Reliability and validity of the Japanese version of the Overwork Climate Scale

Abstract

This study aimed to investigate the validity and reliability of the Japanese version of the Overwork Climate Scale. Japanese workers were invited to participate in online surveys at baseline and 1-month follow-up. The Overwork Climate Scale was translated into Japanese, according to international guidelines. Reliability was assessed using Cronbach’s alpha and the intra-class correlation coefficient (ICC), while structural validity was evaluated through confirmatory factor analysis (CFA). Psychological job demands, work engagement, psychological safety, and workaholism were assessed for convergent validity. The number of respondents was 302 at baseline and 169 at follow-up. Results indicated robust Cronbach’s alpha values of 0.86 (for overwork endorsement) and 0.80 (for lacking overwork reward) at baseline, complemented by ICC of 0.89 and 0.82, respectively. CFA confirmed the suitability of the two-factor model. Moreover, the Japanese Overwork Climate Scale exhibited significant correlations with anticipated constructs. Structural equation modeling revealed a consistent association between overwork climate and both workaholism and work engagement, similar to the original version. In conclusion, the Japanese version of the Overwork Climate Scale demonstrates acceptable levels of reliability and validity, warranting its potential adoption among Japanese workers.

Introduction

In modern work environments, advancements in information technology have empowered workers to work flexibly, transcending traditional boundaries of time and location. However, the regulation of working hours remains an important issue for modern industries and workers¹⁾. Notably, teleworking has rapidly increased in Japan since the beginning of 2020, blurring the lines between work and personal time^{2, 3)}.

Long working hours have emerged as a critical issue impacting occupational safety and health. They pose both physical^4,5,6,7) and psychological^8,9,10) risks to workers and can detrimentally affect work quality, productivity, and workers’ personal lives¹⁾. Extended work hours may reduce workers’ time for rest and sleep^{11, 12)}, fostering unhealthy habits such as smoking and excessive drinking¹³⁾. Moreover, they can incur costs for organizations through diminished productivity¹⁴⁾ and disrupt workers’ work-life balance¹⁵⁾. Despite the global urgency to address the prevalence of excessive working hours⁴⁾, research on organizational climates fostering harmful overwork remains limited. Thus, clarifying the circumstances under which workers perceive an overworked climate is imperative for safeguarding workers’ health in the workplace^{16, 17)}

Mazetti et al. introduced the Overwork Climate Scale to assess employees’ perceptions of work environments characterized by excessive demands and inadequate rewards¹⁸⁾. An overwork climate refers to a psychological climate wherein individuals perceive their work environments^{18,19,20,21,22,23)}. This psychological climate influences their interpretation of workplace occurrences and subsequent behavioral responses²¹⁾, impacting their commitment, motivation, and performance^{19,20,21,22,23,24,25)}. Moreover, it shapes specific behaviors such as organizational citizenship behavior, counterproductive work behavior, and workplace incivility^26,27,28,29). A meta-analysis conducted by Parker et al.²¹⁾ identified five cognitive components within the psychological climate: job characteristics (e.g., autonomy), role characteristics (e.g., overload), leadership characteristics (e.g., upper management control), workgroup and social relationships (e.g., cooperation, warmth), and organizational and subsystem characteristics (e.g., disclosure, innovation). This implies that overwork climate may be regarded as a manifestation of role overload and upper management control. Employees perceiving overwork climate tend to engage in excessive work behaviors^{19, 30)}.

The scale comprises two dimensions: overwork endorsement and lacking overwork rewards¹⁸⁾. Overwork endorsement encompasses perceptions of pressure from upper management to work overtime and widespread overtime practices within the workplace¹⁸⁾. Conversely, lacking overwork rewards pertains to employees feeling inadequately compensated for their overtime efforts or not being granted compensatory time off. Moreover, Mazetti et al. revealed that the absence of compensation for overtime work negatively impacts work engagement¹⁸⁾.

Versions of Overwork Climate Scale exist in French and Polish^{31, 32)}. The Polish version maintains the original two-factor structure³²⁾, while the French version proposes a modified structure, focusing solely on overtime endorsement and omitting the lack of overwork rewards³¹⁾. The authors of the French version attribute these differences to cultural variations³¹⁾. Notably, the majority of French participants (n=198) 35.5%, particularly in industries such as insurance and banking³¹⁾, may explain the absence of compensation concerns. Conversely, industries such as transportation, warehousing, lodging, food services, and construction report a higher percentage of unpaid work³³⁾. In the broader Japanese cultural context, organizational norms often prioritize long working hours³⁴⁾, valuing diligence and company commitment over personal time^{34, 35)}. Workers may hesitate to express dissatisfaction with unpaid overtime, as motivation and diligence are integral to personnel evaluations^{35, 36)}. Thus, the overwork climate likely manifests as a two-factor structure.

Regarding the relationship between perceptions of an overwork climate and other variables, individuals experiencing such a climate often encounter heightened psychological job demands. These demands, characterized by frequent, challenging, and psychologically taxing tasks, create pressures within the work environment³⁷⁾. Consequently, workers may be compelled to extend their work hours, including weekends or holidays, to meet these demanding expectations¹⁹⁾. Previous research has indicated a positive correlation between the overwork climate and psychological demands, particularly regarding overwork endorsement and lacking overwork rewards¹⁸⁾.

Employees who perceive an overwork climate tend to experience lower levels of psychological safety^38,39,40). This concept refers to individuals’ ability to navigate their work environment without fear of negative repercussions to their self-image, status, or career^{38,39,40,41,42,43,44,45)}. In a collectivist setting, the absence of someone to speak up about such complaints leads to workers perceiving that voicing will threaten their position, and thus are reluctant to speak^{38,39,40,41,42,43,44,45)}. Consequently, the overwork climate is often negatively associated with workers’ sense of psychological safety^{38, 39)}.

Additionally, employees in an overwork climate tend to exhibit lower levels of work engagement^{20, 46, 47)}. According to the job demands-resources (JD-R) model, the negative impact of high job demands can be mitigated by adequate job resources, such as autonomy and social support. However, in an overwork climate, employees working long hours under intense workload pressure without adequate compensation for their work will feel a lack of support from their supervisors and control over their own work^{46, 47)}. Xanthopoulou et al. demonstrated that work resources are associated with increased work engagement and higher financial returns⁴⁸⁾, but a lack of these resources may lead to depletion and reduced engagement^{46, 47)}.

Moreover, empirical studies suggest a correlation between the overwork climate and workaholism^{16,18,19,20,49,50)}. Workaholics, defined as workers whose thoughts, emotions, and actions are dominated by work²⁰⁾, are more prevalent in environments where overtime work is endorsed and opportunities for rest are limited. Mazetti et al. demonstrated a significant relationship between workaholism and the overwork climate, particularly in terms of overwork endorsement¹⁸⁾. This relationship remained significant even after controlling for individual personal traits¹⁹⁾. Thus, we hypothesized that workaholism has positive correlations with overwork climate.

Objectives

This study aimed to examine the reliability and validity of the Japanese version of the Overwork Climate Scale among Japanese workers. Specifically, we investigated the internal consistency, test-retest reliability, structural validity, and convergent validity of the translated measure. We hypothesized that the Japanese version of the Overwork Climate Scale would demonstrate satisfactory structural validity, internal consistency, test-retest reliability, and convergent validity. Regarding convergent validity, we hypothesized that an overwork climate would have positive correlations with psychological job demands and workaholism and a negative correlation with psychological safety and work engagement. Furthermore, considering the proposition of a one-factor model in the French version³¹⁾, we examined whether a one- or two-factor model would better fit the Japanese version. Additionally, studies have indicated that an overwork climate may significantly contribute to the development and perpetuation of workaholism^{18, 32)}. Consistent with this, previous studies utilizing structural equation modeling (SEM) in the English and Polish versions demonstrated a satisfactory fit in the relationship between overwork climate and workaholism/work engagement^{18, 32)}. Therefore, we conducted exploratory SEM to investigate this relationship further.

Subjects and Methods

Participants

We conducted a baseline survey in March 2022, followed by a 1-month follow-up survey in April 2022, utilizing an internet-based panel. The Japanese version of the Overwork Climate Scale underwent testing for internal consistency, structural validity, and convergent validity during the baseline survey, while test-retest reliability was assessed using longitudinal data.

Participants were invited from workers registered as respondents with an internet-based research company, Cross Marketing Inc., which has 2,190,000 active respondents who have recently completed questionnaires. A total of 314 workers completed a web-based questionnaire in the order of their arrival. The minimum effect size for detection in this study was 0.2 (ρ). We calculated the required sample size using G*Power version 3.1.9.7, and it was estimated to be greater than 262 when considering an α-error probability of 0.05 and power (1-β) of 0.95, according to those in previous studies^51,52,53). The inclusion criteria were employees who typically worked ≥35 h per week, held non-managerial positions, resided in Japan, and were aged between 20 and 65 yr. Given that the standard retirement age for most employees in Japan is 65 yr, individuals above this age were excluded. Managers were also excluded from participation to minimize potential influence on employees’ perceptions of workplace climate. Cross Marketing Inc. recruited eligible workers until the target number of participants was reached. Upon completion of the questionnaires, eligible workers who provided informed consent received 100 points, approximately equivalent to 100 Japanese yen. One month later, participants were invited by Cross Marketing Inc. for the follow-up survey. The study protocol was approved by the Research Ethics Committee of the Faculty of Human Sciences, University of Tsukuba (no. TOU 2021-113).

Measures

Overwork climate

Overwork climate was assessed using 11 items on a five-point Likert scale (1=strongly disagree to 5=strongly agree). It comprises two factors: (a) overwork endorsement (seven items) and (b) lacking overwork rewards (four items)¹⁹⁾. These items assess the extent to which workers perceive their workplaces to align with specific statements. We translated the scale developed by Mazetti et al. from English to Japanese following the guidelines of the International Society of Pharmacoeconomics and Outcomes Research to ensure cross-linguistic equivalence⁵⁴⁾. Initially, permission was obtained from the original authors to translate the scale. Subsequently, two translators, proficient in both Japanese and English, independently translated the content. A unified Japanese version was created by reconciling the sequential translations. Additionally, a translator not involved in the initial translation process conducted a reverse translation of the Japanese version into English. The original authors then assessed the equivalence between the original and back-translated versions, leading to the final Japanese version (see Appendix). To ensure the comprehensibility and cognitive equivalence of the Japanese version, a brief survey was administered to 11 native Japanese speakers. These participants confirmed their understanding of the content and verified that the expressions were natural in Japanese. Scale scores were computed by calculating the average of the individual item responses.

Psychological job demands

Psychological job demands were assessed using the Brief Job Stress Questionnaire⁵⁵⁾. This scale consists of two factors: (a) quantitative and (b) qualitative demands. Each subscale comprises three items rated on a four-point Likert scale (1=strongly disagree to 4=strongly agree), with higher scores indicating greater quantitative/qualitative job demands. The sample items for job demands is “I have an extremely large amount of work to do” for job quantitative, and “I have to pay very careful attention” for qualitative job demands. The mean score for each factor was computed by averaging the individual item responses. The Cronbach’s α coefficient for quantitative and qualitative demands were 0.85 and 0.76, respectively.

Psychological safety

Psychological safety was evaluated using the Japanese version of the Psychological Safety Scale⁵⁶⁾. The scale comprises five items rated on a five-point Likert scale (1=strongly disagree to 5=strongly agree). The items for psychological safety include “In my work unit, I can express my true feelings regarding my job” and “In my work unit, I can freely express my thoughts”. Scale scores were obtained by averaging the responses to the items. The Cronbach’s α coefficient for psychological safety was 0.84.

Work engagement

The nine-item Japanese version of the Utrecht Work Engagement Scale (UWES) was used to assess work engagement⁵⁷⁾. The UWES comprises three subscales: vigor, dedication, and absorption. Each subscale has three items. Responses to all items were provided on a seven-point Likert scale (1=never to 7=always). Sample items for UWES includes “At my job, I feel strong and vigorous” for vigor, “I am enthusiastic about my job” for dedication, and “I am immersed in my work” for absorption. The mean score for each subscale was computed by averaging the responses to the respective items. The Cronbach’s α coefficients were 0.92 for vigor, 0.90 for dedication, and 0.92 for absorption.

Workaholism

Workaholism was assessed using the Dutch Workaholic Scale (DUWAS)⁵⁸⁾. This scale includes two subscales: working excessively (five items) and working compulsively (five items). Participants rated their agreement with each item on a four-point Likert scale (1=[almost] never to 4=[almost] always). Sample items include “I seem to be in a hurry and racing against the clock”, “I find myself continuing to work after my coworkers have called it quits”, “It is important for me to work hard even when I do not enjoy what I am doing”, and “I feel that there is something inside me that drives me to work hard”. The mean score for each subscale was computed by averaging the responses to the respective items. The Cronbach’s α coefficients were 0.82 for working excessively and 0.82 for working compulsively.

Data analyses

To confirm the two-factor structural validity, confirmatory factor analyses were performed on the 11 items using a robust maximum likelihood estimation. Model fit was evaluated using various indices, including Chi-square (χ²), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), Standardized Root Mean Square Residual (SRMR), goodness of fit index (GFI), and adjusted goodness of fit index (AGFI). Both the original two-factor model and a one-factor model, where all 11 items were explained by a single factor, were tested.

The Cronbach’s α and intra-class correlation coefficient (ICC) of the Japanese version of the Overwork Climate Scale were calculated to test reliability.

Convergent validity was examined by calculating Pearson’s correlation coefficients (r) between overwork climate and scores for psychological job demands, psychological safety, work engagement, and workaholism.

Furthermore, exploratory SEM was conducted to investigate the relationships between overwork climate, work engagement, and workaholism. The sample size of 300 was also adequate for SEM analysis^{59, 60)}. Data were analyzed using IBM SPSS Statistics^® version 26 and IBM SPSS Amos^® version 26, with statistical significance set at p<0.05.

Results

Characteristics of participants

From the initial 314 respondents in the baseline survey, responses from 12 respondents were excluded due to inappropriate answers, such as consistently selecting the same options regardless of reversed-coded items. In the follow-up survey, 169 out of 302 participants responded to the questionnaire again (response rate=60.0%). The baseline survey comprised 152 men (50.3%) and 150 women (49.7%), with a mean age of 48.0 (SD=10.6) yr. Most participants were full-time workers (79.1%) or contract/temporary workers (13.6%), with work styles including fixed-time (69.2%) and flextime (12.9%). Industries represented included services (22.2%) and manufacturing (21.5%). Most participants worked between 35 and 50 h per week (86.8%). Among the 302 workers, 42 reported unpaid overtime (defined as work hours exceeding scheduled working hours). The demographic characteristics of participants at baseline and follow-up surveys are summarized in Table 1.

Table 1. Demographic characteristics of the participants.

			Baseline survey (n=302)		Follow-up survey (n=169)
			n (%)	Mean (SD)	n (%)	Mean (SD)
Gender
	Men		152 (50.3)		90 (53.3)
	Women		150 (49.7)		79 (46.7)
	Other
Age				48.0 (10.6)		50.5 (10.4)
Employment status
	Full-time		239 (79.1)		134 (79.3)
	Contract employees		41 (13.6)		28 (16.6)
	Temporary staff		22 (7.3)		7 (4.1)
Work style
	Fixed hours		209 (69.2)		120 (71.0)
	Variable working hour system		28 (9.3)		15 (8.9)
	Flextime system		39 (12.9)		17 (10.1)
	Discretionary labor system		8 (2.6)		6 (3.6)
	Rotation/night shift		17 (5.6)		10 (5.9)
	Other		1 (0.3)		1 (0.6)
Job type
	Professional/Technical		62 (20.5)		37 (21.9)
	Clerical		119 (39.4)		67 (39.6)
	Sales		33 (10.9)		18 (10.7)
	Services		30 (9.9)		14 (8.3)
	Transport/Construction		17 (5.6)		8 (4.7)
	Production/Skilled		27 (8.9)		17 (10.1)
	Other		14 (4.6)		8 (4.7)
Job category
	Construction		20 (6.6)		13 (7.7)
	Manufacturing		65 (21.5)		37 (21.9)
	Services		67 (22.2)		30 (17.8)
	Information and Communication		21 (7.0)		9 (5.3)
	Transport		14 (4.6)		8 (4.7)
	Retail		30 (9.9)		23 (13.6)
	Financial/Insurance		17 (5.6)		8 (4.7)
	Real estate and goods rental		6 (2.0)		3 (1.8)
	Education		12 (4.0)		6 (3.6)
	Medical/Welfare		34 (11.3)		22 (13.0)
	Public service		11 (3.6)		6 (3.6)
	Other		5 (1.7)		4 (2.4)
Working hours
	35–40 h /w		156 (51.7)		90 (53.3)
	41–50 h /w		106 (35.1)		59 (34.9)
	51–60 h /w		23 (7.6)		12 (7.1)
	61 h – /w		17 (5.6)		8 (4.7)
Unpaid overtime
	No		260 (86.1)		145 (85.8)
	Yes		42 (13.9)		24 (14.2)
	Sum of unpaid overtime	1–5 h	29 (9.6)		19 (11.2)
		6–10 h	10 (3.3)		3 (1.8)
		11–30 h	3 (1.0)		2 (1.2)
Size of worksite
	100 employees		113 (37.4)		62 (36.7)
	<500 employees		68 (22.5)		40 (23.7)
	<1,000 employees		17 (5.6)		11 (6.5)
	1,000 emplyees –		93 (30.8)		51 (30.2)
	Public offices		11 (3.6)		5 (3.0)
Marital status
	Married		162 (53.6)		91 (53.8)
	Unmarried		136 (45.0)		75 (44.4)
	Others (divorced, etc.)		4 (1.3)		3 (1.8)

Figures may not always add up to 100% due to rounding data.

SD: standard deviation.

Structural validity

Table 2 displays the results of the confirmatory factor analyses for both the one- and two-factor models. For the one-factor model, the indicators were as follows: χ² [43]=238.008, CFI=0.861, TLI=0.822, RMSEA=0.123, SRMR=0.081, GFI=0.864, and AGFI=0.792. The original two-factor model demonstrated marginally acceptable fit (χ² [42]=151.735, CFI=0.922, TLI=0.898, RMSEA=0.093, SRMR=0.058, GFI=0.916, and AGFI=0.868). Covariance errors were adjusted as items 8 and 9 had similar expressions when translated into Japanese⁶¹⁾.

Table 2. Factor loadings of the 11 overwork climate scale items, factor correlations, and model fit in confirmatory factor analyses.

Internal consistency and test-retest reliability

Table 3 presents the mean scores, Cronbach’s α, ICCs, SEMs, and smallest detectable changes (SDCs) for the Overwork Climate Scale. Cronbach’s α coefficients were 0.86 for overwork endorsement and 0.80 for lacking overwork rewards. The ICCs for each factor were 0.89 for overwork endorsement and 0.82 for lacking overwork rewards, respectively, indicating that more than 80% of the variance in the two measurements can be attributed to individual differences. SDCs were 0.12 and 0.16.

Table 3. Mean (SD), factor loadings in confirmatory factor analysis, and internal consistency for each item of the overwork climate scale.

Item no.	Item	Min – Max	Baselinea		Follow-upb	Test-retest Reliabilityb
			Mean (SD)	Cronbach’s α	Mean (SD)	ICC	SEM	SDC
Overwork endorsement (7 items)			2.35 (0.88)	0.86	2.37 (0.88)	0.89	0.04	0.12
1	Almost everybody expects that employees perform overtime work.	1–5	2.82 (1.30)		2.72 (1.30)	0.81	0.08	0.23
2	Management encourages overtime work.	1–5	2.17 (1.14)		2.18 (1.13)	0.82	0.07	0.19
3	It is considered normal for employees to take work home.	1–5	1.80 (0.96)		1.83 (0.96)	0.74	0.06	0.18
4	Most employees work beyond their official work hours.	1–5	2.81 (1.29)		2.90 (1.28)	0.83	0.08	0.21
5	Performing overwork is important for being promoted.	1–5	2.15 (1.10)		2.25 (1.10)	0.77	0.07	0.20
6	It is considered normal to work on weekends.	1–5	2.25 (1.25)		2.33 (1.26)	0.72	0.09	0.26
7	It is difficult to take a day off or paid holidays.	1–5	2.43 (1.23)		2.38 (1.23)	0.85	0.07	0.19
Lacking overwork rewards (4 items)			2.68 (1.01)	0.80	2.68 (0.98)	0.82	0.06	0.16
8	Overtime work is fairly compensated by extra time off work or by other perks. (R)	1–5	2.75 (1.26)		2.67 (1.28)	0.63	0.10	0.28
9	Working overtime is fairly compensated financially. (R)	1–5	2.73 (1.29)		2.69 (1.24)	0.70	0.09	0.26
10	(Almost) nobody needs to do unpaid overtime work. (R)	1–5	2.42 (1.31)		2.44 (1.34)	0.77	0.09	0.24
11	A policy exists to restrict overtime work. (R)	1–5	2.82 (1.24)		2.91 (1.23)	0.76	0.08	0.23

a: n=302, b: n=169, SD: standard deviation; ICC: intra-class correlation; SEM: standard error of measurement; SDC: smallest detectable change; R: reverse coded.

Convergent validity

Table 4 displays Pearson’s correlation coefficients (r) among the overwork climate factors, workaholism, work engagement, psychological job demands, and psychological safety. Each Overwork Climate Scale factor moderately correlated with psychological safety (overwork endorsement, r=−0.32; lacking overwork rewards, r=−0.36). Moreover, overwork endorsement exhibited moderate correlations with workaholism (working excessively, r=0.38; compulsively, r=0.27) and weak correlations with lacking overwork rewards (r=0.19 and r=0.16, respectively). Additionally, psychological job demands correlated with overwork endorsement (quantitative job demands, r=0.26; qualitative job demands, r=0.25) but not with lacking overwork rewards. Interestingly, work engagement negatively correlated with lacking overwork rewards, while neither vigor nor absorption correlated with overwork endorsement.

Table 4. Convergent validity of the Japanese version of the overwork climate scale.

	Psychological job demands		Work engagement			Psychological safety	Workaholism
	Quantative demands	Qualitative demands	Vigor	Dedication	Absorption		Working excessively	Working compulsively
Overwork endorsement	0.26**	0.25**	–0.11	–0.13*	–0.06	–0.36**	0.38**	0.27**
Lacking overwork rewards	0.09	–0.01	–0.18**	–0.20**	–0.16**	–0.32**	0.19**	0.16**

*p<0.05, **p<0.01.

Relationships between overwork climate and work engagement/workaholism

The results of SEM exploring the two hypotheses are shown in Fig. 1. The model exhibited marginal fit (χ² [100]=296.94, p<0.01, TLI=0.93, CFI=0.92, and RMSEA=0.08). Consistent with the original version, lacking overwork rewards exhibited a negative correlation with work engagement (r=−0.21, p<0.01), while overwork endorsement exhibited a positive correlation with workaholism (r=0.39, p<0.01) and was not significantly related to work engagement. However, unlike the original version, lacking overwork rewards showed no relationship with workaholism.

Fig. 1.

The relationships between overwork climate and work engagement/workaholism.

n=302, χ²(100)=296.94, p<0.01, TLI=0.93, CFI=0.92, RMSEA=0.08, **p<0.01 Values in parenthesis were not included in the analysis because they were not significant.

Discussion

In this study, the Japanese version of the Overwork Climate Scale demonstrated acceptable structural validity and good reliability, along with convergent validity. The structural validity was acceptable. Confirmatory analyses favored the two-factor model over the one-factor model, aligning with findings from the original version. The CFI, TLI, and SRMR demonstrated a good fit, while the RMSEA suggested otherwise. This discrepancy in RMSEA may be attributed to factors such as sample size and the number of items. RMSEA’s sensitivity to small sample sizes or fewer variables underscores the need for further research with larger sample sizes to corroborate these findings. Moreover, SRMR, known for its stability regardless of sample size or the number of variables, offers a more robust indicator of model fit, supporting the overall adequacy of the structural validity demonstrated by the Japanese version of the Overwork Climate Scale^59,60,61,62).

The measure demonstrated strong internal consistency and was stable for more than a month. As for test-retest reliability, DeVet et al. of the COSMIN workgroup found 0.70 to be acceptable and values above 0.80 to be much better, thus affirming the adequacy of this scale’s reliability^{63, 64)}. In this study, the ICCs for each factor were 0.89 for overwork endorsement and 0.82 for lacking overwork rewards, further supporting the reliability and stability of the overwork climate scale over one month. Notably, Cronbach’s α values were slightly higher than that of the original version (Cronbach’s α of the original version: 0.80 and 0.66, Japanese version: 0.86 and 0.80, respectively).

Convergent validity findings were also substantiated in our study. We observed a positive correlation between psychological job demands and overwork endorsement. The results suggest that when faced with demanding expectations from supervisors, employees may experience heightened pressure, leading to an inclination towards an overwork climate. Consequently, they may invest substantial efforts in their work to meet these expectations. To protect workers from exhaustion or burnout under such circumstances, it becomes imperative for supervisors to ensure the provision of adequate resources, such as support from supervisors, learning opportunities, or autonomy over their work^{19, 46)}.

Further, lacking overwork rewards, a subscale of overwork climate, displayed a negative correlation with work engagement. According to the JD-R model, work engagement thrives in environments rich in workplace resources⁴⁶⁾. These resources, including opportunities for professional development and job security, serve to mitigate the negative impact of high job demands on workers’ well-being⁴⁷⁾. However, in the context of an overwork climate, workers find themselves unrewarded for their excessive work hours and may feel a sense of deprivation regarding recognition and support from their supervisors^{46, 47)}. In such scenarios, if workers cannot augment or access necessary work resources, their level of work engagement may deteriorate^{46, 47)}.

Additionally, our findings revealed a negative correlation between psychological safety and workers’ perceptions of the overwork climate. Psychological safety, as measured by the scale employed in this study, delves into employees’ willingness to express their true feelings or opinions within the workplaces^{39, 40)}. This result indicates that employees may be reluctant to express their concerns or opinions to supervisors or upper management, even when grappling with issues such as uncompensated overtime. This reluctance stems from a recognition of the potential repercussions, such as social isolation or exclusion from the workplace. Such employee silence has garnered considerable attention from researchers and underscores the importance of fostering environments where open communication is encouraged^65,66,67).

Additionally, workaholism exhibited a positive correlation with the overwork climate. The endorsement of excessive work may fuel workaholic tendencies, as organizational factors play a significant role in promoting workaholism^{19, 20)}. In contemporary work environments, advancements in information technology have facilitated constant communication between supervisors and employees, blurring the boundaries between work and personal life. This persistent connectivity often leads employees to compulsively check their smartphones and other devices, driven by the fear of negative job evaluations, thereby exacerbating anxiety and other negative effects^{68, 69)}. Supervisors, inadvertently, can contact their subordinates at any time and from anywhere, contributing to the perpetuation of an overwork climate. Such a climate poses challenges for workers striving to maintain a healthy work-life balance, disengage from work psychologically, and cope with stress⁶⁹⁾. To avoid fostering an overwork climate, establishing guidelines prohibiting work-related communication outside of designated work hours may be necessary^{68, 69)}.

The relationships observed between the overwork climate and work engagement/workaholism in our study mirrored those reported in the original version. Specifically, we found that lacking overwork rewards had a significantly negative effect on work engagement, while overwork endorsement had a positive effect on workaholism. However, unlike the original version, the lack of overwork rewards did not exhibit a significant effect on workaholism in our study. Notably, the impact indicator for this relationship in the original version was not particularly robust (i.e., r=1.0, p<0.05). This discrepancy might be attributed to differences in sample size, with the original version comprising a larger sample¹⁸⁾ (n=791) compared to our study (n=302).

The current study has some limitations. First, the participants were recruited via an internet-based survey company, which may have introduced selection bias. Despite excluding managers, the participants had an average age of 48.0 (SD=10.6) yr, with an unmarried rate of 45.0% for the first survey. The Basic Survey on Wage Structure from the Ministry of Health, Labour and Welfare in 2021 indicates that the average age of a waged employee is 43.4 yr⁷⁰⁾, and the data from the Basic Survey of Employment Structure from the Statistics Bureau in 2017 indicate that the unmarried rate of full-time workers in their 40s was 22.2%⁷¹⁾; Our sample, on average, was 5 yr older than the general population and had a notably higher (23%) rate of unmarried individuals. This discrepancy could imply that our participants, potentially facing lower incomes compared to their peers, might have perceived themselves as under-compensated for overwork. While internet-based surveys offer a convenient and widely used data collection method, they are susceptible to issues such as sampling error, measurement error, and under-coverage error^72,73,74). In addition, sampling bias may be present, as individuals experiencing perceived overwork climates may be too occupied with work to participate. This bias could potentially lead to an underestimation of overwork climate scores in our sample. Second, we could not calculate the response rate because we relied on the survey company to recruit participants until the required sample size was reached. Third, the nature of contemporary industries, where data management policies often discourage employees from taking work home, may have caused confusion in responding to question #3, “It is considered normal for employees to take work home”. Finally, since our study solely conducted correlation analyses, the causal relationships between overwork climate and other variables remain unknown, necessitating further research on the topic.

In conclusion, the Japanese version of the Overwork Climate Scale demonstrated good reliability and validity. This scale holds promise for future research into the drivers of extended working hours in Japanese workplaces and could play a vital role in efforts to mitigate the adverse health effects associated with overwork climates.

Conflict of Interest

The authors declare no conflicts of interest.

Appendix

過重労働風土日本語版

あなたが働いている組織についてお聞きします。次の文章を読み，あなたの考えに最も近いものに〇をつけてください。

【時間外労働の奨励】

1. 私が働いている職場では，従業員が時間外労働を行うのは当たり前だと考えられている。

2. 私が働いている職場では，経営層は時間外労働を推奨している。

3. 私が働いている職場では，仕事を自宅に持ち帰るのは当たり前のことだと考えられている。

4. 私が働いている職場では，多くの従業員が決められた時間以上に働いている。

5. 私が働いている職場では，時間外労働を行うことは昇進するのに重要である。

6. 私が働いている職場では，週末に仕事をすることは当然のことと考えられている。

7. 私が働いている職場では，休日や有給休暇を取得することは難しい。

【時間外労働に対する報酬の欠如】

8. 私が働いている職場では，時間外労働は代休や手当などで十分補償されている。（逆転項目）

9. 私が働いている職場では，時間外労働は金銭的に十分補償されている。（逆転項目）

10. 私が働いている職場では，無給の時間外労働を行う必要はない。（逆転項目）

11. 私が働いている職場では，時間外労働を規制する会社のポリシーがある。（逆転項目）