement to adopt a more humanistic approach to their sustainability efforts, focusing on people and their work-related wellbeing perceptions [111]. If an organisation pays attention to promoting leadership practises oriented to wellbeing and to choosing work modes adjusted to its employees’ needs, it means that the organisation becomes engaged in sustainable development and cares for the quality of working life. Nowadays, in the VUCA era, these engagements do not just represent a necessary investment for organisations that want to develop sustainably, but they are also a part of a successful employer branding strategy, fundamental for retaining and attracting talents and preventing turnover [112,113], and on which the very same sustainability of organisations depends.