• 沒有找到結果。

The influence of organizational factors on safety in Taiwanese high-risk industries

N/A
N/A
Protected

Academic year: 2021

Share "The influence of organizational factors on safety in Taiwanese high-risk industries"

Copied!
8
0
0

加載中.... (立即查看全文)

全文

(1)

The in

fluence of organizational factors on safety in Taiwanese high-risk industries

Shang Hwa Hsu

a

, Chun-Chia Lee

a,b,*

, Muh-Cherng Wu

a

, Kenichi Takano

c

aDepartment of Industrial Engineering and Management, National Chiao Tung University, 1001 Ta Hsueh Road, Hsinchu City 300, Taiwan, ROC bDepartment of Information Management, Fooyin University, 151, Chinhsueh Rd., Ta-liao, Kaohsiung County 831, Taiwan, ROC

cGraduate School of System Design and Management, Keio University, 2-15-45 Mita, Minato-ku, Tokyo 108-8345, Japan

a r t i c l e i n f o

Article history:

Received 31 January 2010 Received in revised form 25 May 2010 Accepted 22 June 2010 Keywords: Organizational factors Safety climate Safety management High-risk industry Guanxi-oriented culture

a b s t r a c t

Organizational theorists have recently begun to pay research attention to the workforces of emerging countries in Asia, especially the greater China regions. However, little research has focused on the influence of organizational factors on safety in these countries. This study hypothesized that organiza-tional factors, characterized as safety leadership perspective (management commitment, blame culture) and safety climate perspective (harmonious relationships), would influence group-level safety management, which would in turn influence individual level safety awareness and practices. To test this hypothesis, a safety climate questionnaire was distributed to ten Taiwanese plants in high-risk industries. Structural equation modeling (SEM) was used to analyze the linkages among organizational level factors, work-group-level factors, and workers’ safety awareness and practices. We found that safety leadership style and organizational harmony in Taiwanese high-risk industries can exert significant influences on work-group processes, which in turn have greater effects on individual safety awareness and practices. Thesefindings and implications can serve as a basis for safety improvement in areas of the greater China region.

Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction

One analysis of recent safety accidents of large-scale complex systems indicates that 30e40% of accidents can be attributed to organizational factors (Hollnagel & Woods, 2005). In the past several decades, organizational factors have been found by several safety climate studies to serve as antecedents of accidents and injuries (Flin, Mearns, O’Connor, & Bryden, 2000; Hofmann & Stetzer, 1996; Lee & Harrison, 2000; Tomas, Melia, & Oliver, 1999; Zohar, 1980), and have become leading indicators of performance in evaluating occupational safety climate (Flin et al., 2000).

Recently, the workforces of emerging countries in Asia, espe-cially the greater China areas (including China, Hong Kong, and Taiwan) have drawn attention from organizational theorists; however, they have paid little attention in safety managementfield. With an increase in moving foreign operated factories from developed countries to emerging countries in greater China areas, the influence of organizational factors on safety in these areas is

becoming an increasing important management issue. After review previous studies of safety management (Guldenmund, 2000; Seo, 2005), we found little research has focused on how organiza-tional factors in greater China areas might affect operaorganiza-tional safety in high-risk industries.

Therefore, this study attempts to investigate relationships between organizational factors and safety awareness and practices in greater China areas. Taiwanese high-risk industries, including chemical processing plants and semiconductor and steel foundries, were chosen as samples. For a long time, Confucianism exerts a strong influence in greater China region, which can also be called as “guanxi-oriented culture”. Previous research indicated the characteristics of guanxi-oriented culture, such as emphasizing on harmonious working relationships, loyalty and the maintenance of hierarchical order, exert strong influences on organizational structure, the interaction among members, and relationship between leaders and subordinates (Helmreich & Merritt, 1998; Hofstede, 1980, Hwang, 1987; Tsui & Farh, 1997; Tsui, Wang, Xin, Zhang, & Fu, 2004; Westwood, 1997). Taiwan is a region in which a guanxi-oriented culture associated with Confucian dynamism has been found to be important (Hofstede & Bond, 1988). Exploring the influence of organizational factors on safety management in Taiwanese high-risk industries might provide implications of safety management programs in some other guanxi-oriented counties where Confucianism continues to exert an influence.

* Corresponding author at: Chun-Chia Lee, Ph.D., Department of Information Management, Fooyin University, 151, Chinhsueh Rd., Ta-liao, Kaohsiung County 831, Taiwan, ROC. Tel.:þ886 7 7811151x6410; fax: þ886 7 7821301.

E-mail addresses:shhsu@mail.nctu.edu.tw(S.H. Hsu),dereklee@mail.nctu.edu. tw, chunchia.derek@gmail.com(C.-C. Lee),mcwu@mail.nctu.edu.tw(M.-C. Wu),

k.takano@sdm.keio.ac.jp(K. Takano).

Contents lists available atScienceDirect

Journal of Loss Prevention in the Process Industries

j o u rn a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j l p

0950-4230/$e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jlp.2010.06.018

(2)

2. Theoretical framework and hypotheses 2.1. Identifying organizational level factors

To meet the purpose of this research, we attempted to identify organizational level factors from the perspectives of safety leader-ship and safety climate, which have often been mentioned by prior studies (Barling & Zacharatos, 1999; Hofmann & Morgeson, 1999; Neal & Griffin, 2002; Neal, Griffin, & Hart, 2000; Zohar, 1980). In a guanxi-oriented culture, the leadership style tends to be pater-nalistic, a style of leadership that combines strong discipline and authority with fatherly benevolence and moral integrity couched (Farh & Cheng, 2000). An ideal paternalistic leader should also demonstrate holistic concern for their subordinates’ personal and family well-being (Farh & Cheng, 2000). Therefore, we identified two organizational factors related to safety climate in Taiwan, management commitment and blame culture. Management commitment to safety is considered critical to employee safety performance (Wiegmann, Zhang, von Thaden, Sharma, & Mitchell, 2002; Zohar, 1980), and is thought to play an important role in safety climate research in different cultures (Guldenmund, 2000). Management commitment to safety in a guanxi-oriented culture may exert a strong influence on the interaction among members and relationship between leaders and subordinates. Therefore, we will explore the roles of management commitment on safety management.

In addition, leaders in guanxi-oriented societies are inclined to reprimand subordinates for poor performance (Farh & Cheng, 2000), they may not necessarily praise them for jobs well done. Blame culture, from a western cultural perspective, may have a negative effect on the safety operation of large, complex systems in the workplace. However, there is very little empirical evidence of what influence blame culture exerts in these environments. In this study, we were interested in exploring how blame culture affect the safety performance of workers.

A guanxi-oriented culture tends to be collectivist (Helmreich & Merritt, 1998; Hofstede, 1980). Members in these guanxi-oriented cultures place a very high priority on harmonious working rela-tionships (Ho & Chiu, 1994; Hwang, 1987; Tsui & Farh, 1997; Westwood, 1997). The term, guanxi, refers to a special kind of relationship characterized by implicit rules of obligations and reciprocities (Hwang, 1987; Xin & Pearce, 1996; Yeung & Tung, 1994). Maintaining the harmonious interpersonal relationships, reciprocity and face (a kind of reputation, status or dignity) can be seen as a necessary mechanism for coping with highly non-codified social order and getting things done (Hwang, 1987; Shi & Westwood, 2000; Westwood, 1997). Therefore, we would like to explore the influence of harmonious relationships on safety management in the present study.

2.2. Hypotheses

From the concepts how an organization operates, we found top-level management makes policies to determine strategic goals and the means to achieve the goals in an organization. They formulate operating procedures to provide tactical policy action guidelines. Line managers in the work-group-level execute policies and procedures, give directives to frontline workers, and supervise the work process to ensure safe and reliable operation (Zohar & Luria, 2005). Therefore, this study postulated that organizational factors would influence group processes, which in turn would affect safety practices and safety awareness (Helmreich & Merritt, 1998; Takano, Kojima, Hasegawa, & Hirose, 2001).Fig. 1is a graphical depiction of the relationships between organizational factors and safety performance, and specified hypotheses are described as follows.

2.2.1. Linkage between organizational level factors and work-group-level factors

Upper management’s commitment to safety indicates upper-and middle-level management identifies safety as a core value of company and demonstrates positive and supportive safety atti-tudes. Because Taiwan is characterized as having a paternalistic guanxi-oriented corporate culture, upper managers are expected to act like fathers of a family and therefore set good examples for their subordinates (Farh & Cheng, 2000). Upper managers in Taiwanese plants who consider safety important actively participate in safety activities and frequently express their own concerns about safety in their companies (Hsu, Lee, Wu, & Takano, 2008). Greater upper management commitment to safety should be positively related to the supervisory practices of the company’s line managers (Hsu et al., 2008; Simard & Marchand, 1994), making them more serious about safety supervision, including task instructions and progress monitoring. We proposed the following hypothesis. Hypothesis 1a. In companies higher management commitment to safety will be positively related to safety supervision of line managers.

Furthermore, in a patriarchic guanxi-oriented culture, the authority and power of the upper manager is accepted as natural, proper, absolute, paramount, unchallengeable and inviolate (Farh & Cheng, 2000; Westwood, 1997). When upper managers are perceived as having a high commitment to safety, employees may want to meet upper management expectations by increasing their willingness to report safety problems and improving team inter-action such as coordination and communication among team members (Wiegmann et al., 2002). Therefore, the following hypotheses were proposed.

Hypothesis 1b. In companies higher management commitment to safety will be positively related to the employees’ willingness to report safety problems.

Hypothesis 1c. In companies higher management commitment to safety will be positively related to collaboration among team members.

In patriarchal guanxi-oriented cultures, employees are often blamed or punished for their mistakes or for violating safety rules (Farh & Cheng, 2000). Blame culture refers to the tendency for management to punish employees when they make mistakes. Reason (1997)has suggested that blame culture might discourage employees from reporting workplace safety problems and thus have a negative effect on employees’ safety performance. To avoid blame or punishment from the management, employees may selectively decide what news to report, sharing the good news and hiding problems. Therefore, we postulated the following hypothesis.

(3)

Hypothesis 2a. In companies that tend to blame or punish workers for their mistakes will be negatively associated with the willingness to report safety problems.

Previous studies have indicated that maintaining harmonious relationships is an important social value embedded in guanxi-oriented cultures (Ho & Chiu, 1994; Hwang, 1987; Westwood, 1997). Having a harmonious relationship refers to the extent to which there is a harmonious atmosphere among coworkers and super-visors in an organization.Helmreich and Merritt (1998)found that Taiwanese pilots placed high value on maintaining good relation-ships with managers and coworkers. Valuing harmony in rela-tionships in the workplace encourages members of an organization to develop good interpersonal relationships and mutual trust (Tsui & Farh, 1997). And trust among employees should in turn increase the willingness to report safety problems in the workplace (Reason, 1997), and also facilitate group processes, e.g., team being well-coordinated and openly sharing safety information (Hsu et al., 2008). Therefore, we proposed the following hypotheses.

Hypothesis 3a. In companies with greater harmony among team members is positively related to the willingness to report safety problems among team members.

Hypothesis 3b. In companies with greater harmony among team members is positively related to collaboration among team members.

2.3. Linkage between work-group-level factors and individual level factors

Simard and Marchand (1994)have maintained that employee safety performance improves when supervisors promote safety frequently. When supervisors give more feedback and spend more time monitoring safety in the workplace, safety performance is improved (Mattila, Hyttinen, & Rantanen, 1994). Therefore, an increased in task instruction and progress monitoring might encourage employees to comply with the safety procedures and regulations and increase the safety awareness. We proposed the following two hypotheses.

Hypothesis 4a. In companies more safety supervision will be positively related to safety awareness among the employees. Hypothesis 4b. In companies with more safety supervision will be positively related to safety practices among the employees.

Furthermore, safety reporting is critical to safety performance (Reason, 1997; Wiegmann et al., 2002). According toReason (1997), safety reporting facilitates an information sharing and organiza-tional learning for incidents occurring in the workplace, proactively preventing future incidents and accidents. Safety reporting might lead to an increase in employee safety awareness and employee safety practices. Therefore, the following hypotheses were proposed.

Hypothesis 5a. In companies more safety reporting will be posi-tively related to safety awareness of employees.

Hypothesis 5b. In companies more safety reporting will be positively related to safety practices of employees.

Finally, team collaboration plays an important role in the safe operation of process control systems and is assessed by commu-nication and coordination among team members. Several accidents in high-risk systems have been related to teamwork failures (Helmreich & Merritt, 1998). With good team collaboration, it is more likely that a group-level safety climate will be established

(Lee & Harrison, 2000). This should in turn increase employee safety awareness and improve employee safety practices.

Hypothesis 6a. In companies greater collaboration among team members will be positively related to safety awareness of employees.

Hypothesis 6b. In companies greater collaboration among team members will be positively related to safety practices of employees.

3. Method 3.1. Participants

The participants in this study were frontline workers serving in high-risk industries in Taiwan. They were the workers in four chemical processing plants, five steel foundries, and one semi-conductor foundry. We used a stratified random sampling method to select the participants; thus, the number of samples randomly selected from a department was proportional to the sizes of their departments. The questionnaires were administered during work hours. The investigators described the procedures of the study, and the process was supervised by a research team. Questionnaires were completed anonymously and collected immediately by investigators who promised absolute confidentiality. Finally, three hundred twelve samples (312) were drawn from chemical plants, 289 from steel foundries, and 89 from semiconductor foundries. Respondent characteristics are shown inTable 1.

3.2. Materials and measures

This study also used a safety climate survey questionnaire, adapted from a safety climate tool developed by Takano et al. (2001) and Takano, Tsuge, Hasegawa, and Hirose (2004). The initial questionnaire was divided into three categories: organiza-tional climate, safety management, and individual performance. Each questionnaire item uses a 5-point Likert scale, with answers ranging from 1 (strongly disagree) to 5 (strongly agree), indicating the degree to which the respondent agreed with statement of certain opinions. The reliability and validity of Taiwanese-version

Table 1

Profile of respondents from frontline employees. Demographic variable All (N¼ 690)

Frequency (%) Gender Male 649 94 Female 41 6 Age 21e30 41 6 31e40 179 26 41e50 242 35 51e60 214 31 >61 14 2 Job categories Plant services 62 9 Shopfloor 580 84 Other 48 7 Work experience <5 years 62 9 6e10 83 12 11e15 117 17 16e20 159 23 21e25 145 21 >25 124 18

(4)

questionnaire has been demonstrated in earlier studies of different high-risk industries (Hsu, 2005; Hsu et al., 2008). To meet the needs of this study, the items on that questionnaire were reorganized into the eight factors under categories defined by our above-mentioned hypotheses at organizational level, group level, and individual level. These eight factors are characterized in detail as follows.

Organizational level factors were divided into three factors: management commitment, blame culture, and harmonious rela-tionships. Management commitment included six items designed to measure the extent to which upper- and middle-level manage-ment identifies safety as a core value and demonstrates positive and supportive safety attitudes. Blame culture consisted of three items designed to measure the extent to which management blamed employees for making mistakes or for unsafe behavior. Harmonious relationship consisted of five items designed to measure the extent to which there was a harmonious work atmo-sphere among coworkers and supervisors in their organization. Work-group-level factors were divided into three factors: safety supervision, safety reporting, and team collaboration. Safety supervision consisted of six items designed to measure the extent to which supervisors continually provided instructions and moni-tored employee safety. Safety reporting consisted of three items designed to measure the extent to which employees were willing to honestly report safety problems. Team collaboration consisted of four items reflecting the extent of collaboration, including coordi-nation and communication, among team members. Individual level factors were divided into two factors: risk awareness and safety practices. Risk awareness consisted of four items measuring employee perception of risk at work. Safety practices consisted of five items measuring employee risk-taking and compliance to safety rules and procedures. All the items of eight factors are listed inAppendix.

3.3. Data analysis

A common analytical technique used for scale development is to split the data at random into two subsamples, and then to use the first subsample to explore the factor structure using exploratory factor analysis (EFA) and the second subsample to confirm the factor structure using confirmatory factor analysis (CFA) (Gerbing & Hamilton, 1996; Ghiselli, Campbell, & Zedeck, 1981; Joreskog & Sorbom, 1993). To ensure the maximum internal consistency reli-ability of the constructs, researchers suggested the relireli-ability indicators should be calculated not only the Cronbach alpha coef-ficient (Churchill, 1991; Nunnally, 1978), but also the composite reliability index and the average variance extracted (AVE) (Fernández-Muñiz, Montes-Peón, & Vázquez-Ordás, 2007; Hair, Anderson, Tatham, & Black, 1998).

The validity of the constructs will follow the procedures of concept validity analysis recommended by prior research (Fernández-Muñiz et al., 2007; Hair et al., 1998), which can be conducted by studying the content validity, convergent validity and discriminant validity. Content validity refers to the degree to which individual scale items cover the range of meanings included in the concept. In this case, the constructs were refined by means of in-depth interviews with safety experts and a pilot test in various companies (Hsu, 2005; Hsu et al., 2008). The convergent validity of a concept indicates the extent to which two measures of the same concept are correlated. This convergent validity can be analyzed by using the standardized lambda coefficients and t-values. Research suggested these coefficients must exceed 0.5 and t-values superior to 1.96 are required (Anderson & Gerbing, 1988). Discriminant validity evaluates the degree to which two conceptually similar concepts differ, which can be verified by estimating the confidence interval of the correlation existing between the proposed

constructs. Research suggested no interval of correlation containing 1.0 has good discriminant validity (Anderson & Gerbing, 1988).

After confirming the measurement model, structural equation modeling (SEM) was used to examine hypothetical relationships among organizational factors and safety performance. The study used several goodness-of-fit indices to evaluate CFA and SEM, as suggested by Joreskog and Sorbom (1993), Bentler and Bonett (1980) andMaruyama (1998): chi-square (

c

2), normed fit index (NFI); non-normed fit index (NNFI); comparative fit index (CFI); incremental fit index (IFI); and root-mean-squared error of approximation (RMSEA).Bentler (1992)suggested that NFI, NNFI, CFI, IFI scores of 0.90 or more are indicative acceptable datafit. A RMSEA value up to 0.05 indicates a good-modelfit, a value of 0.08 or less indicates a reasonable modelfit, and a value greater than 0.10 indicates poor modelfit (Joreskog & Sorbom, 1993).

4. Results

4.1. Measurement model

Maximum likelihood extraction and varimax rotation tech-niques were employed for exploratory factor analysis (EFA). At the organizational level, three factors (management commitment, harmonious relationship, blame culture) had eigenvalues greater than 1, accounting for 49% of the total variance (19%, 16%, and 14%, respectively). At the work-group level, three factors (safety super-vision, team collaboration, safety reporting) had eigenvalues greater than 1, accounting for 45% of total variance (21%, 15%, and 9%, respectively). At the individual level, two factors (safety prac-tices and safety awareness) had eigenvalues greater than 1, accounting for 42% of total variance (24% and 18%, respectively). According to the results of confirmatory factor analysis (CFA), the overall measurement modelfit was assessed as

c

2(601)¼ 1623.87

(p< 0.01). Since

c

2is affected by sample size, otherfit indices had

to be used. The RMSEA values of 0.070 (less than 0.08) indicated that the measurement model had a reasonable modelfit. Others indices (NFI¼ 0.92; NNFI ¼ 0.93; CFI ¼ 0.94; IFI ¼ 0.94) were greater than 0.9 (as mentioned criteria in data analysis section), also indicating that the measurement model was acceptable.

Descriptive statistics and Cronbach’s alpha coefficients among the various factors can be found in Table 2. Cronbach’s alpha coefficient of each factor was greater than 0.6. The composite reliability index and AVE of factors can be found inTable 3. The composite reliability index of each factor was greater than 0.6, and that the most AVE values are close to 0.5. The results indicated that internal consistency reliability indicators of the measure model were confirmed to have acceptable reliability (Bagozzi & Yi, 1988; Churchill, 1991; Nunnally, 1978). Table 3 also showed that all standardized lambda coefficients exceeded 0.5 and t-values were greater than 1.96, confirming the convergent validity of the factors. Table 4 revealed that no interval includes the value 1.0, thereby

Table 2

Descriptive statistics and Cronbach’s alpha for all constructs.

Constructs M SD Cronbach’s alpha Management commitment 3.96 0.67 0.87 Blame culture 3.62 0.71 0.68 Harmonious relationship 3.89 0.58 0.82 Safety supervision 3.93 0.64 0.86 Safety reporting 3.70 0.50 0.63 Team collaboration 3.97 0.51 0.70 Risk awareness 3.90 0.55 0.75 Safety practices 4.17 0.48 0.83

(5)

confirming the discriminant validity of the factors. Thus, the concept validity of measurement model can be confirmed.

4.2. Structural model

According to the results of SEM, the overall fit indices of the structural model was assessed as

c

2 (614)¼ 3328.63 (p < 0.01). Because

c

2tends to be affected by sample size, we used otherfit

indices. The values of RMSEA were 0.078 (less than 0.08), indi-cating the structural model was reasonable. Other indices (NFI¼ 0.92; NNFI ¼ 0.93; CFI ¼ 0.93; IFI ¼ 0.93) were greater than 0.9, indicating that the structural model had an acceptablefit. In summary, test results indicate that the structural model was adequate. The coefficients of all original structural paths were significant except the path from blame culture to safety reporting. The effect of blame culture on safety performance may be medi-ated by other variables.

We found a higher correlation coefficient between blame culture and supervision. Therefore, we tried to add the path of blame culture and supervision in structural model. Thefit indices of the modified model were

c

2 (613)¼ 3329.07, p < 0.01,

RMSEA¼ 0.078, NFI ¼ 0.92; NNFI ¼ 0.93; CFI ¼ 0.93; and IFI ¼ 0.93, indicating it was an acceptable model. The differences (

c

diff2 )

between hypothetical model and modified model were not signif-icant, suggesting that the modified model was a better choice. The standardized path coefficients in the modified model are presented inFig. 2.

Table 3

Measurement scales offirst-order confirmatory factor model. Constructs Standardized lambda (l) t-Value Composite reliability AVE Management commitment 0.86 0.51 Commitment 1 0.71 20.70 Commitment 2 0.72 21.01 Commitment 3 0.63 17.87 Commitment 4 0.66 19.14 Commitment 5 0.81 25.62 Commitment 6 0.75 22.31 Blame culture 0.66 0.41 Blame 1 0.56 12.69 Blame 2 0.69 17.84 Blame 3 0.65 13.31 Harmonious relationship 0.82 0.50 Harmonious 1 0.61 17.53 Harmonious 2 0.73 21.25 Harmonious 3 0.78 23.07 Harmonious 4 0.71 20.31 Harmonious 5 0.68 19.14 Safety supervision 0.85 0.49 Supervision 1 0.62 17.49 Supervision 2 0.73 22.01 Supervision 3 0.72 21.49 Supervision 4 0.68 19.76 Supervision 5 0.78 24.05 Supervision 6 0.67 19.62 Safety reporting 0.66 0.40 Reporting 1 0.58 14.37 Reporting 2 0.63 15.77 Reporting 3 0.69 17.15 Team collaboration 0.71 0.38 Collaboration 1 0.53 14.43 Collaboration 2 0.71 20.86 Collaboration 3 0.63 17.33 Collaboration 4 0.57 15.60 Risk awareness 0.76 0.45 Awareness 1 0.60 16.08 Awareness 2 0.72 20.06 Awareness 3 0.67 18.51 Awareness 4 0.68 18.73 Safety practices 0.84 0.53 Practices 1 0.72 19.71 Practices 2 0.82 23.76 Practices 3 0.68 18.52 Practices 4 0.74 20.26 Practices 5 0.67 18.25 Table 4

Discriminant validity offirst-order confirmatory factor model. Constructs Correlation Standard

error Confidence interval Management commitmenteblame culture 0.40 0.05 0.30e0.50 Management commitmenteharmonious relationship 0.53 0.03 0.47e0.59 Management commitmentesafety supervision 0.86 0.02 0.82e0.90 Management commitmentesafety reporting 0.37 0.04 0.29e0.45 Management commitmenteteam collaboration 0.60 0.03 0.54e0.66 Management commitmenterisk awareness 0.59 0.03 0.53e0.65 Management commitmentesafety practices 0.38 0.05 0.28e0.48 Blame cultureeharmonious relationship 0.34 0.04 0.26e0.42 Blame cultureesafety supervision 0.43 0.03 0.37e0.49 Blame cultureesafety reporting 0.11 0.04 0.03e0.19 Blame cultureeteam collaboration 0.32 0.05 0.22e0.42 Blame cultureerisk awareness 0.32 0.05 0.22e0.42 Blame cultureesafety practices 0.14 0.04 0.06e0.22 Harmonious relationshipesafety supervision 0.58 0.03 0.52e0.64 Harmonious relationshipesafety reporting 0.37 0.05 0.27e0.47 Harmonious relationshipeteam collaboration 0.67 0.03 0.61e0.73 Harmonious relationshiperisk awareness 0.61 0.03 0.55e0.67 Harmonious relationshipesafety practices 0.35 0.04 0.27e0.43 Safety supervisionesafety reporting 0.37 0.04 0.29e0.45 Safety supervisioneteam collaboration 0.64 0.03 0.58e0.70 Safety supervisionerisk awareness 0.63 0.03 0.57e0.69 Safety supervisionesafety practices 0.40 0.04 0.32e0.48 Safety reportingeteam collaboration 0.38 0.05 0.28e0.48 Safety reportingerisk awareness 0.36 0.04 0.28e0.44 Safety reportingesafety practices 0.47 0.04 0.39e0.55 Team collaborationerisk awareness 0.66 0.03 0.60e0.72 Team collaborationesafety practices 0.52 0.04 0.44e0.60 Risk awarenessesafety practices 0.42 0.04 0.34e0.50

(6)

5. Discussion

To summarize the results, we found that in companies (a) with more management commitment to safety have significantly posi-tive effects on safety supervision line managers, the willing of safety reporting, and team collaboration, which in turn have significantly positive effects on safety awareness and employee safety practices among the employees; (b) with more harmonious relationships have significantly positive relationships on safety reporting and team collaboration, which in turn have significantly positive effects on safety awareness and employee safety practices among the employees; (c) tendency to blame or punish workers for their mistakes has significantly effect on safety supervision line managers, which in turn have significantly positive effects on safety awareness and employee safety practices among the employees.

These results are in line with those reported by other safety climate studies, which indicated that management’s involvement and commitment to safety in safety management processes have critical influences on safety performance (Dedobbeleer & Beland, 1991; Flin et al., 2000; Zohar, 1980). In addition, these results are consistent with the prior research (Hsu et al., 2008; Von Thaden, Li, Li, & Lei, 2006), which indicated traditional social values, social harmony and reciprocity in guanxi-oriented culture play important roles in influencing safety management and employees’ safety attitudes and behavior. Contrary to our hypothesis, we did notfind in companies that tended to blame or punish workers for their mistakes have significantly effect on the willingness of work to report safety prob-lems. The effect of blame culture on safety reporting may be moderated by other variables, such as harmonious relationships.

This study has several practical implications for improving safety climate in Taiwanese high-risk industries as well as in paternalistic guanxi-oriented companies. First, the leadership style of Taiwanese management tends to be top-down paternalistic leadership (Farh & Cheng, 2000; Hsu et al., 2008). Although we found such leadership can increase positive effects on safety awareness and safety practices, we suggest that any safety improvement program for Taiwanese plants should place increased emphasis on the importance of behavior-based safety (DeJoy, 2005; DePasquale & Geller, 1999). To improve the safety performance, employees should be empowered by management. When people feel be empowered, their sense of personal responsibilities and accountability are increased and perform actively caring behavior (Geller, 2001). Therefore, management may consider taking bottom-up involvement approach to enhance safety management besides top-down directive safety leadership.

Secondly, blame culture can also have a positive effect on safety supervision in guanxi-oriented companies.Geller (2001)indicates blaming an individual or group for an injury-producing incident is not consistent with a systems approach to safety. Instead, an inci-dent provides an opportunity to gather facts from all aspects of the system which might have contributed to the incident. Therefore, management should turn their thoughts from faultingfinding to factfinding. Reward systems might consider focusing on workplace hazards rather than human errors. Avoid mentioning the people involved in an incident, as a means of saving the face, can be more acceptable to employees in harmonious organizations.

Thirdly, as above-mentioned states, maintaining the harmo-nious relationship is an important social value embedded in guanxi-oriented cultures (Ho & Chiu, 1994; Hwang, 1987; Westwood, 1997). However, maintaining the harmonious rela-tionship should transfer effectively into building trust relation-ships. Building a trust relationship is a prerequisite of improving teamwork effectiveness (Geller, 2001). Good interpersonal rela-tionships can increase group cohesiveness, which facilitates group collaboration, information sharing, and safety reporting. Therefore,

safety training should put more emphasis on how the trust rela-tionship of teamwork is built. They may develop effective team training program on safety through implementing Crew Resource Management (CRM). CRM has been widely used to improve the operation of teamwork (Helmreich & Foushee, 1993). It emphasizes the role of human factors in high-stress, high-risk environments, and encompasses team training as well as interactive group debriefings and measurement of team performance.

The current study has several potential limitations. One limita-tion is that this study is a cross-seclimita-tional study. It clearly limits the degree of inference on relationships among organizational factors. To determine the relationships, longitudinal studies should be performed to study the effects over time. The second limitation is that our samples are drawn from workers in Taiwanese high-risk industries only. Although people in other greater China regions share the same cultural characteristics as Taiwanese, they differ in some extent. Therefore, further comparative studies in other greater China regions are needed. Thirdly, the group and organizational structure of the sample is not incorporated in the analysis. Con-ducting the study within a single strong culture might actually reduce the variance in variables and therefore reduce the chance of observing important effects. Finally, some cultural factors, such as paternalistic leadership and guanxi-oriented culture, are not con-ducted in the analysis. Paternalistic leadership has been found to have a significant effect on employees’ behavior in guanxi-oriented cultures (Farh & Cheng, 2000; Westwood, 1997). What effect these factors may have on safety climate may need further study. Never-theless, the empirical framework of this study is valuable because it provides a systematic method of investigating how organizational factors influence employees’ awareness and practices through work-group-level factors in a guanxi-oriented culture.

6. Conclusion

This study identifies the valuable mechanisms through which organizational level factors influence individual awareness of safety and safety practices in Taiwanese high-risk industries. Our results support our hypotheses that safety leadership style (management commitment, blame culture) and organizational harmony (harmonious relationships) can exert significant influences on work-group processes, which in turn have greater effects on indi-vidual safety awareness and practices. Thefindings and implica-tions of this study can be used to improve safety management programs to accommodate organizational characteristics inherent in guanxi-oriented cultures.

Appendix

Questionnaire items for this research 1. Management commitment to safety (6 items)

1.1 Management places a high priority on safety operations in company.

1.2 Management cares about the safety welfare of their employees.

1.3 Management works to upgrade the safety of its facilities or reduce safety problems.

1.4 Management provides resources to prevent the occurrence of safety-related incidents.

1.5 Management gets personally involved in safety training. 1.6 Management holds quality management activities about

safety at workplace. 2. Blame culture (3 items)

2.1 Management seldom praises employees for good safety behavior.

(7)

2.2 Management punishes employees involved in safety-related incidents or near misses.

2.3 Management blames employees for making mistakes, regardless of whether or not an incidence occurs. 3. Harmonious relationship (5 items)

3.1 Supervisors and workers can discuss any issue openly in the workplace.

3.2 Team members maintain a harmonious atmosphere among themselves.

3.3 A high priority is placed on maintaining harmonious rela-tionships at work.

3.4 There are good interpersonal relationships in the workplace. 3.5 The team members at workplace are open-minded. 4. Safety supervision (5 items)

4.1 Supervisor revises related safety rules periodically. 4.2 Supervisor frequently holds safety training activities in the

workplace.

4.3 Supervisor frequently moves around inspecting the workplace.

4.4 Supervisor diligently reviews the safety behaviors of the employees.

4.5 Supervisor reports cases or shares safety-related experi-ences in the workplace.

4.6 Supervisor makes ongoing safety instruction at workplace. 5. Safety reporting (3 items)

5.1 Coworkers are willing to report workplace safety problems to management.

5.2 I often report to management or coworkers when receiving useful safety information.

5.3 Coworkers are willing to making reports to upper management regarding safety mistakes of other coworkers. 6. Team collaboration (4 items)

6.1 Team members help each otherfinish their work. 6.2 There are good communications among team members. 6.3 There are clear task assignments among team members. 6.4 Teamwork is well-coordinated.

7. Risk awareness (4 items)

7.1 I am aware of coworkers who do not comply with the safety rules and procedures.

7.2 I ask my supervisors when I have safety concerns at work. 7.3 I stop working if I am not sure that it is safe to continue. 7.4 When in doubt about safety is in question, I proceed with

great caution. 8. Safety practices (5 items)

8.1 I comply with safety rules and procedures.

8.2 I routinely review standard operating procedures before starting work.

8.3 I actively look for areas in which the facilities may be unsafe.

8.4 I keep myself in a good mental and physical state. 8.5 I often disregard safety rules or procedures in order to

catch up on work.

References

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: a review and recommend two-step approach. Psychological Bulletin, 103, 411e423. Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models.

Journal of the Academic of Marketing Science, 16, 74e94.

Barling, J., & Zacharatos, A. (1999). High-performance safety system: Management practices for achieving optimal safety performance. Paper presented at the 25th annual meeting of the Academy of Management, Toronto.

Bentler, P. M. (1992). On thefit of models to covariances and methodology to the bulletin. Psychological Bulletin, 112, 400e404.

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness-of-fit in the analysis of covariance structures. Psychological Bulletin, 88, 588e606.

Churchill, G. A. (1991). Marketing research: Methodological foundation (5th ed.). New York: The Dryden Press.

Dedobbeleer, N., & Beland, F. (1991). A safety climate measure for construction sites. Journal of Safety Research, 22, 97e103.

DeJoy, D. M. (2005). Behavior change versus culture change: divergent approaches to managing workplace safety. Safety Science, 43, 105e129.

DePasquale, J. P., & Geller, E. S. (1999). Critical success factors for behavior-based safety: a study of 20 industry-wide applications. Journal of Safety Research, 30, 237e249.

Farh, J. L., & Cheng, B. S. (2000). A cultural analysis of paternalistic leadership in Chinese organizations. In J. T. Li, A. S. Tsui, & E. Weldon (Eds.), Management and organizations in the Chinese context (pp. 84e127). New York: Macmillan. Fernández-Muñiz, B., Montes-Peón, J. M., & Vázquez-Ordás, C. J. (2007). Safety

management system: development and validation of a multidimensional scale. Journal of Loss Prevention in the Process Industries, 20, 52e68.

Flin, R., Mearns, K., O’Connor, P., & Bryden, R. (2000). Measuring safety climate: identifying the common features. Safety Science, 34, 177e193.

Geller, E. S. (2001). Working safe: How to people actively care for health and safety. Boca Raton: CRC Press.

Gerbing, D. W., & Hamilton, J. G. (1996). Viability of exploratory factor analysis as a precursor to confirmatory factor analysis. Structural Equation Modeling, 3, 62e72. Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981). Measurement theory for the

behavioral sciences. San Francisco: W.H. Freeman and Co.

Guldenmund, F. W. (2000). The nature of safety culture: a review of theory and research. Safety Science, 34, 215e257.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis. Upper Saddle River: Prentice-Hall International Inc.

Helmreich, R. L., & Foushee, H. C. (1993). Why crew resource management? Empirical and theoretical bases of human factors in training and aviation. In E. Wiener, B. G. Kanki, & R. L. Helmreich (Eds.), Cockpit resource management (pp. 3e45). San Diego, CA: Academic Press.

Helmreich, R. L., & Merritt, A. C. (1998). Culture at work in aviation and medicine: National, organizational, and professional influences. Aldershot: Ashgate. Ho, D. Y. F., & Chiu, C. Y. (1994). Component ideas of individualism, collectivism, and

social organization: an application in the study of Chinese culture. In U. Kim, H. C. Triandis, C. Kagitcibasi, S. C. Choi, & G. Yoon (Eds.), Individualism and collectivism: Theoretical and methodological issues (pp. 137e156). Thousand Oaks, CA: Sage.

Hofmann, D. A., & Morgeson, F. P. (1999). Safety-related behavior as social exchange: the role of perceived organizational support and leader-member exchange. Journal of Applied Psychology, 84, 286e296.

Hofmann, D. A., & Stetzer, A. (1996). A cross-level investigation of factors influ-encing unsafe behaviors and accidents. Personnel Psychology, 49, 307e339. Hofstede, G. (1980). Culture’s consequences: International differences in work-related

values. Newbury Park, CA: Sage.

Hofstede, G., & Bond, M. H. (1988). The Confucius connection: from cultural roots to economic growth. Organizational Dynamics, 16, 4e21.

Hollnagel, E., & Woods, D. D. (2005). Joint cognitive systems: Foundations of cognitive systems engineering. Boca Raton, FL: CRC Press.

Hsu, S. H. (2005). Establishing safety knowledge management system for improving safety culture within high-risk technical industry (Report no. NSC 91-2213-E-009-103). Taiwan, ROC: National Science Council.

Hsu, S. H., Lee, C. C., Wu, M. C., & Takano, K. (2008). A cross-cultural study of organizational factors on safety: Japanese vs. Taiwanese oil refinery plants. Accident Analysis and Prevention, 40, 24e34.

Hwang, K. K. (1987). Face and favor: the Chinese power game. American Journal of Sociology, 92, 944e974.

Joreskog, K., & Sorbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Chicago: Scientific Software International, Inc. Lee, T., & Harrison, K. (2000). Assessing safety culture in nuclear power stations.

Safety Science, 34, 61e97.

Maruyama, G. M. (1998). Basics of structural equation modeling. Thousand Oaks, CA: Sage Publications.

Mattila, M., Hyttinen, M., & Rantanen, E. (1994). Effective supervisory behavior and safety at the building site. International Journal of Industrial Ergonomics, 13, 85e93.

Neal, A., & Griffin, M. A. (2002). Safety climate and safety behavior. Australian Journal of Management, 27, 67e76.

Neal, A., Griffin, M. A., & Hart, P. M. (2000). The impact of organizational climate on safety climate and individual behavior. Safety Science, 34, 99e109.

Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill. Reason, J. (1997). Managing the risks of organizational accidents. Aldershot: Ashgate. Seo, D. C. (2005). An explicative model of unsafe work behavior. Safety Science, 43,

187e211.

Shi, X., & Westwood, R. I. (2000). International business negotiation in the Chinese context. In J. T. Li, A. S. Tsui, & E. Weldon (Eds.), Management and organizations in the Chinese context (pp. 185e221). New York: Macmillan.

Simard, M., & Marchand, A. (1994). The behavior offirst-line supervisors in accident prevention and effectiveness in occupational safety. Safety Science, 17, 169e185. Takano, K., Kojima, M., Hasegawa, N., & Hirose, A. (2001). Interrelationships between organizational factors and major safety indicators: a preliminaryfield study. In B. Wilpert, & N. Itoigawa (Eds.), Safety culture in nuclear power oper-ations (pp. 189e205). London: Taylor & Francis.

Takano, K., Tsuge, T., Hasegawa, N., & Hirose, A. (2004). Development of a safety assessment system for promoting a safe organizational climate and culture. In

(8)

N. Itoigawa, B. Wilpert, & B. Fahlbruch (Eds.), Emerging demands for nuclear safety of nuclear power operations: Challenge and response (pp. 57e71). Boca Raton: CRC Press.

Tomas, J. M., Melia, J. L., & Oliver, A. (1999). A cross-validation of a structural equation model of accidents: organizational and psychological variables as predictors of work safety. Work and Stress, 13, 49e58.

Tsui, A. S., & Farh, J. L. (1997). Where guanxi matters: relational demography and guanxi in the Chinese context. Work and Occupations, 24, 56e79.

Tsui, A. S., Wang, H., Xin, K., Zhang, L., & Fu, P. P. (2004). Let thousandflowers bloom: variation of leadership styles among Chinese CEOs. Organizational Dynamics, 33, 5e20.

Von Thaden, T. L., Li, Y. J., Li, J., & Lei, D. (2006). Validating the commercial aviation safety survey in the Chinese context. Technical report HFD-06-09. Savoy, IL: Human Factors Division, University of Illinois.

Westwood, R. (1997). Harmony and patriarchy: the cultural basis for“paternalistic headship” among the overseas Chinese. Organization Studies, 18, 445e480. Wiegmann, D. A., Zhang, H., von Thaden, T., Sharma, G., & Mitchell, A. (2002). A

synthesis of safety culture and safety climate research (ARL-0203/FAA-02-2). Savoy, IL: University of Illinois.

Xin, K. R., & Pearce, J. L. (1996). Guanxi: connections as substitutes for formal institutional support. Academy of Management Journal, 39, 1641e1658. Yeung, Y. M., & Tung, R. L. (1994). Achieving business success in Confucian societies:

the importance of guanxi (connections). Organizational Dynamics, 25, 54e65. Zohar, D. (1980). Safety climate in industrial organizations: theoretical and applied

implications. Journal of Applied Psychology, 65, 96e102.

Zohar, D., & Luria, G. (2005). A multilevel model of safety climate: cross-level relationships between organization and group-level climates. Journal of Applied Psychology, 90, 616e628.

數據

Fig. 1. The proposed structural model of the present research.
Fig. 2. Results of the structural model with standardized path coefficients. *p &lt; 0.01.

參考文獻

相關文件

Therefore, this study is focusing on designing the bicycle traffic safety Lesson Plan to enhance the bicycle riding safety of students.. Through the pre-teaching test and the

Based on the defects of the safety control in the semiconductor electric circuit industry and the application of swarm Intelligence and knowledge management SECI model, the

implementation of public safety for the buildings under comprehensive reviews on fire prevention and evacuation and compares with related literature to establish the check items

【47】Thompson, Richard C; Hilton, Thomas F; Witt, L.Alan, “Where the Safety rubber meets the shop floor: A confirmatory model of management influence on workplace safety”.Jouranl of

In terms of external cognitive factors, this research confirmed that assurance, apathy and price reasonability as part of the service quality dimension have influence on

We have to discuss the influence of flood probability and Regional Drainage.We have to notice the property and the safety of people on campus, so my studies analyze

Therefore, this study proposes to unify the implementation schedule of the traffic safety education through adopting “Road Safety Education Week” in the school

Considering the safety of construction and the limitation of the narrow site, we build these land-improving supporters on the inner side of the slurry wall, and through the