Praise and recognition from supervisors buffers employee psychological strain: A two-sample investigation with tourism workers

Abstract

BACKGROUND:

Focusing on employees with psychological strain, this research draws on Fredrickson’s ‘undoing hypothesis’ to examine praise and recognition from one’s supervisor as an organizational resource.

OBJECTIVE:

A model is tested in which psychological strain is a mediator in the positive relationship between role demands and employees’ intentions to take sick leave and seek medical advice, and positions supervisor praise and recognition as a buffer of psychological strain on such intentions.

METHODS:

The model was tested using two Australian samples in the tourism sector, consisting of motel workers (n = 104) and museum workers (n = 168).

RESULTS:

For museum workers, but not motel workers, there was a positive indirect effect of each role demand on sick leave intentions through psychological strain that weakened as a function of supervisor praise and recognition. The proposed moderated mediated model was supported for both samples in regards to intentions to seek medical advice.

CONCLUSIONS:

This research contributes new evidence regarding the antecedents of employees’ intentions to take sick leave and seek medical advice for work stress-related problems. It also contributes to the limited evidence regarding supervisor praise and recognition as a protective factor for employees exhibiting the symptoms of psychological strain.

Keywords

Occupational stress role demands employee strain sick leave medical advice

1 Introduction

The significant impact that adverse working conditions have on employees’ mental health is well-established [1]. Psychological strain related to excessive job demands has deleterious outcomes for the afflicted individual, employers, and the economy as a whole [2]. In examining organizational resources that can assist distressed employees, the present research focuses on praise and recognition from one’s supervisor. Drawing on several tenets of two prominent occupational stress theories: the job demands-resources (JD-R) model [3] and conservation of resources (COR) theory [4, 5], as well as Fredrickson’s [6] broaden-and-build theory, a set of direct, indirect, and conditional predictions are made (see Fig. 1). First, the direct effect of role demands on psychological strain is examined. Second, it is proposed that psychological strain is the mediating mechanism between role demands and two common behavioral responses to experiencing distress: taking sick leave and seeking medical advice. Third, it is proposed that employees who experience psychological strain as a result of high role demands will be less likely to be ruminating about such behaviors when they perceive praise and recognition from their supervisor. To this end, a moderated mediation is proposed with praise and recognition from supervisors as a second-stage moderator and tested using two different Australian samples of employees from the tourism sector (motel and museum workers).

Fig. 1

Proposed moderated mediation model.

In testing this model, we focus on three aspects of role-related stress: role overload, role conflict, and role ambiguity. Role overload occurs when an employee feels pressured by excessive workloads, difficult deadlines, and is unable to fulfill organizational expectations in the time available [7]. Role conflict reflects the degree to which an employee is expected to simultaneously perform two or more mutually exclusive tasks, instructions, or requests, and has been described as incompatible demands and expectations placed on an employee [8]. Role ambiguity refers to a lack of clarity or uncertainty with respect to job responsibilities, the perceived lack of important job-related information, or unclear or constantly changing specifications regarding expectations and duties defining the employee’s job [9]. According to the JD-R model [3], chronic job demands require employees to expend continual physical or psychological effort that results in adverse employee outcomes. In support of the JD-R model and the ‘health impairment process’, substantial meta-analytic research indicates that hindrances in the form of these role stressors results in psychological strain [10] and job demands predict both cross-sectional and longitudinal job burnout [11].

Tourism has been identified as a stressful work environment for employees [12]. Ariza-Montes et al. [12] note that this sector is unique due to its high level of labor turnover; labor and skill shortages; high proportion of seasonal, part-time, and on-call workforce; and poor conditions of employment. Hotel and motel employees are subjected to intense time pressures, a demanding work tempo, and long and unsociable work hours (see Zhao and Ghiselli [13] for a review of hospitality job characteristics). Role conflict and role ambiguity are common too, because hotel employees are often required to deal with demanding guests and supervisors, and follow complex procedures [13]. Several questionnaire studies support the positive association between role demands and poor psychological well-being in hotel and hospitality employees (e.g., Ariza-Montes et al. [12]). And more recent research, in light of the COVID-19 pandemic, has demonstrated that hotel employees now face even more unstable employment, demanding hygiene policies, and unethical labor practices, in addition to their traditonal hotel-work stressors [14].

Empirical studies also support the prevalence of role demands for museum employees. For example, earlier qualitative research has found that having too much to do and conflicting tasks, and ambiguous demands in the nature of the job role, were all problems for UK museum personnel [15]. Subsequent interview studies have noted pressure related to managing multiple and time-intensive projects [16, 17] and balancing visitors’ high expectations with the reality of historical facts, as a form of role conflict [18]. Questionnaire studies also have demonstrated a positive association between role overload and psychological strain in museum volunteers [19]. As such, the first hypothesis aims to confirm that:

H1. Role demands will have positive associations with psychological strain.

1.1 Psychological strain as a mediator

The JD-R model further proposes that job demands lead –via psychological strain –to undesirable employee behavioral responses [20]. In relation to sickness absence, the JD-R stress-reaction hypothesis proposes that such behavior is a reaction to distress caused by a failure to cope with job demands. Absenteeism is thus a coping mechanism to manage psychological strain caused by stressful job demands [20]. COR theory [4, 5] provides an additional theoretical framework for understanding the relationship between psychological strain and subsequent employee withdrawal. Individuals strive to obtain and retain resources, which include material objects, social conditions, personal characteristics, and energies that are valued by the individual. Psychological strain occurs when individuals experience depletion of resources. As such, psychological strain is a manifestation of when resources are inadequate to meet job demands, and employees suffering from such resource loss will attempt to conserve resources and prevent further resource loss by withdrawing from resource-depleting situations. Accordingly, distressed employees may take sick leave in an attempt to replenish depleted resources and prevent further resource losses.

There is empirical evidence for the positive association between psychological strain and sickness absence. For instance, a meta-analysis of 153 studies found positive associations for a composite indicator of psychological illness with the occurrence and duration of absenteeism [21]. In a sample of 519 Brazilian elementary school teachers, Coledam and da Silva [22] found that teachers with common mental health problems had a higher rate ratio of health-related absenteeism over the previous 12 months. Gohar et al. [23] used interviews to identify four main variables that predicted sickness absence in nurses and personal support workers, one of which was the emotional toll on mental well-being. In addition to taking sick leave, employees may seek medical advice for work-stress related problems in an attempt to replenish depleted resources. For example, qualitative research with 45 employees from a range of different occupations identified seeking medical advice as a coping method when under stress resulting from work [24]. In a sample of 8,070 Dutch employees, Andrea et al. [25] found that experiencing mental health problems predicted future occupational physician and general practitioner visits. Similarly, in a nationally representative sample of the Canadian population, high and medium job strain was associated with the number of visits to both a general practitioner (GP) and a specialist, with the number of GP visits up to 26%more for employees with high strain compared to low strain [26].

Medical advice seeking may be particularly salient for Australian tourism employees because the majority of the Australian tourism workforce (just under two thirds) are part-time or casual employees [27] and in the Australian context, casual employees are not eligible for paid sick leave [28]. In addition, as in most Western countries, Australian GPs have a central gatekeeping function for their employed patients who are seeking workers’ compensation. In Australia, work-related mental health conditions are the second most common workers’ compensation disease claim, after injury and musculoskeletal disorders [29]. To make a workers’ compensation claim in Australia, an initial certificate of capacity completed by a medical practitioner must be filed. Thus, although jurisdictional variation exists throughout Australian states and territories, GPs are almost always the first point of contact with the healthcare system, and are the main gatekeepers to compensable benefit entitlements. As such, the following is hypothesized:

H2. Psychological strain will have a positive association with sick leave intentions and medical advice intentions.

The theoretical underpinnings of both the JD-R model and COR theory also provide the grounds for hypothesizing that psychological strain therefore mediates the relationship between job demands and such intentions, and has been examined in relation to absenteeism. For example, results of Darr and John’s [21] meta-analysis demonstrated that psychological illness mediated the relationship between work strain (an individual’s subjective evaluation of their work as harmful or threatening due to work-related factors) and absenteeism. Bakker et al. [30] reported that job burnout mediated the positive effects of workload and reorganization on absence duration for industrial workers. Schaufeli et al. [20] found increases in job demands for Dutch managers predicted their job burnout over a 1-year period, which went on to predict future absence duration. Vignoli et al. [31] found support for a positive indirect effect relationship of job demands (in this case, operationalized as perceived effort) on emotional exhaustion that, in turn, predicted absence duration for retail employees, finding no direct relationship between job demands and absenteeism. Thus, the following is hypothesized:

H3. Role demands will have positive indirect effects on sick leave intentions and medical advice intentions through psychological strain.

1.2 Supervisor praise and recognition

Employee recognition has been defined as “the assignment of personal non-monetary rewards (i.e., interest, approval, and appreciation) for individual efforts and work accomplishment to recognize and reinforce the desired behaviors displayed by an employee” [32 p. 356]. Giving praise and recognition is a behavioral skill of supportive managers [33] that also is captured in the notion of health-promoting leadership [34] and the individualized consideration component of transformational leadership [35]. Praise and recognition is one of three main characteristics of humble leaders [36] and also a form of appreciative leadership [37, 38]. In this respect, Stocker et al. [37] used employee verbal descriptions to identify six categories of appreciative acts in the work context that included praise and gratitude, which was found to be the most frequent type. Praise and recognition also is aligned with the ‘esteem’ component of reward in Siegrist’s Effort-Reward Imbalance Model [39]. Indeed, van Vegchel et al. [40] demonstrated that esteem, in the form of feeling appreciated, was the strongest of rewards (compared to remuneration and secure employment) for reduced risk of physical and psychosomatic complaints and emotional exhaustion for healthcare workers.

Empirical evidence demonstrating the importance of praise and recognition and related constructs for employees is well-established. For example, studies using broader measures of good or supportive leadership that include items about appreciating employees are associated with lower job strain, as found in a sample of Canadian 247 child welfare agency employees [41], and less sickness absence due to overstrain and fatigue in Finnish blue- and white-collar workers and Swedish blue-collar workers, all recruited from a forestry organization [42]. Supervisor appreciation has been demonstrated to predict lower depression in a sample of 194 German employees [43], and lack of acknowledgement from superiors predicts long-term (more than one month) sick leave among French private-sector employees [44]. In their meta-analysis examining the association of psychosocial stressors at work with sickness absence due to a diagnosed mental disorder, Duchaine et al. [45] found workers exposed to low reward (3 studies; N = 38,659) were at a higher risk of sickness absence, compared to non-exposed workers.

1.3 Supervisor praise and recognition and the role of positive emotions

There are several theoretical explanations behind the benefits of praise and recognition for employees, pointing to the role of positive emotions. As discussed by Boudens [46], a positive response to one’s work by another person is required in order for employees to experience positive feelings for a job well done. According to the Stress-as-Offence-to-Self framework [47], appreciation protects employees’ social esteem and boosts positive emotions. Supervisor behavior also can be construed as an affective event in the workplace [48]. As such, supervisors can be viewed as sources of uplifts if positive emotions are evoked [49]. Effective leaders are those who provide regular small uplifts that serve to mitigate hassles [49], and thus increase the likelihood that employees will engage in positive workplace behaviors [48].

Event-based studies using experience sampling designs have demonstrated that receiving praise and recognition generates positive emotions in employees. In this regard, Ohly and Schmitt [50] found that praise/appreciation/positive feedback was the second most frequent positive affective work event, and was related to higher levels of enthusiasm and feeling at rest, after controlling for trait positive affect. In hotel employees, Basch and Fisher [51] found receiving recognition was related to feelings of pleasure, happiness, pride, enthusiasm, and affection. Arguing that praise and recognition facilitates a positive state of mind, Stocker et al. [37] demonstrated that small appreciative acts predicted the extent to which employees (from a range of occupations) felt serene at the end of their work shift. In a sample of 200 full-time German employees, Wang et al. [52] found that recognition events predicted positive affect and employee engagement. Studies using longer time-lags also have shown that appreciation predicts higher levels of enthusiasm and contentment [53], and feelings of success [54] in Swiss employees.

1.4 Supervisor praise and recognition as a moderator

Drawing on Fredrickson’s [6] broaden-and-build theory, it is proposed that supervisor praise and recognition will weaken the positive association between psychological strain and intentions to take sick leave and seek medical advice due to the various adaptive functions of positive emotions. According to broaden-and-build theory, positive emotions can ‘undo’ the adverse effects of negative emotions. In support of the ‘undoing hypothesis’, experimental studies have shown that positive affect can downregulate the negative physiological impact of anxiety and fear on the cardiovascular system [55, 56]. Overall, accumulating research suggests that targeting the positive affect system may be an important, yet unexplored, avenue in the treatment of anxiety and depression [57]. For example, using a short-term prospective design with university students, Riskind et al. [58] found that positive affect buffered against the effects of negative affect on gains in depression symptoms six weeks later.

Our proposition is further supported by the ‘broaden’ aspect of Fredrickson’s [6] broaden-and-build theory, as positive emotions are argued to broaden an individual’s immediate behavioral tendencies or ‘momentary thought-action repertoires’. Negative emotions call forth specific action tendencies, such as fight or flee. In contrast, positive emotions guide people towards a wider range of cognitions (see Isen [59] for a review) and actions [60] than is typical, and function as internal signals for individuals to approach or continue to engage with their environments and participate in activities [6]. As such, positive emotions instigate the use of adaptive coping strategies, such as problem-solving, positive reframing, and seeking counsel, rather than maladaptive and avoidant coping strategies [61]. The role of the supervisor in employees’ use of coping strategies has been demonstrated by Stare et al. [62] who found that individualized consideration predicted positive reappraisal and planful problem-solving, as well as less escapist and avoidant coping in a sample of employees in IT organizations in Slovenia.

In light of this past research, praise and recognition from supervisors should buffer the negative effects of psychological strain. However, the extant literature examining the interaction between psychological strain and appreciative supervisor behavior on sickness absence is inconsistent. For example, Väänänen et al. [63] found supervisor support moderated the effect of psychological strain on risk of long-term sickness absences among male, but not female employees working for a multinational forest industry enterprise in Finland. Munir et al. [64] found that quality of leadership did not significantly interact with employee depressive symptoms in the prediction of 2-year incidence of long-term sickness absence in a representative sample of the Danish workforce. Of interest, follow-up stratified analyses showed that the effect size for leadership was significant for those with moderate depression, but not for the groups classified as free of depression or severe clinical depression, suggesting that leadership was protective up to a certain point of depression. In a study of female Danish eldercare workers, Hjarsbech et al. [65] found no significant interaction of depressive symptoms and psychosocial working conditions (that included high quality of leadership) on risk of long-term sickness absence.

It is argued that these previous studies have a number of characteristics that are distinct from the present research. First, although praise and recognition has been investigated as a construct in its own right in relation to direct effects, investigations into its moderating effects have examined aggregated measures of leadership behavior, and have not focused on supervisors’ specific appreciative behaviors such as giving praise and recognition. Indeed, as noted by Pfister and colleagues, appreciation tends to be examined in the context of broader leadership constructs and not in isolation [54, 53]. Second, these studies have focused on long-term sickness absence, a distal outcome variable with other preceding conditions. Third, these studies have not examined psychological strain in the context of job or role demands driving the health-impairment process. Psychological strain has a multifactorial etiology and can be triggered by a wide range of psychosocial stressors outside of the work domain [66]. It is proposed that it is when psychological strain is demonstrated to arise from the work domain that praise and recognition from the supervisor will weaken employees’ intentions to take sick leave for stress-related problems, thus leading to the final hypothesis:

H4. Praise and recognition will weaken the positive indirect effects of role demands on sick leave intentions and medical advice intentions through psychological strain because of its stress-buffering influence on the relationship between psychological strain and intentions.

2 Method

2.1 Sample 1

The participating organizations were eight motels located in Queensland and New South Wales. In the Australian context, motels are establishments with 15 or more rooms and are targeted to motorists, with car parking provided [67]. The motels were three- or four-star motels with a workforce average of 21.88 employees per motel (SD = 15.82; range = 8 to 57 employees). The researchers contacted the motel managers, and all accepted the invitation to collaborate with the researchers. All frontline and back-of-house employees (175 employees) were then invited to participate and supplied with a paper questionnaire to complete in their own time, independent of their employer and the researchers, along with a postage-paid envelope addressed direct to the researchers. A total of 111 surveys were returned, yielding an overall response rate of 63.4%(range was 30%to 100%across the eight motels). Two cases were removed: one identified through Mahalanobis distance as a multivariate outlier due to an unusual pattern of responding and one who did not have a supervisor. The final sample of 109 comprised participants from the front office (36.7%), housekeeping (35.8%), kitchen (12.8%), restaurant (11.9%), and maintenance (2.8%). The sample consisted of 83 males (76.1%) and 27 females (24.8%). Age ranged from 17 to 72 years (M = 37.45; SD = 14.74). The majority of employees were employed on a casual basis (60.6%), with 31.2%full-timers and 8.3%part-timers. Participants worked an average of 28.61 hours per week (SD = 15.56). Average organizational tenure was 2.05 years (SD = 2.55). Average tenure with current supervisor was 2.05 years (SD = 2.32). The University Human Research Ethics Committee of The University of Queensland approved the research.

2.2 Sample 2

The participating organization was a museum that oversees a number of tourist attractions in Australia. An online questionnaire was distributed to all 300 employees. A total of 190 responses were returned, giving a response rate of 63.3%. Five cases were removed due to being high-level managers with no supervisor. Thus, the final sample size was 185. The sample consisted of 105 women (56.8%) and 69 men (37.3%), with 11 participants (5.9%) not indicating their gender. Age ranged from 20 to 65 years (M = 43.96; SD = 11.95). Most were employed full-time (63.2%), with the remaining employed on a part-time (17.8%) or casual (13.0%) basis (missing = 5.9%). Participants worked an average of 34.17 hours per week (SD = 13.12). Average organizational tenure was 9.11 years (SD = 8.52). Participants had been working with their current supervisor for an average of 3.70 years (SD = 3.62). Occupations included: marketing, accounting, IT, publishing, product development, learning officers, librarians, and donor relations (24.3%); customer service officers (23.8%); curators and collections personnel (14.6%); maintenance, facilities, and technical officers (8.6%); managers (6.5%); researchers (4.9%); administrators (4.9%); and retail assistants (3.2%), with 9.2%not reporting a specific occupational role. The University Human Research Ethics Committee of Queensland University of Technology approved the research.

2.3 Measures

Role overload was measured using four items from the Demands subscale of the UK’s Health and Safety Executive (HSE) Indicator Tool [68], rated from 1 (strongly disagree) to 7 (strongly agree). A sample item is: “I have unrealistic time pressures.”

Role conflict was measured using items (Sample 1: four items; Sample 2: five items) from the perceived work characteristics scale developed by Haynes et al. [69], rated from 1 (strongly disagree) to 7 (strongly agree). A sample item is: “Different people at work expect conflicting things from me.”

Role ambiguity was measured with items (Sample 1: four items; Sample 2: five items) from the Role subscale of the UK’s HSE Indicator Tool [68], rated from 1 (strongly disagree) to 7 (strongly agree). A sample item is: “I am clear about what is expected of me at work.” All items were reverse-scored.

Psychological strain was measured using the 12-item General Health Questionnaire [70]. Respondents rated how they had generally been feeling over the past four weeks, ranging from 1 (never) to 7 (always). A sample item is: “How often have you been feeling unhappy or depressed?” Six positively worded items were reverse-scored.

Praise and recognition from one’s supervisor was measured using three items rated from 1 (strongly disagree) to 7 (strongly agree). Employees were asked to rate their supervisor in relation to whether they felt their supervisor: encouraged them with praise and thanks, valued their contributions to the organization, and gave them sufficient credit for their hard work. For Sample 2, two extra items were included concerning the supervisor appreciating their work at the organization and recognizing their accomplishments. Items were taken from various praise and recognition scales [71 , 40].

Sick leave intentions were measured with one item asking participants if they seriously believed that, in the near future, they would take sick leave for a stress-related problem, rated from 1 (extremely unlikely) to 7 (extremely likely).

Medical advice intentions were measured with one item asking participants if they seriously believed that, in the near future, they would seek medical advice for a stress-related problem, rated from 1 (extremely unlikely) to 7 (extremely likely).

3 Results

3.1 Data analysis overview

Table 1 displays the descriptive statistics, reliabilities, and correlations for both samples. Hayes’ (2018) PROCESS (Version 3.4) [74] macro for SPSS was applied to each of the three role demands in separate models. First, the hypothesized direct and indirect effects (H1–H3) were examined through simple mediation (see Tables 2 3). In a subsequent analysis, the moderator variable was integrated into the model so that Stage 2 moderated mediation (H4) could be tested (see Tables 4 5), along with the associated conditional indirect effects (see Tables 6 7). All analyses utilized the percentile method for bootstrapping with 10,000 resamples to determine 95%confidence intervals (CIs). According to Yzerbyt et al. [75], the percentile bootstrap method offers good performance against both Type I and Type II error. Gender and age were included as control variables. Due to listwise deletion, the useable sample size for the PROCESS analyses was 104 for Sample 1 and 168 for Sample 2.

Table 1
Means, standard deviations, reliabilities, and correlations

Variable Sample 1 Sample 2 1 2 3 4 5 6 7 8 9

M SD α M SD α

1. Gender^a – – – – – – –0.09 –0.03 –0.10 0.01 –0.05 0.04 0.20^* 0.01

2. Age 37.56 14.72 – 43.96 11.95 – –0.27^* 0.03 –0.14 –0.31^ –0.38^* 0.13 –0.13 –0.05

3. Role overload 2.32 1.24 0.92 3.42 1.49 0.89 –0.12 0.07 0.52^* 0.19^* 0.40^* –0.39^* 0.01 0.47^*

4. Role conflict 2.41 1.23 0.92 3.70 1.51 0.92 –0.12 0.02 0.59^* 0.33^ 0.38^* –0.55^*** 0.21^* 0.51^*

5. Role ambiguity 1.51 0.82 0.90 2.31 1.03 0.85 –0.12 –0.02 0.36^* 0.45^* 0.35^* –0.40^* 0.08 0.49^*

6. Psychological strain 2.57 0.98 0.88 3.00 1.16 0.93 –0.12 –0.13 0.50^* 0.55^* 0.60^* –0.40^* 0.18 0.69^*

7. Praise and recognition 5.58 1.28 0.97 5.04 1.77 0.97 0.03 0.03 –0.23^ –0.41^* –0.53^* –0.56^*** –0.24^* –0.49^*

8. Sick leave intentions 1.35^b 0.76 – 2.30^d 1.67 – –0.03 –0.04 0.35^* 0.46^* 0.47^* 0.72^* –0.53^* 0.76^*

9. Medical advice intentions 1.42^c 0.89 – 2.51^e 1.87 – 0.21 –0.19 0.15 0.26^ 0.15 0.37^*** –0.25^* 0.57^***

Variable	Sample 1	Sample 2	1	2	3	4	5	6	7	8	9
1. Gender^a	–	–	–	–	–	–		–0.09	–0.03	–0.10	0.01	–0.05	0.04	0.20^*	0.01
2. Age	37.56	14.72	–	43.96	11.95	–	–0.27^***		0.03	–0.14	–0.31^**	–0.38^***	0.13	–0.13	–0.05
3. Role overload	2.32	1.24	0.92	3.42	1.49	0.89	–0.12	0.07		0.52^***	0.19^*	0.40^***	–0.39^***	0.01	0.47^***
4. Role conflict	2.41	1.23	0.92	3.70	1.51	0.92	–0.12	0.02	0.59^***		0.33^**	0.38^***	–0.55^***	0.21^*	0.51^***
5. Role ambiguity	1.51	0.82	0.90	2.31	1.03	0.85	–0.12	–0.02	0.36^***	0.45^***		0.35^***	–0.40^***	0.08	0.49^***
6. Psychological strain	2.57	0.98	0.88	3.00	1.16	0.93	–0.12	–0.13	0.50^***	0.55^***	0.60^***		–0.40^***	0.18	0.69^***
7. Praise and recognition	5.58	1.28	0.97	5.04	1.77	0.97	0.03	0.03	–0.23^**	–0.41^***	–0.53^***	–0.56^***		–0.24^*	–0.49^***
8. Sick leave intentions	1.35^b	0.76	–	2.30^d	1.67	–	–0.03	–0.04	0.35^***	0.46^***	0.47^***	0.72^***	–0.53^***		0.76^***
9. Medical advice intentions	1.42^c	0.89	–	2.51^e	1.87	–	0.21	–0.19	0.15	0.26^**	0.15	0.37^***	–0.25^*	0.57^***

Notes: Correlations for Sample 1 motel workers (n = 109) are presented above diagonal; correlations for Sample 2 museum employees (n = 185) are presented below diagonal; ^aGender coded as –1 = male, 1 = female; ^bRange = 1–4; ^cRange = 1–5. ^dRange = 1–7; ^eRange = 1–7. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

Table 2

Simple mediation results for Sample 1 motel workers (n = 104)

	Model 1: Psychological strain		Model 2a: Sick leave intentions		Model 2b: Medical advice intentions
Role overload	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.29 (0.67)	13.70^***	0.08 (0.55)	2.18	0.19 (0.70)	5.83^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.09 (0.09)	–0.26; 0.10	0.18 (0.11)	–0.02; 0.40	0.24 (0.12)	0.02; 0.48
Age	–0.03 (0.01)	–0.04; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role overload	0.34 (0.08)	0.19; 0.49	–0.05 (0.07)	–0.19; 0.10	–0.03 (0.10)	–0.21; 0.18
PS			0.16 (0.09)	–0.01; 0.35	0.37 (0.13)	0.11; 0.61
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.06 (0.04)	–0.002; 0.14	0.12 (0.05)	0.03; 0.23
Role conflict	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.25 (0.71)	11.29^***	0.11 (0.53)	2.99^*	0.21 (0.68)	6.68^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.06 (0.09)	–0.23; 0.11	0.19 (0.10)	–0.001; 0.41	0.25 (0.12)	0.03; 0.48
Age	–0.02 (0.01)	–0.03; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role conflict	0.26 (0.08)	0.10; 0.41	0.12 (0.06)	0.01; 0.24	0.12 (0.09)	–0.05; 0.29
PS			0.08 (0.08)	–0.06; 0.24	0.30 (0.11)	0.08; 0.52
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.02 (0.02)	–0.01; 0.08	0.08 (0.04)	0.02; 0.16
Role ambiguity	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.21 (0.75)	8.83^***	0.08 (0.55)	2.07	0.19 (0.70)	5.80^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.09 (0.09)	–0.27; 0.09	0.18 (0.10)	0.02; 0.39	0.24 (0.12)	0.01; 0.48
Age	–0.02 (0.01)	–0.03; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role ambiguity	0.37 (0.14)	0.11; 0.68	0.00 (0.14)	–0.27; 0.29	0.01 (0.15)	–0.25; 0.34
PS			0.14 (0.08)	–0.01; 0.29	0.35 (0.12)	0.11; 0.59
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.05 (0.04)	–0.003; 0.14	0.13 (0.07)	0.02; 0.29

Notes: PS = Psychological Strain; MSE = Mean Squared Error; B = Unstandardized Regression Coefficient; SE = Standard Error; CI = Confidence Interval; BSE = Boot Standard Error. 95%CI is significant for results in bold. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

Table 3

Simple mediation results for Sample 2 museum workers (n = 168)

	Model 1: Psychological strain		Model 2a: Sick leave intentions		Model 2b: Medical advice intentions
Role overload	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.28 (1.01)	21.65^***	0.55 (1.32)	49.91^***	0.53 (1.71)	45.37^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.27 (0.17)	–0.61; 0.07	0.28 (0.19)	–0.09; 0.65	0.48 (0.22)	0.04; 0.89
Age	–0.02 (0.01)	–0.03; –0.01	0.01 (0.01)	–0.002; 0.02	0.01 (0.01)	–0.01; 0.02
Role overload	0.39 (0.05)	0.29; 0.49	–0.04 (0.08)	–0.20; 0.11	0.19 (0.09)	0.01; 0.36
PS			1.11 (0.09)	0.94; 1.29	1.04 (0.10)	0.84; 1.23
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.43 (0.07)	0.30; 0.57	0.40 (0.07)	0.28; 0.54
Role conflict	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.33 (0.95)	26.49^***	0.55 (1.31)	50.66^***	0.53 (1.69)	46.66^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.24 (0.16)	–0.55; 0.09	0.29 (0.19)	–0.08; 0.66	0.49 (0.21)	0.07; 0.91
Age	–0.02 (0.01)	–0.03; –0.003	0.01 (0.01)	–0.004; 0.02	0.01 (0.01)	–0.01; 0.02
Role conflict	0.41 (0.05)	0.31; 0.52	0.09 (0.06)	–0.03; 0.21	0.23 (0.08)	0.08; 0.38
PS			1.02 (0.09)	0.84; 1.19	1.00 (0.10)	0.81; 1.18
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.42 (0.07)	0.30; 0.55	0.41 (0.07)	0.29; 0.55
Role ambiguity	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.42 (0.81)	40.28^***	0.55 (1.32)	50.01^***	0.52 (1.72)	44.86^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.22 (0.15)	–0.51; 0.07	0.28 (0.19)	–0.08; 0.66	0.47 (0.22)	0.03; 0.90
Age	–0.01 (0.01)	–0.03; –0.001	0.01 (0.01)	–0.004; 0.02	0.01 (0.01)	–0.01; 0.03
Role ambiguity	0.71 (0.07)	0.57; 0.87	0.08 (0.14)	–0.18; 0.35	0.28 (0.16)	–0.02; 0.61
PS			1.04 (0.11)	0.81; 1.25	1.01 (0.11)	0.77; 1.21
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
Indirect effect			0.74 (0.11)	0.54; 0.98	0.72 (0.11)	0.51; 0.95

Table 4

Moderated mediation results for Sample 1 motel workers (n = 104)

	Model 1: Psychological strain		Model 2a: Sick leave intentions		Model 2b: Medical advice intentions
Role overload	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.29 (0.67)	13.70^***	0.13 (0.53)	2.39^*	0.30 (0.62)	6.93^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.09 (0.09)	–0.26; 0.09	0.18 (0.10)	–0.02; 0.39	0.22 (0.11)	0.005; 0.45
Age	–0.03 (0.01)	–0.04; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role overload	0.34 (0.08)	0.18; 0.49	–0.08 (0.08)	–0.23; 0.07	–0.05 (0.09)	–0.20; 0.13
PS			0.10 (0.08)	–0.05; 0.27	0.32 (0.12)	0.10; 0.55
PR			–0.11 (0.07)	–0.26; 0.02	0.02 (0.08)	–0.13; 0.17
PS×PR			–0.05 (0.06)	–0.17; 0.05	–0.20 (0.07)	–0.30; 0.001
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.02 (0.02)	–0.06; 0.02	–0.07 (0.03)	–0.12; 0.001
Role conflict	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.25 (0.71)	11.29^***	0.13 (0.53)	2.39^*	0.31 (0.61)	7.20^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.06 (0.09)	–0.23; 0.11	0.18 (0.10)	–0.01; 0.40	0.23 (0.11)	0.01; 0.45
Age	–0.02 (0.01)	–0.03; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role conflict	0.26 (0.08)	0.10; 0.41	0.08 (0.08)	–0.06; 0.24	0.09 (0.08)	–0.06; 0.26
PS			0.06 (0.07)	–0.09; 0.21	0.28 (0.11)	0.08; 0.51
PR			–0.06 (0.08)	–0.22; 0.09	0.07 (0.10)	–0.11; 0.26
PS×PR			–0.05 (0.06)	–0.18; 0.04	–0.20 (0.07)	–0.29; –0.02
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.01 (0.02)	–0.05; 0.01	–0.05 (0.02)	–0.10; –0.003
Role ambiguity	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.21 (0.75)	8.83^***	0.12 (0.54)	2.26^*	0.30 (0.62)	6.89^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.09 (0.09)	–0.27; 0.09	0.18 (0.10)	–0.02; 0.39	.22 (0.11)	.01; 0.44
Age	–0.02 (0.01)	–0.03; –0.01	–0.00 (0.00)	–0.01; 0.01	–0.00 (0.01)	–0.01; 0.01
Role ambiguity	0.37 (0.14)	0.11; 0.68	–0.08 (0.14)	–0.35; 0.21	–0.05 (0.16)	–0.33; 0.32
PS			0.08 (0.07)	–0.06; 0.23	0.31 (0.11)	0.09; 0.52
PR			–0.11 (0.06)	–0.24; 0.01	0.02 (0.09)	–0.14; 0.20
PS×PR			–0.05 (0.06)	–0.18; 0.05	–0.20 (0.07)	–0.30; –0.01
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.02 (0.03)	–0.09; 0.02	–0.08 (0.04)	–0.16; –0.001

Notes: PS = Psychological Strain; PR = Praise and Recognition; MSE = Mean Squared Error; B = Unstandardized Regression Coefficient; SE = Standard Error; CI = Confidence Interval; IMM = Index of Moderated Mediation; BSE = Boot Standard Error. 95%CI is significant for results in bold. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Table 5

Moderated mediation results for Sample 2 museum workers (n = 168)

	Model 1: Psychological strain		Model 2a: Sick leave intentions		Model 2b: Medical advice intentions
Role overload	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.28 (1.01)	21.65^***	0.60 (1.17)	41.06^***	0.56 (1.62)	33.77^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.27 (0.18)	–0.61; 0.08	0.29 (0.19)	–0.07; 0.65	0.47 (0.21)	0.06; 0.89
Age	–0.02 (0.01)	–0.03; –0.004	0.01 (0.01)	–0.001; 0.02	0.01 (0.01)	–0.01; 0.02
Role overload	0.39 (0.05)	0.29; 0.49	–0.02 (0.07)	–0.17; 0.13	0.21 (0.09)	0.06; 0.36
PS			0.90 (0.10)	0.70; 1.09	0.85 (0.12)	0.62; 1.08
PR			–0.12 (0.06)	–0.23; –0.01	–0.13 (0.08)	–0.29; 0.02
PS×PR			–0.12 (0.03)	–0.18; –0.07	–0.09 (0.04)	–0.16; –0.02
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.05 (0.01)	–0.07; –0.02	–0.03 (0.01)	–0.06; –0.01
Role conflict	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.33 (0.95)	26.49^***	0.61 (1.17)	41.49^***	0.56 (1.62)	33.73^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.24 (0.16)	–0.55; 0.10	0.30 (0.19)	–0.06; 0.67	0.49 (0.22)	0.07; 0.91
Age	–0.02 (0.01)	–0.03; –0.003	0.01 (0.01)	–0.002; 0.02	0.01 (0.01)	–0.01; 0.02
Role conflict	0.41 (0.05)	0.31; 0.52	0.07 (0.06)	–0.05; 0.19	0.21 (0.08)	0.06; 0.36
PS			0.85 (0.09)	0.67; 1.02	0.87 (0.11)	0.63; 1.10
PR			–0.11 (0.06)	–0.23; –0.002	–0.09 (0.08)	–0.25; 0.07
PS×PR			–0.13 (0.03)	–0.18; –0.07	–0.09 (0.04)	–0.16; –0.02
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.05 (0.01)	–0.08; –0.03	–0.04 (0.01)	–0.07; –0.01
Role ambiguity	R² (MSE)	F	R² (MSE)	F	R² (MSE)	F
	0.42 (0.81)	40.28^***	0.60 (1.17)	41.05^***	0.54 (1.67)	32.00^***
	B (SE)	95%CI	B (SE)	95%CI	B (SE)	95%CI
Gender	–0.22 (0.15)	–0.52; 0.08	0.29 (0.19)	–0.08; 0.66	0.47 (0.22)	0.03; 0.90
Age	–0.01 (0.01)	–0.03; –0.002	0.01 (0.01)	–0.002; 0.02	0.01 (0.01)	–0.01; 0.03
Role ambiguity	0.71 (0.08)	0.57; 0.87	–0.03 (0.13)	–0.27; 0.23	0.21 (0.16)	–0.09; 0.53
PS			0.90 (0.10)	0.69; 1.10	0.91 (0.13)	0.64; 1.14
PR			–0.12 (0.06)	–0.24; –0.01	–0.08 (0.08)	–0.24; 0.07
PS×PR			–0.12 (0.03)	–0.18; –0.07	–0.08 (0.04)	–0.16; –0.01
			Effect (BSE)	95%CI	Effect (BSE)	95%CI
IMM			–0.09 (0.02)	–0.13; –0.05	–0.06 (0.03)	–0.11; –0.004

Table 6

Conditional indirect effects of role demands on sick leave intentions through psychological strain at three values of praise and recognition for Sample 2 museum workers (n = 168)

Role overload	Effect (se)	95%CI
Praise and recognition 16th percentile	0.45 (0.07)	0.32; 0.59
Praise and recognition 50th percentile	0.32 (0.06)	0.21; 0.44
Praise and recognition 84th percentile	0.25 (0.06)	0.15; 0.38
Role conflict	Effect (se)	95%CI
Praise and recognition 16th percentile	0.45 (0.07)	0.32; 0.58
Praise and recognition 50th percentile	0.31 (0.05)	0.22; 0.42
Praise and recognition 84th percentile	0.25 (0.05)	0.14; 0.36
Role ambiguity	Effect (se)	95%CI
Praise and recognition 16th percentile	0.82 (0.11)	0.61; 1.06
Praise and recognition 50th percentile	0.58 (0.10)	0.40; 0.79
Praise and recognition 84th percentile	0.47 (0.10)	0.27; 0.69

Notes: se = Standard Error; M = Mean; SD = Standard Deviation; CI = Confidence Interval. 95%CI is significant for results in bold.

Table 7

Conditional indirect effects of role demands on medical advice intentions through psychological strain at three values of praise and recognition for Sample 1 motel workers (n = 104) and Sample 2 museum workers (n = 168)

	Sample 1		Sample 2
Role overload	Effect (se)	95%CI	Effect (se)	95%CI
Praise and recognition 16th percentile	–	–	0.40 (0.07)	0.27; 0.54
Praise and recognition 50th percentile	–	–	0.31 (0.06)	0.20; 0.44
Praise and recognition 84th percentile	–	–	0.26 (0.07)	0.14; 0.41
Role conflict	Effect (se)	95%CI	Effect (se)	95%CI
Praise and recognition 16th percentile	0.13 (0.06)	0.03; 0.25	0.43 (0.07)	0.30; 0.58
Praise and recognition 50th percentile	0.05 (0.03)	0.02; 0.14	0.33 (0.06)	0.22; 0.47
Praise and recognition 84th percentile	0.00 (0.04)	–0.06; 0.08	0.29 (0.07)	0.16; 0.43
Role ambiguity	Effect (se)	95%CI	Effect (se)	95%CI
Praise and recognition 16th percentile	0.20 (0.09)	0.03; 0.40	0.77 (0.13)	0.53; 1.03
Praise and recognition 50th percentile	0.09 (0.06)	0.004; 0.22	0.61 (0.11)	0.39; 0.85
Praise and recognition 84th percentile	0.01 (0.06)	–0.09; 0.15	0.53 (0.12)	0.29; 0.79

Notes: se = Standard Error; M = Mean; SD = Standard Deviation; CI = Confidence Interval. 95%CI is significant for results in bold.

In regards to Sample 1, because data were from eight different motels, variance components analyses were conducted to investigate the presence of group-level effects of having employees nested within motels. All intra-class correlations explained less than 5%of the variance at the motel level, and none achieved statistical significance. Design effects, which account for within-group sample size [76], were all less than 2, reinforcing that the influence of clustering need not be controlled for in the analyses.

3.2 Simple mediation results

Direct effects (Hypotheses 1 and 2). For Sample 1, as per H1, role overload, role conflict, and role ambiguity each had significant positive associations with psychological strain (see Table 2, model 1). For Sample 2, the role demands each had significant positive direct effects on psychological strain, supporting H1 (see Table 3, model 1). For Sample 1, psychological strain was not significant in the prediction of sick leave intentions (see Table 2, model 2a), but did predict medical advice intentions, regardless of which role demand was in the model (see Table 2, model 2b). For Sample 2, psychological strain had a positive direct effect on sick leave intentions (see Table 3, model 2a) and medical advice intentions (see Table 3, model 2b) regardless of which role demand was in the model. Thus, H2 received partial support.

Indirect effects (Hypothesis 3). For Sample 1, there was partial support for H3. The positive indirect effect of each role demand on sick leave intentions through psychological strain was not significant (see Table 2, model 2a). However, the positive indirect effect of each role demand on medical advice intentions through psychological strain was significant: role overload (indirect effect = 0.12, SE = 0.05, 95%CI), role conflict (indirect effect = 0.08, SE = 0.04, 95%CI [0.02, 0.16]), and role ambiguity (indirect effect = 0.13, SE = 0.07, 95%CI [0.02,0.29]), as seen in Table 2, model 2b.

For Sample 2, H3 received full support. All three role demands had a significant positive indirect effect on sick leave intentions through psychological strain: role overload (indirect effect = 0.43, SE = 0.07, 95%CI [0.30, 0.57]), role conflict (indirect effect = 0.42, SE = 0.07, 95%CI [0.30, 0.55]), and role ambiguity (indirect effect = 0.74, SE = 0.11, 95%CI [0.54,0.98]) (see Table 3, model 2a). Similarly, the positive indirect effect of each role demand on medical advice intentions through psychological strain was significant: role overload (indirect effect = 0.40, SE = 0.07, 95%CI [0.28, 0.54]), role conflict (indirect effect = 0.41, SE = 0.07, 95%CI [0.29, 0.55]), and role ambiguity (indirect effect = 0.72, SE = 0.11, 95%CI [0.51, 0.95]) (see Table 3, model 2b).

3.3 Moderated mediation results (Hypothesis 4)

For Sample 1, Hayes’ index of moderated mediation (IMM) [77] was nonsignificant across all three of the role demands on sick leave intentions (see Table 4, model 2a). The IMM was non-significant on medical advice intentions for role overload, but was significant for role conflict (IMM = –0.05, SE = 0.02, 95%CI [–0.10, –0.003]) and role ambiguity (IMM = –0.08, SE = 0.04, 95%CI [–0.16, –0.001]), as shown in Table 4, model 2b. For Sample 2, the IMM was significant on sick leave intentions for all role demands: role overload (IMM = –0.05, SE = 0.01, 95%CI [–0.07, –0.02]), role conflict (IMM = –0.05, SE = 0.01, 95%CI [–0.08, –0.03]), and role ambiguity (IMM = –0.09, SE = 0.02, 95%CI [–0.13, –0.05]), as shown in Table 5, model 2a. The IMMs also were significant for medical advice intentions for all three role demands: role overload (IMM = –0.03, SE = 0.01, 95%CI [–0.06, –0.01]), role conflict (IMM = –0.04, SE = 0.01, 95%CI [–0.07, –0.01]), and role ambiguity (IMM = –0.06, SE = 0.03, 95%CI [–0.11, –0.004]), as shown in Table 4, model 2b.

For each moderated mediation in which there was a significant IMM, the significant conditional indirect effects of the role demand on sick leave intentions and medical advice intentions through psychological strain were further examined at three values of praise and recognition: the 16th, 50th, and 84th percentiles. In relation to sick leave intentions for museum workers, the mediation remained significant at low, moderate, and high praise and recognition (see Table 6). However, the positive indirect effect weakened as a function of increasing levels of praise and recognition from one’s supervisor, in line with general expectations outlined in H4. In relation to medical advice intentions, across both samples, the positive indirect effect of role demands on medical leave intentions weakened as a function of increasing levels of supervisor praise and recognition (see Table 7). For Sample 1 motel workers, at low and moderate levels of psychological strain, the positive indirect effect was significant, while at high levels the positive indirect effect was, in fact, not significant, suggesting a full buffering effect. This pattern of results supports H4.

3.4 Supplementary analysis

Although Stage 1 moderated mediation was not hypothesized, such an interaction has theoretical justification in light of research that has examined the stress-buffering role of job resources in the relationship between job demands and various indicators of employee strain, as per the JD-R model (e.g., Bakker et al. [78], and also the effort-reward imbalance model [39] that incorporates reward and recognition with job demands to predict health-adverse outcomes. Thus, the extent to which praise and recognition moderated the relationship between each of the three role demands and psychological strain in the context of the proposed mediation was tested in both samples, but none of the IMMs were significant, and Stage 1 moderated mediation was ruled out.

4 Discussion

In light of the theoretical tenets contained in Fredrickson’s [6] undoing hypothesis and broadened thought-action repertoires as it relates to positive emotions, it was proposed that employees under psychological strain would be less inclined to be considering sick leave as a coping response when their supervisor is a source of praise and recognition. Significant moderated mediations were observed for museum employees, indicating that the positive indirect effect of all three role demands on sick leave intentions through psychological strain weakened as supervisor praise and recognition increased. Overall, these findings are important because previous studies that have found limited support for a moderating effect of supervisor behavior in reducing absence behavior for distressed employees have focused on long-term sickness absence (> 3 weeks; e.g., Munir et al., [64]; Hjarsbech et al., [65]). These results suggest that supervisor praise and recognition is beneficial for mitigating short-term absence intentions. Although it should be acknowledged that, whilst beneficial, even moderate and high levels of supervisor praise and recognition were nonetheless insufficient to fully prevent this specific representation of the stress process.

In contrast, there were no significant moderated mediations on sick leave intentions for motel workers. These non-significant findings may be because 60.6%of motel workers in our sample were employed on a casual basis and worked an average of 28.61 hours per week. As such, these employees may be reluctant to take sick leave due to the associated loss of income, given that casual employees are not eligible for paid sick leave in the Australian context [28]. Furthermore, employees who are working fewer hours each week may have sufficient time during non-work hours to replenish depleted resources. Overall, the contrast in our findings across samples points to the importance of focussing on homogenous occupational groups because the benefits of a specific job resource might be influenced by the employment context.

The present research found that, as replicated across two samples, praise and recognition buffered the positive indirect effect of role demands on medical advice intentions through psychological strain (with the exception of role overload in the motel worker sample). The deleterious effect of role demands on medical advice intentions through psychological strain became weaker, although again remained significant, as supervisor praise and recognition increased for museum workers. For motel workers in somewhat more precarious employment with limited access to sick leave, high levels of supervisor praise and recognition was effective in fully eliminating the relationship between psychological strain and medical advice ruminations.

4.1 Practical implications

The present research has clear practical implications for supervisors who find themselves having to support distressed employees. First, a thorough psychosocial risk analysis and minimizing job design factors that lead to role demands is essential. Second, for those employees experiencing psychological strain at work, these findings highlight the importance of line managers making strong and concerted efforts to acknowledge distressed employees via positive feedback and other gestures of appreciation. Praise and recognition is a simple, inexpensive form of appreciation [79], and qualitative research indicates that employees prefer verbal one-on-one expressions of appreciation to other forms of recognition [80]. In order to support employees under psychological strain, it is important that praise and recognition is not limited to major accomplishments or to the highest achievements within the workplace. Rather, to best assist, employees need to be recognized for a range of achievements as well as performance improvements, and even commendable efforts that were unsuccessful [79]. Overall, this research reinforces the need for HR training programs that coach supervisors in the evaluation of their own approach to performance management so as to implement appropriate employee recognition practices.

It should be clarified that taking sick leave or seeking medical advice can serve important functions in supporting employees to recuperate from stress and should be utilized when needed, especially since presenteeism is not an ideal situation for both employees and employers [81]. Nevertheless, Darr and Johns’ meta-analysis [21] concluded that, even though immediate withdrawal might provide short-term respite, later withdrawal in response to weakened physical and psychological states might maintain, rather than relieve, psychological strain. When the psychosocial job environment is health-promoting, remaining at work can improve common health conditions, and should be encouraged [82]. Furthermore, the consequences of needing to be absent from work –such as returning to an increased workload –can soon re-establish stress levels. Overall, therefore, while taking sick leave and seeking medical advice for stress-related problems should be encouraged when needed, efforts should be made to prevent the exacerbation of psychological strain to the point where employees feel the need to pursue such actions in order to manage their job demands.

4.2 Limitations and future research

Several methodological limitations must be noted. The cross-sectional design means that causal inferences inherent in a mediation model cannot be made [83]. Future research should use a temporal sequence with appropriate time intervals. Furthermore, self-report measures at a single time point means that significant associations may have been inflated by common method variance. A related limitation is that intentions to take sick leave and seek medical advice were not followed up with an objective assessment of these behaviors. Nevertheless, it could be argued that there is practical value to focusing on intentions so as to provide a tangible point of intervention for organizations, before actual withdrawal behaviors ensue, and is indeed theoretically aligned with the broadening of momentary thought-action repertoires via positive emotions (generated via praise and recognition) by undoing preparation for specific action (i.e., sick leave intentions).

Because the proposed indirect effect was not eliminated in full at moderate and high levels of supervisor praise and recognition in the museum sample of employees, other job resources might be needed to compliment praise and recognition from supervisors. Thus, future research is needed to examine if other job resources, either from the supervisor or elsewhere in the organization, are needed to prevent psychological strain leading to sick leave and medical advice ruminations. In addition, further research is needed to confirm the theoretical mechanisms through which supervisor praise and recognition weakens the association between psychological strain and stress-related intentions, for example via generating positive emotions and reducing negative emotions, and encouraging adaptive coping rather than maladaptive coping. In addition, given that Younies and Al-Tawil [84] found that hotel workers in the UAE rated the importance of receiving praise was more pronounced for those with higher organizational tenure, it also would be beneficial to examine how to design non-material reward systems so as to cater for individual differences in an organization’s workforce.

5 Conclusion

The present research proposed and tested a theoretical model in which psychological strain mediated the positive relationship between role demands and employees’ intentions to take sick leave, and supervisor praise and recognition moderated the psychological strain–stress-related intentions relationship. Overall, results confirm previous findings that high levels of role demands are detrimental to the psychological health of employees, and provides consistent new evidence that the health impairment process influences intentions to take sick leave and seek medical advice for work stress-related problems. The results provide partial support for the provision of praise and recognition from supervisors in order to reduce employees’ psychological strain from deteriorating to the point where they are considering the need to take sick leave and visit a doctor. Improving the praise and recognition skills and practices of managers is a simple yet effective tactic by which organizations could assist employees experiencing occupational stress and psychological strain.

Conflict of interest

None to report.

Funding

This research was supported by Linkage Project (LP0775049); Linkage Project (LP120100575); Future Fellowship (FT110100083), all funded by the Australian Research Council and awarded to the third author. The funders had no involvement in any parts of the research.

References

Harvey

, Sellahewa

, Wang

, Milligan-Saville

, Bryan

, Henderson

, Hatch

, Mykletun

. The role of job strain in understanding midlife common mental disorder: A national birth cohort study. Lancet Psychiatry. 2018;5:498–506.

OECD. Sick on the job? Myths and 849 realities about mental health and work. Paris: OECD Publishing; 2012.

Demerouti

, Bakker

, Nachreiner

, Schaufeli

. The job demands-resources model of burnout. J Appl Psychol. 2001;86:499–512.

Hobfoll

. Conservation of resources: A new attempt at conceptualizing stress. Am Psychol. 1989;44:513–24.

Hobfoll

. The influence of culture, community, and the nested self in the stress process: Advancing conservation of resources theory. Appl Psychol: An Int Rev.. 2001;50:337–70.

Fredrickson

. The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. Am Psychol. 2001;56:218–26.

Gilboa

, Shirom

, Fried

, Cooper

. A meta-analysis of work demand stressors and job performance: Examining main and moderating effects. Pers Psychol. 2008;61:227–71.

Rizzo

, House

, Lirtzman

. Role conflict and ambiguity in complex organizations. Adm Sci Q. 1970;15:161–3.

Kahn

, Wolfe

, Quinn

, Snoek

, Rosenthal

. Occupational stress: Studies in role conflict and ambiguity. New York: Wiley; 1964.

10.

Gonzalez-Mulé

, Kim

M(M)

, Ryu

. A meta-analytic test of multiplicative and additive models of job demands, resources, and stress. J Appl Psychol 2021;106:1391–411.

11.

Lesener

, Gusy

, Wolter

. The job demands-resources model: A meta-analytic review of longitudinal studies. Work Stress. 2019;33:76–103.

12.

Ariza-Montes

, Arjona-Fuentes

, Han

, Law

. Work environment and well-being of different occupational groups in hospitality: Job-demand-control-support model. Int J Hosp Manag. 2018;73:1–11.

13.

Zhao

, Ghiselli

. Why do you feel stressed in a ‘smile factory’? Hospitality job characteristics influence work-family conflict and job stress. Int J Contemp Hosp Manag. 2016;28:305–26.

14.

Wong

AKF

, Kim

, Han

. How the COVID-19 pandemic affected hotel employee stress: Employee perceptions of occupational stressors and their consequences. Int J Hosp Manag. 2021;93:102798.

15.

Kahn

, Garden

. Job attitudes and occupational stress in the United Kingdom museum sector. Mus Manag Curator. 1993;12:285–302.

16.

Bailey

. Researching museum educators’ perceptions of their roles, identity and practice. J Mus Educ. 2006;31:175–97.

17.

Munro

. Doing emotion work in museums: Reconceptualizing the role of community engagement practitioners. Mus Soc. 2014;12:44–60.

18.

Murdy

, Alexander

, Bryce

. Role conflict and changing heritage practice: Ancestral tourism in Scotland. J Mark Manag. 2016;32:1494–512.

19.

Chen

, Yu

. Effects of positive vs negative forces on the burnout-commitment-turnover relationship. J Serv Manag. 2014;25:388–410.

20.

Schaufeli

, Bakker

, van Rhenen

. How changes in job demands and resources predict burnout, work engagement, and sickness absenteeism. J Organ Behav. 2009;30:893–917.

21.

Darr

, Johns

. Work strain, health and absenteeism: A meta-analysis. J Occup Health Psychol. 2008;13:293–318.

22.

Coledam

DHC

, da Silva

. Predictors of health-related absenteeism, presenteeism and sick leave among Brazilian elementary school teachers: A cross-sectional study. Work: J Prev, Assess Rehabil. 2020;67:709–19.

23.

Gohar

, Larivière

, Lightfoot

, Wenghofer

, Larivière

, Nowrouzi-Kia

. Understanding sickness absence in nurses and personal support workers: Insights from Northeastern Ontario. Work: J Prev, Assess Rehabil. 2020;66:755–66.

24.

Kinman

, Jones

. Lay representations of workplace stress: What do people really mean when they say they are stressed? Work Stress. 2005;19:101–20.

25.

Andrea

, Metsemakers

JFM

, Kant

, Beurskens

AJHM

, Swaen

GMH

, van Schayck

. Seeking help in relation to work –visiting the occupational physician or the general practitioner. Occup Med. 2004;54:419–21.

26.

Azagba

, Sharaf

. Psychosocial working conditions and the utilization of health care services. BMC Public Health. 2011;11:642–642.

27.

Australian Trade Commission. Australian tourism labour force report: 2015–2020. October 2015. Retrieved from https://www.tra.gov.au/Archive-TRA-Old-site/Research/View-all-publications/All-Publications/australian-tourism-labour-force-report-2015-2020.

28.

Australian Government Fair Work Ombudsman. Casual Employees. (nd) Retrieved from https://www.fairwork.gov.au/employee-entitlements/types-of-employees/casual-part-time-and-full-time/casual-employees#

29.

Safe Work Australia. Australian workers compensation statistics 2018-2019, Canberra, Australian Capital Territory: Author. 2021.

30.

Bakker

, Demerouti

, de Boer

, Schaufeli

. Job demands and job resources as predictors of absence duration and frequency. J Vocat Behav. 2003;62:341–56.

31.

Vignoli

, Guglielmi

, Bonfiglioli

, Violante

. How job demands affect absenteeism? The mediating role of work-family conflict and exhaustion. Int Arch Environ Occup Health. 2016;89:22–31.

32.

Montani

, Boudrias

, Pigeon

. Employee recognition, meaningfulness and behavioral involvement: Test of a moderated mediation model. Int J Hum Resour Manag. 2020;31:356–84.

33.

Rooney

, Gottlieb

. Development and initial validation of a measure of supportive and unsupportive managerial behaviors. J Vocat Behav. 2007;71:186–203.

34.

Jiménez

, Winkler

, Dunkl

. Creating a healthy working environment with leadership: The concept of health-promoting leadership. Int J Hum Resour Manag. 2017;28:2430–48.

35.

Bass

, Avolio

. Full range leadership development: Manual for the multifactor leadership questionnaire. Redwood City CA: Mind Garden. 1997.

36.

Owens

, Hekman

. Modeling how to grow: An inductive examination of humble leader behaviors, contingencies, and outcomes. Acad Manag J. 2012;55:787–818.

37.

Stocker

, Jacobshagen

, Krings

, Pfister

, Semmer

. Appreciative leadership and employee well-being in everyday working life. German J Hum Resour Manag Res. 2014;28:73–95.

38.

Stocker

, Jacobshagen

, Semmer

, Annen

. Appreciation at work in the Swiss armed forces. Swiss J Psychol. 2010;69:117–24.

39.

Siegrist

. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996;1:27–41.

40.

van Vegchel

, de Jonge

, Bakker

, Schaufeli

. Testing global and specific indicators of rewards in the Effort-Reward Imbalance Model: Does it make any difference? Eur J Work and Organ Psychol. 2002;11:403–21.

41.

Rooney

, Gottlieb

, Newby-Clark

. How support-related managerial behaviors influence employees: An integrated model. J Manag Psychol. 2009;24:410–27.

42.

Westerlund

, Nyberg

, Bernin

, Hyde

, Oxenstierna

, Jäppinen

, Väänänen

, Theorell

. Managerial leadership is associated with employee stress, health, and sickness absence independently of the demand-control-support model. Work: J Prev, Assess Rehabil. 2010;37:71–9.

43.

Kranabetter

, Niessen

. Appreciation and depressive symptoms: The moderating role of need satisfaction. J Occup Health Psychol. 2019;15:106–15.

44.

Duchemin

, Bar-Hen

, Lounissi

, Dab

, Hocine

. Hierarchizing determinants of sick leave: Insights from a survey on health and well-being at the workplace. J Occup Environ Med. 2019;61:340–7.

45.

Duchaine

, Aubé

, Gilbert-Ouimet

, Véina

, Ndjaboué

, Massamba

, Talbot

, Lavigne-Robichaud

, Trudel

, Bruno Pena-Gralle

A-P

, Lesage

, Moore

, Milot

, Laurin

, Brisson

. Psychosocial stressors at work and the risk of sickness absence due to a diagnosed mental disorder: A systematic review and meta-analysis. JAMA Psychiatry. 2020;77:842–51.

46.

Boudens

. The story of work: A narrative analysis of workplace emotion. Organ Stud. 2005;26:1285–306.

47.

Semmer

, Jacobshagen

, Meier

, Elfering

. Occupational stress research: The “Stress-As-Offense-to-Self” perspective. In: Houdmont J, McIntyre S. editors. Occupational health psychology: European perspectives on research, education and practice. Castelo da Maia: ISMAI; 2007;43-60.

48.

Weiss

, Cropanzano

. An affective events approach to job satisfaction. In Staw BM, Cummings LL. editors. Research in organizational behaviour. Greenwich, CT: JAI Press; 1996;1–74.

49.

Dasborough

. Cognitive asymmetry in employee emotional reactions to leadership behaviors. Leadersh Q. 2006;17:163–78.

50.

Ohly

, Schmitt

. What makes us enthusiastic, angry, feeling at rest or worried?: Development and validation of an affective work events taxonomy using concept mapping methodology. J Bus Psychol. 2015;30:15–35.

51.

Basch

, Fisher

. Affective events-emotions matrix: A classification of work events and associated emotions. In Ashkanasy N, Hartel, C, Zerbe W. editors. Emotions in the workplace : Theory, research and practice. Westport CT: Quorum Books; 2000;36–49.

52.

Wang

, Zhu

, Dormann

, Song

, Bakker

. The daily motivators: Positive work events, psychological needs satisfaction, and work engagement. Appl Psychol: An Int Rev. 2020;69:508–37.

53.

Pfister

, Jacobshagen

, Kälin , Stocker

, Meier

, Semmer

. Appreciation and illegitimate tasks as predictors of affective well-being: Disentangling within- and between-person effects. J Work Organ Psychol.. 2020;36:63–75.

54.

Pfister

, Jacobshagen

, Kälin

, Semmer

. How does appreciation lead to higher job satisfaction. J Manag Psychol. 2020;35:465–79.

55.

Fredrickson

, Levenson

. Positive emotions speed recovery from the cardiovascular sequelae of negative emotions. Cogn Emot. 1998;12:191–220.

56.

Fredrickson

, Mancuso

, Branigan

, Tugade

. The undoing effect of positive emotions. Motiv Emot. 2000;24:237–58.

57.

Taylor

, Lyubomirsky

, Stein

. Upregulating the positive affect system in anxiety and depression: Outcomes of a positive activity intervention. Depress Anxiety. 2017;34:267–280.

58.

Riskind

, Kleiman

, Schafer

. ‘Undoing’ effects of positive affect: Does it buffer the effects of negative affect in predicting changes in depression? J Soc Clin Psychol. 2013;32:363–80.

59.

Isen

. Positive affect and decision-making. In Lewis M, Haviland-Jones JM, editors. Handbook of emotions. New York: Guilford Press. 2000;417–35.

60.

Fredrickson

, Branigan

. Positive emotions broaden the scope of attention and thought-action repertoires. Cogn Emot. 2005;19:313–32.

61.

Burns

, Brown

, Sachs-Ericsson

, Plant

, Curtis

, Fredrickson

, Joiner

. Upward spirals of positive emotion and coping: Replication, extension, and initial exploration of neurochemical substrates. Pers Individ Dif. 2008;44:360–70.

62.

Stare

, Pezdir

, Boštjančič

. Links between perceived leadership styles and self-reported coping strategies. Psychol Top. 2013;22:413–30.

63.

Väänänen

, Toppinen-Tanner

, Kalimo

, Mutanen

, Vahtera

, Piero

. Job characteristics, physical and psychological symptoms, and social support as antecedents of sickness absence among men and women in the private industrial sector. Soc Sci Med. 2003;57:807–24.

64.

Munir

, Burr

, Hansen

, Rugulies

, Nielsen

. Do positive psychosocial work factors protect against 2-year incidence of long-term sickness absence among employees with and those without depressive symptoms? A prospective study. J Psychosom Res. 2011;70:3–9.

65.

Hjarsbech

, Christensen

, Andersen

, Borg

, Aust

, Rugulies

. Do psychosocial working conditions modify the effect of depressive symptoms on long-term sickness absence? Am J Ind Med. 2013;56:1329–40.

66.

Sandin

, Chorot

, Santed

, Valiente

. Differences in negative life events between patients with anxiety disorders, depression and hypochondriasis. Anxiety Stress Coping. 2004;17:37–47.

67.

Australian Bureau of Statistics. Tourist accommodation, Australia. 2015-16, cat no. 8635.0 available at: http://www.abs.gov.au/ausstats/abs@.nsf/mf/8635.0/.

68.

Cousins

, Mackay

, Clarke

, Kelly

, McCaig

. Management standards and work-related stress in the UK: Practical development. Work Stress.. 2004;18:113–36.

69.

Haynes

, Wall

, Bolden

, Stride

, Rick

. Measures of perceived work characteristics for health services research: Test of a measurement model and normative data. Br J Health Psychol. 1999;4:257–75.

70.

Goldberg

. The detection of psychiatric illness by questionnaire. London: Oxford University Press. 1972.

71.

Kuper

, Singh-Manoux

, Siegrist

, Marmot

. When reciprocity fails: effort-reward imbalance in relation to coronary heart disease and health functioning within the Whitehall II Study. Occup Environ Med. 2002;59:777–84.

72.

Migneault

, Rousseau

, Boudrias

. The influence of work climate components on individual’s empowerment. Eur Rev App Psychol. 2009;59:239–52.

73.

Sveinsdóttir

, Ragnarsdóttir

, Blöndal

. Praise matters: The influence of nurse unit managers’ praise on nurses’ practice, work environment and job satisfaction: A questionnaire study. J Adv Nurs. 2016;72:558–68.

74.

Hayes

, Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. 2nd ed. New York: Guilford Press. 2018.

75.

Yzerbyt

, Muller

, Batailler

, Judd

. New recommendation for testing indirect effects in mediational models: The need to report and test component paths. J Pers Soc Psychol: Attitudes and Social Cognition. 2018;115:929–43.

76.

Muthén

, Satorra

. Complex sample data in structural equation modelling. In: Marsden PV. editor. Social methodology. Oxford: Blackwell; 1995;267–316.

77.

Hayes

. An index and test of linear moderated mediation. Multivariate Behav Res. 2015;50:1–22.

78.

Bakker

, Demerouti

, Euwema

. Job resources buffer the impact of job demands on burnout. J Occup Health Psychol. 2005;10:170–180.

79.

Yukl

. Leadership in organizations. Boston: Pearson. 2013.

80.

Beck

. Perceptions of thanks in the workplace: Use, effectiveness, and dark sides of managerial gratitude. Corp Comm Int J. 2016;21:333–51.

81.

Johns

. Presenteeism in the workplace: A review and research agenda. J Organ Behav. 2010;31:519–42.

82.

Whysall

, Bowden

, Hewitt

. Sickness presenteeism: Measurement and management challenges. Ergonomics. 2018;61:341–54.

83.

Spector

, Meier

. Methodologies for the study of organizational behavior processes: How to find your keys in the dark. J Organ Behav. 2014;35:1109–19.

84.

Younies

, Al-Tawil

. Hospitality workers’ reward and recognition. Int J Law Manag. 2021;63:157–71.