Safety Science 65 (2014) 10–19
Contents lists available at ScienceDirect
journal homepage: www.elsevier .com/locate /ssc i
Relationships between safety climate and safety performance of building repair, maintenance, minor alteration, and addition (RMAA) works
0925-7535/$ – see front matter � 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ssci.2013.12.012
⇑ Corresponding author. Tel.: +852 2766 5872; fax: +852 2764 5131. E-mail address: firstname.lastname@example.org (C.K.H. Hon).
Carol K.H. Hon ⇑, Albert P.C. Chan, Michael C.H. Yam Department of Building and Real Estate, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China
a r t i c l e i n f o
Article history: Received 22 June 2012 Received in revised form 5 November 2013 Accepted 16 December 2013 Available online 21 January 2014
Keywords: Safety climate Safety performance Repair and maintenance
a b s t r a c t
The importance of repair, maintenance, minor alteration, and addition (RMAA) works is increasing in many built societies. When the volume of RMAA works increases, the occurrence of RMAA accidents also increases. Safety of RMAA works deserves more attention; however, research in this important topic remains limited. Safety climate is considered a key factor that influences safety performance. The present study aims to determine the relationships between safety climate and safety performance of RMAA works, thereby offering recommendations on improving RMAA safety. Questionnaires were dispatched to private property management companies, maintenance sections of quasi-government developers and their subcontractors, RMAA sections of general contractors, small RMAA contractors, building ser- vices contractors and trade unions in Hong Kong. In total, data from 396 questionnaires were collected from RMAA workers. The sample was divided into two equal-sized sub-samples. On the first sub-sample SEM was used to test the model, which was validated on the second sub-sample. The model revealed a significant negative relationship between RMAA safety climate and incidence of self-reported near misses and injuries, and significant positive relationships between RMAA safety climate and safety participation and safety compliance respectively. Higher RMAA safety climate was positively associated with a lower incidence of self-reported near misses and injuries and higher levels of safety participation and safety compliance.
� 2014 Elsevier Ltd. All rights reserved.
Repair, maintenance, minor alteration, and addition (RMAA) works have been largely overlooked during the construction mar- ket boom. In fact, the volume of RMAA works often accounts for a considerable size of the total construction volume in many devel- oped societies. For example, the RMAA sector accounted for an average of 50.2% of the construction volume in Hong Kong from 2006 to 2010 (Census and Statistics Department, 2007, 2008, 2009, 2010, 2011). The RMAA sector is expected to expand further due to the rising concerns for the safety of aging buildings and sus- tainability in the built environment. Repair and maintenance of dilapidated buildings is needed to protect the safety of the occu- pants and the public; whereas remodeling and retrofitting is needed to preserve or upgrade the building value (Yiu, 2007). With the rising importance of the RMAA sector, safety problems of this sector deserve more attention (Hon et al., 2010). The RMAA sector accounted for six out of ten (66.7%) fatal cases in the construction industry of Hong Kong in 2010. The RMAA sector accounted for 44.7% of accidents in the construction industry in 2011 while it
only accounted for 39.2% of the construction volume in the same period (Legislative Council, 2011a,b). Research into safety of the RMAA sector; however, remains scarce.
Unsafe behavior is a decisive factor for accident to occur (Rea- son, 1995). Unsafe behavior often occurs because safety measures are likely to entail modest benefits but immediate costs, such as slower pace, extra effort or personal discomfort. If the likelihood of injury is underestimated in a seemingly safe environment, the expected utility of the unsafe behavior exceeds that of the safe behavior. Unsafe behavior is also naturally reinforced because peo- ple tend to place higher value on short-term results. In this sense, deterring unsafe behavior is a significant managerial challenge (Zohar, 2002).
Since RMAA works mainly involve labor rather than machines, most of the accidents occurred because of unsafe behavior rather than machine failure. However, unsafe behavior is only an ostensi- ble cause or symptom, and other more fundamental factors also need to be considered. For example, building design affects safety of construction workers. As suggested by Behm (2005), safety haz- ards are often ‘‘designed into’’ the construction projects. A holistic approach of accident causation should be adopted (Reason, 1997). Broader organizational and contextual factors leading to unsafe behavior should not be neglected. A behavioral approach, which considers how employees think, behave, respond to situations,
C.K.H. Hon et al. / Safety Science 65 (2014) 10–19 11
and how the work environment impacts upon personnel attitudes and behavior, would likely be more effective in managing safety of RMAA works (Lingard and Rowlinson, 2005).
Safety climate has been a useful construct to improve safety in the past few decades (Zohar, 2010). A handful of research studies show that a positive relationship between safety climate and safety performance exists in construction. For example, Mohamed (2002), Chan et al. (2005), and Choudhry et al. (2009) have successfully established a positive relationship between safety climate and safety performance on construction projects; however, little re- search has been done in the RMAA sector, which is increasingly important not only in Hong Kong, but also in other developed societies.
Safety practices of RMAA works differ from those in new con- struction works. Most RMAA contracting companies are small/med- ium-sized specialty contractors of RMAA works. The small/ medium-sized companies often have limited resources for safety (Lamm, 1997). Unlike greenfield projects, RMAA job sites are often found in occupied buildings (Chan et al., 2010). RMAA workers may underestimate the risks of working in an occupied environment which does not resemble a construction site. Small size and widely scattered locations of RMAA projects make safety supervision more difficult, inefficient, and costly than those of new works. Close safety supervision on a RMAA contract with small contract sum and short duration of work is not cost effective (Hon et al., 2012). In light of these subtle differences, previous safety climate research findings on new construction projects may not be fully relevant to RMAA works. The relationships between safety climate and safety performance of the RMAA sector require further investigation.
This paper reports part of the findings of a wider scope safety research project on RMAA works in Hong Kong. It aims to deter- mine the relationships between safety climate and safety perfor- mance of RMAA projects. The current study fills the knowledge gap of limited safety climate research in the RMAA sector of con- struction. A model unveiling the relationship of safety climate and safety performance of RMAA works would be useful for safety professionals in the industry to measure, monitor, and improve the safety performance of RMAA works.
2. Safety climate
Zohar (1980) applied the concept of behavioral climate for safety and produced a seminal paper on safety climate in the early 80s. Since then, safety climate has been widely applied in different contexts. Zohar (1980, p. 96) defines safety climate as ‘‘a summary of molar perceptions that employees share about their work envi- ronments. . . a frame of reference for guiding appropriate and adap- tive task behaviors’’. As stated by Zohar (2003), safety climate reflects the true perceived priority of safety in an organization. Some researchers defined safety climate as a current-state reflec- tion of the underlying safety culture (e.g., Mearns et al., 2001, 2003).
There is little consensus on the number and content of safety climate factors. Flin et al. (2000) identified five most frequently- occurring factors from 18 safety climate scales of different indus- tries, they were: management/supervision, the safety system, risk, work pressure and competence. As reviewed by Hon et al. (2013), management commitment to safety, safety rules and procedures, and workers’ involvement in safety, were the three most common safety climate factors found in construction (Dedobbeleer and Béland, 1991; Mohamed, 2002; Fang et al., 2006; Choudhry et al., 2009; Zhou et al., 2011). Safety climate studies in the construction industry have been focusing on new construction projects (e.g. Chan et al., 2005; HSE, 2012) but our understanding of the safety climate of RMAA works is largely unrealised.
3. Safety performance
Earlier safety studies tended to use statistical data of accidents or injuries to measure safety performance. By contrast, apart from actual injury records, more recent studies have also used alterna- tive data such as self-reported injury data collected through ques- tionnaires (e.g. Siu et al., 2004; Huang et al., 2006) and self- reporting has been shown to be a reliable and valid source of injury data (Begg et al., 1999; Gabbe et al., 2003) According to Gabbe et al. (2003), the accuracy of self-reported injuries could be as high as 80%. However, accidents or injuries are reactive measures and are relatively infrequent. They may not be effective indicators of safety because they only reflect occurrences of failures (Cooper and Phillips, 2004). They are also ‘‘insufficiently sensitive, of dubi- ous accuracy, retrospective, and ignore risk exposure’’ (Glendon and Litherland, 2001, p. 161). Lingard et al. (2011) have also re- ported that injuries resulting in lost time and medical treatment occur infrequently and are ineffective indicators of safety perfor- mance. They suggested using a more fine-grained measure of workgroup safety performance, such as micro-accidents or minor (non-reportable) injuries in future research. According to Beus et al. (2010, p. 717) ‘‘safety climate should be more effective in pre- dicting injuries of a less serious nature’’. It is because minor inju- ries, which often come before serious ones, are more proximal to safety climate than serious injuries.
In light of the deficiency in using injury as a proxy of safety per- formance, a growing number of studies have attempted to use safety behavior as a measure of safety performance. Safety perfor- mance can be defined as evaluative ‘‘actions or behaviors that indi- viduals exhibit in almost all jobs to promote the health and safety of workers, clients, the public, and the environment’’ (Burke et al., 2002, p. 432). According to Neal and Griffin (2004), safety perfor- mance can be measured with safety compliance and safety participation.
Safety compliance is defined by Griffin and Neal (2000) as fol- lowing rules in core safety activities. This includes ‘‘obeying safety regulations, following correct procedures, and using appropriate equipment’’ (Neal and Griffin, 2004, p. 16). It refers to ‘‘the core activities that individuals need to carry out to maintain workplace safety. These procedures include adhering to standard work proce- dures and wearing personal protective equipment’’ (Neal and Griffin, 2006, p. 947). Safety participation refers to ‘‘behaviors that do not directly contribute to an individual’s personal safety but that do help to develop an environment that supports safety’’ (Neal and Griffin, 2006, p. 947).
4. Relationships between safety climate and safety performance
4.1. Theoretical linkages
Social exchange theory and expectancy-valence theory are two theoretical mechanisms that may help to explain and predict the relationship between safety climate and safety behavior (Neal and Griffin, 2006). Social exchange theory postulates that, when an organization cares for the well-being of employees (i.e., the organization has a positive safety climate), the employees are likely to develop implicit obligations to perform duties, using behavior beneficial to the organization. Apart from their standard core work duties, they also perform organizational citizenship behavior, i.e., extra-role functions other than core work activities. Hofmann and Morgeson (1999) have found that when an organiza- tion emphasizes safety, its employees reciprocate by complying with established safety procedures (Neal and Griffin, 2006).
The expectancy-valence theory postulates that motivation is a combination of employees’ valence, expectancy and instrumentality.
12 C.K.H. Hon et al. / Safety Science 65 (2014) 10–19
Valence is the depth of want that an employee has for extrinsic (e.g. promotion) or intrinsic (e.g. satisfaction) rewards. Expectancy refers to level of confidence an employee is capable of achieving the goal. Instrumentality is an employee’s perception of being rewarded as promised by the management (Vroom, 1964). In terms of safety per- formance, employees will behave safely when they perceive that such behavior will bring valued intrinsic or extrinsic outcomes. When an organization truly values safety, there is a high level of safety climate in the organization. Based on behavior-outcome expectancies, employees are likely to behave safely because they ex- pect that their safety behavior would be rewarded and such behav- ior would bring a valuable outcome to them (Neal and Griffin, 2006).
4.2. Empirical relationships
The influence of safety climate on safety performance varies across different work settings and environments. Some studies found significant relationship between safety climate and safety performance (Gillen et al., 2002; Siu et al., 2004; Pousette et al., 2008) whereas some did not (Glendon and Litherland, 2001; Coo- per and Phillips, 2004). Comprehensive meta-analysis studies of Clarke (2006) and Christian et al. (2009) on safety climate and safety performance indicated that safety climate is a significant factor affecting safety performance. Clake’s study (2006) revealed that safety climate and safety performance is consistently posi- tively-related in prospective studies. A more recent meta-analysis study of Christian et al. (2009) also established significant relation- ships between safety climate and safety participation and safety compliance respectively. Safety climate helps to raise safety moti- vation and safety knowledge, leading to safer behavior and fewer accidents and injuries.
5. Research hypotheses
Based on the literature, a research model showing three re- search hypotheses for empirical testing is shown in Fig. 1.
The safety climate generally shows the importance of safety perceived by the employees in an organization. The level of safety climate affects the safety behavior and safety attitudes of employ- ees in an organization. When safety perceptions are more favor- able, employees are less likely to engage in unsafe acts, resulting in a lower chance of injury (Clarke, 2006). The first hypothesis, Hypothesis 1 (H1), can thus be generated as: RMAA safety climate is negatively related to self-reported near misses and injuries.
Safety participation is more on voluntary basis, and, perhaps outside of one’s formal role. When managers and supervisors dem- onstrate their commitment to safety, their subordinates are more likely to reciprocate by participating in safety activities. The more positive the safety climate, the higher the level of the safety partic- ipation (Clarke, 2006), as hypothesized in Hypothesis 2 (H2): RMAA safety climate is positively related to safety participation.
Near misses & injuries
RMAA safety climate
F1 Management commitment
F2 Safety rules
F3 Safety responsibility
Fig. 1. Research model and hypotheses. Note. F1 = Management commitment to OHS and employee involvement. F2 = Applicability of safety rules and work practices. F3 = Responsibility for health and safety. H1 = Hypothesis 1. H2 = Hypoth- esis 2. H3 = Hypothesis 3.
A higher level of safety climate may imply better safety man- agement, safety knowledge and awareness of safety within the company. In such a case, people are more likely to comply with safety rules and regulations (Clarke, 2006). Thus, Hypothesis 3 (H3) can be generated as: RMAA safety climate is positively related to safety compliance.
6. Research methods
6.1. Questionnaire design
A questionnaire was designed to explore the relationship be- tween safety climate and safety performance of RMAA works. The questionnaire consisted of three parts. Part A asked 13 ques- tions on personal attributes. Part B adopted 38 questions of the Safety Climate Index (SCI) survey developed by the Occupational Safety and Health Council (OSHC) of Hong Kong to measure safety climate of RMAA works. SCI was selected because it was written in both English and Chinese and was designed in the context of the construction industry of Hong Kong. Its validity and reliability had been verified by prior research of the OSHC. Part C consisted of three broad indicators to measure safety performance: injuries, safety participation, and safety compliance.
6.1.1. RMAA safety climate A second-order safety climate model for RMAA works which
consisted of three factors encapsulating 22 questions (Appendix A) of the Safety Climate Index (SCI) survey of the Occupational Safety and Health Council (OSHC) of Hong Kong (OSHC, 2008) was employed. Detailed development of the second-order RMAA safety climate model has been reported in Hon et al. (2013). Three RMAA safety climate factors were: (F1) Management commitment to occupational health and safety (OHS) and employee involvement; (F2) Applicability of safety rules and work practices; and (F3) Respon- sibility for health and safety. These questions were evaluated by the respondents in a five-point Likert scale, with ‘‘1’’ being ‘‘strongly disagree’’ and ‘‘5’’ being ‘‘strongly agree’’. Measurement of safety climate is assumed at the individual level rather than at group le- vel. According to James (1982, p. 219), the appropriate unit of the- ory for climate is individual because ‘‘climate involves a set of macro perceptions that reflect how environments are cognitively represented in terms of their psychological meaning and signifi- cance to the individual’’. Cronbach’s Alpha of the RMAA safety cli- mate was .89. Cronbach’s Alpha values of safety climate factors F1 to F3 were .88, .79 and .67 respectively.
6.1.2. Self-reported near misses and injuries Four questions were utilized to capture near misses and occupa-
tional injuries of the respondents in the last 12 months with a 5-point ordinal scale (0 = Never; 1 = 1 time; 2 = 2–3 times; 3 = 4–5 times; 4 = Over 5 times). The questions were: ‘‘How many times have you exposed to a near miss incident of any kind at work?’’; ‘‘How many times have you suffered from injury of any kind at work, but did not require absence from work?’’; ‘‘How many times have you suffered from injury, which require absence from work not exceeding 3 consecutive days?’’ and ‘‘How many times have you suffered from injuries, which require absence from work exceeding 3 consecutive days?’’. The questions were set in an ascending degree of injury severity with reference to the existing injury reporting requirement to the Labor Department. Since none respondents experienced injuries which require absence from work exceeding 3 consecutive days, this variable was excluded from the data analysis. Cronbach’s Alpha of the near misses and injuries was .80.
C.K.H. Hon et al. / Safety Science 65 (2014) 10–19 13
6.1.3. Safety participation Two statements from Neal and Griffin (2006, p. 953) were
modified to measure safety participation of the respondents with a 5-point ordinal scale (Appendix B). Having considered that small RMAA projects may not have formal safety programs, one of the statements listed in Neal and Griffin (2006, p. 953), ‘‘I promote the safety program within the organization’’, was not selected. With examples given to enhance clarity, the two selected state- ments were posed as questions regarding the frequency of putting extra effort to improve safety of the workplace, and the frequency of voluntarily carrying out tasks or activities to improve workplace safety. Cronbach’s Alpha of the safety participation was .73.
6.1.4. Safety compliance Two questions adopted from Mohamed (2002) were utilized to
measure in terms of time (0–100%) the degree of safety compliance to all safety procedures by the respondents and their co-workers respectively (Appendix B). The first question was regarding the percentage of time the respondents follow all of the safety proce- dures for the jobs or tasks that the respondents perform, whereas the second question was regarding the percentage of time their coworkers follow all of the safety procedures for the jobs or tasks that they perform. Cronbach’s Alpha of the safety compliance was .88.
6.2. Participants and procedures
The survey was administered between April and August in 2009. A pilot questionnaire was reviewed by 13 advisory group members in a focus group meeting. These advisory group members provided advice and industrial support for the research team. They were well experienced, including senior management of clients, RMAA contracting companies, property management companies, and government officials concerning OHS. First, a sampling frame- work consisting of clients, property management companies, RMAA contractors and subcontractors was designed. Then ques- tionnaires were dispatched through industrial links of the advisory group members. With their facilitation, several private property management companies, maintenance sections of quasi-govern- ment developers and their subcontractors, RMAA sections of gen- eral contractors, small RMAA contractors, building services contractors and trade unions in Hong Kong participated in this study. In total, 844 questionnaires were sent out and 814 of them were duly returned from managers, supervisors and workers. The response rate was 96.3%. In order to establish the relationships be- tween safety climate and safety performance of workers in the RMAA sector, a total of 396 completed questionnaires of frontline workers were selected for analysis in this paper.
6.3. Data analysis
Quantitative survey data were analyzed with statistical pack- ages SPSS 18.0 (SPSS Inc., Chicago, IL, USA) for descriptive
Table 1 Mean, standard deviation, and correlation of latent variables.
Mean Standard deviation F1
F1 Management commitment 3.90 (3.89) 0.46 (0.45) F2 Safety rules 3.15 (3.19) 0.71 (0.64) .52** (.46**) F3 Safety responsibility 3.55 (3.61) 0.67 (0.65) .46** (.48**) Near misses and injuries 1.31 (1.24) 0.62 (0.54) �.25** (�.08) Safety participation 3.35 (3.44) 1.13 (1.08) .20** (.12) Safety compliance 4.48 (4.55) 0.50 (0.48) .48** (.43**)
Note. Values not in parentheses are from the calibration sub-sample. Values in parenthe * P < .05. ** P < .01.
statistical analysis and LISREL 8.80 (Jöreskog and Sörbom, 2006) for structural equation modeling (SEM).
SEM technique was employed to test the relationship between safety climate and safety performance of RMAA works. SEM was selected because it can examine a series of separate, but interde- pendent, multiple regression equations simultaneously by specify- ing the structural model. It can take into account of latent variables and provide explicit estimates of error variance parameters (Byrne, 2009).
As for this study, safety climate is a latent variable that cannot be directly observed and measured. With SEM, a hypothetical mod- el with multiple latent variables of safety climate and safety per- formance was constructed and tested against empirical data. Interdependencies of observed variables and latent variables were estimated simultaneously. As measurement errors were also con- sidered, parameter estimates were more accurate.
Normality checking showed that the data of this study were not normally distributed. As the dataset of this study is reasonably large, Satorra–Bentler scaled chi-square (v2) was chosen for assessing goodnesss-of-fit of the SEM model. It is an adjusted v2
statistic which attempts to correct for the bias introduced when data are markedly non-normal in distribution (Satorra and Bentler, 2001). A non-significant, small v2 value indicates that the observed data are not significantly different from the hypothesized model. However, as formula of computing v2 is related to sample size, nearly all models are evaluated as incorrect as sample size in- creases. For this reason, the ratio of v2 to the degrees of freedom (v2/df) has been commonly used as an alternative fit index. If v2/ df is less than 2, the model is a good fit (Ullman, 2006). In addition to v2/df, root mean square error of approximation (RMSEA), Com- parative Fit Index (CFI) and Non-normed Fit Index (NNFI) were chosen to assess goodness-of-fit of the SEM model. As a rule, RMSEA value of less than .05 (lower value of 90% confidence inter- val (CI) of RMSEA no greater than .05 and upper value less than .08), CFI and NNFI of .95 or greater indicate good fit (Diamantopo- ulos and Siguaw, 2000).
To test for model stability, cross-validation was done by a split- sample approach (Diamantopoulos and Siguaw, 2000). Data were randomly split into a calibration sub-sample and a validation sub-sample. Tight cross-validation, which is the most rigorous way to examine the extent to which a model replicates in samples other than the one which it was derived, was conducted by fixing all the parameters across the calibration sub-sample and the vali- dation sub-sample (Diamantopoulos and Siguaw, 2000).
Descriptive statistics and the correlations of the included variables are shown in Table 1. As shown in Table 2, the good- ness-of-fit statistics indicated that the hypothesized model fits the calibration sub-sample and the validation sub-sample well. The model has good model stability. Results of tight cross- validation, that is, fixing all the parameters across the calibration
F2 F3 Near misses and injuries Safety participation
.52** (.48**) �.21** (�.22**) �.11 (.03)
.17* (.10) .08 (.03) �.10 (�.08)
.44** (.46**) .26** (.14) �.20** (�.20**) .18* (.24**)
ses are from the validation sub-sample.
Table 2 Goodness-of-fit of the structural equation model.
Goodness-of-fit measures Calibration sub-sample Validation sub-sample Tight cross-validation
v2 454.06 (p < .001) 562.72 (p < .001) 1531.98 (p < .001) v2/df 1.22 1.52 1.90 RMSEA .03 .05 .07 90% CI for RMSEA .02; .04 .04; .06 .06; .07 CFI .99 .97 .95 NNFI .99 .97 .95
14 C.K.H. Hon et al. / Safety Science 65 (2014) 10–19
and validation sub-samples, show reasonable fit. Full structural equation models of the calibration and validation sub-samples can be found in Appendices C and D.
Hypothesis (H1) is supported. The relationship between RMAA safety climate and self-re-
ported near misses and injuries was significantly negative. As shown in Fig. 2, the standardized path coefficient from RMAA safety climate to injuries was �.35 in the calibration sub-sample and �.21 in the validation sub-sample. That means, one unit of in- crease in the RMAA safety climate led to an approximately .2–.4 unit of decrease in the number of self-reported near misses and injuries. Hypothesis (H2) is supported. The relationship between RMAA safety climate and safety participation is significantly posi- tive. Referring to Fig. 2, the standardized path coefficient from RMAA safety climate to safety participation was .28 in the calibra- tion sub-sample and .18 in the validation sub-sample. That means, one unit of increase in the RMAA safety climate led to an approx- imately .2–.3 unit of increase in safety participation. Hypothesis (H3) is supported. The relationship between RMAA safety climate and safety compliance is significantly positive. With reference to Fig. 2, the standardized path coefficient from RMAA safety climate to safety compliance was .65 in the calibration sub-sample and .62 in the validation sub-sample. That means, one unit of increase in the RMAA safety climate will lead to an approximately .6–.7 unit of increase in safety compliance. When estimating the relation- ships between RMAA safety climate and safety performance, safety compliance has the highest standardized path coefficient (.65 in the calibration sub-sample; .62 in the validation sub-sample) when compared with near misses and injuries, and safety participation. The relationship between RMAA safety climate and safety partici- pation was the weakest, and was in fact even weaker than near misses and injuries.
It is acknowledged that the statistical relationship between safety climate and safety performance may be biased due to
R2 = .71 (.57)
R2 =.69 (.55)
R2 = .73 (.85)
RMAA s clima
F1 Management commitment
F2 Safety rules
F3 Safety responsibility
Fig. 2. Empirically tested structural equation model on the calibration sub-sample an employee involvement. F2 = Applicability of safety rules and work practices. F3 = Respon coefficients. All paths are significant at .01 level. Values in parentheses are from the presentation.
common method variance problem of single source data collection. Harmen’s one factor test, a statistical post hoc remedy to control for common method variance, was conducted. If there is substan- tial amount of common method variance, only a single factor or a general factor accounting for a large portion of variance will emerge from the factor analysis (Podsakoff and Organ, 1986). The factor analysis of the dependent and independent variables re- sulted in seven factors with the first factor accounting for 28.74% out of 62.78% of the total variance explained. Although the Har- men’s one factor test has limitations (Podsakoff and Organ, 1986), the result indicates that common method variance due to single source biases is not substantial in this study. The result is in line with Christian et al. (2009) that common methods bias is not a major concern in the safety domain.
8. Discussion and concluding remarks
Echoing with the meta-analysis of Beus et al. (2010), this study shows that the relationship between safety climate and safety per- formance is generally established in the context of RMAA works. Safety climate of RMAA works was positively related to safety par- ticipation and safety compliance but negatively related to self-re- ported near misses and injuries. However, as Beus et al. (2010) suggested, there are potential moderators to the relationship be- tween safety climate and safety performance. In view of the varia- tions of the empirical relationships between safety climate and safety performance in different industry contexts, industry context may be such a potential moderator; however, further investigation would be required to draw a more solid conclusion.
Christian et al. (2009) found that safety climate was inclined to have a stronger relationship with safety participation than with safety compliance. They suggested that complying with safety rules and regulation was the obligation of workers, so that safety climate had little influence on such compulsory behavior. In stark
R2 = .43 (.38)
R2 = .08 (.03)
R2 = .12 (.05)
Near misses & injuries
d (the validation sub-sample). Note. F1 = Management commitment to OHS and sibility for health and safety. Values on the arrows are the standardized (beta) path
validation sub-sample. Error terms and disturbances are omitted for clarity of
C.K.H. Hon et al. / Safety Science 65 (2014) 10–19 15
contrast with Christian et al. (2009), the findings of this study show that the safety climate of RMAA works only exert a small influence on the level of safety participation, but a much stronger influence on the level of safety compliance. This indicates that safety partic- ipation may be predominantly affected by variables other than the RMAA safety climate. One possible variable could be personal attitude towards safety. Besides, typical characteristics of RMAA
Fig. C1. Structural equation model
works, such as short duration and minute tasks, also restrain RMAA workers from safety participation.
Unlike safety compliance, which is considered the obligation of the employee, safety participation involves extra-role activities that are voluntary. More self-motivation is needed to perform safety par- ticipation than safety compliance. Workers are, by definition, ob- liged to comply with safety rules and regulation; however, it is not
of the calibration sub-sample.
16 C.K.H. Hon et al. / Safety Science 65 (2014) 10–19
always the case in the context of RMAA works. Very often, RMAA workers tend to rely on their experience rather than complying with safety rules and regulations, due to low safety awareness of RMAA workers (Hon et al., 2012). The prevailing safety climate level of RMAA works successfully motivates RMAA workers to comply with safety rules and regulations; however, the motivation is not suffi- ciently strong to encourage them to participate in extra safety activities.
Fig. D1. Structural equation model
As in the construction industry, Choudhry et al. (2009) found that the safety climate factor management commitment and employ- ee involvement had a stronger relationship with perceived safety performance than the other safety climate factor inappropriate safety procedures and work practices. The current study on RMAA works; however, has found that (F2) applicability of safety rules and practices has a stronger relationship with safety compliance, reflecting the peculiar situation of the RMAA sector. Appropriate
of the validation sub-sample.
C.K.H. Hon et al. / Safety Science 65 (2014) 10–19 17
safety rules and clear practices for RMAA works are currently lack- ing. RMAA works practitioners perceive that appropriate safety rules and clear practices for RMAA works will enhance their safety performance.
Although the construction industry generally has clear existing safety rules and best practice guidelines, the RMAA sector urgently needs a set of safety rules and practice guidelines that can better meet the specific needs of the RMAA works to follow. Despite the presence of some practice guidelines for implementation of prop- erty management companies, such efforts are yet to be compre- hensive. Moreover, many small/medium-sized RMAA contracting companies may not even be aware that these practice guidelines exist. Referring to Legislative Council (2011c), small RMAA con- tractors are generally less attentive to occupational safety and health legislation. Proper RMAA safety rules and safety practices should be laid down and promoted in the RMAA sector.
Safety climate factors, if managed properly, can result in better safety performance of RMAA works. Deficiencies in management procedures and safety system can be detected in the measure- ment of safety climate (Choudhry et al., 2009). Management com- mitment to OHS was perceived by the RMAA workers as an important factor of safety climate. It stems from genuine concern for the well-being of the employee. Such management commit- ment only occurs when top management truly believes that good safety performance is not a random occurrence but a calculated result of specific management actions (Hinze, 2006). Transparent and good communication with workers and supervisors is neces- sary. Safety should be integrated with other company goals. To enlist employee involvement, the RMAA workers need to have a clear understanding of their OHS responsibilities and the health and safety risks they will face and they should be assessed and praised for working safely.
The factor applicability of safety rules and work practices contrib- utes significantly to safety performance. It is important that safety rules and work practices should be up-to-date, technically correct, and clear (Choudhry et al., 2009), and should help the RMAA work- ers avoid potential risks and hazards, and conduct tasks safely. They also need to be upheld and properly enforced. To have a po- sitive perception of responsibility for health and safety, RMAA work- ers must have a correct assessment to risk and a locus of control for accidents (Hinze, 2006). In addition, the contracting companies also need to properly bear the responsibility for health and safety. Accident investigation should identify the root causes of accidents, not who should be blamed. Proactive safety measures should be performed on a daily basis, not be delayed until someone is injured.
This study exerts profound impact on construction safety of developed societies. It is expected that the RMAA sector will play an increasingly important role in the construction market of devel- oped societies because there are fewer buildings to be built but more to be repaired and maintained. In general, safety of new con- struction sites in developed societies has been improved; this is not the case for RMAA works. Although this study was conducted in Hong Kong, findings can be extrapolated to other developed cit- ies which have expanding RMAA sectors and increasing accidents of RMAA works.
To conclude, with the help of SEM, the intricate relationships of safety climate of RMAA works, and its safety climate factors and multifaceted safety performance have been simultaneously esti- mated. To the author’s knowledge, this is the first study of its kind in the RMAA sector of the construction industry to successfully test
the theoretical model of safety climate and safety performance using SEM techniques.
The work described in this paper was fully supported by a grant from the Research Grants Council of the Hong Kong Special Admin- istrative Region, China (RGC Project No. PolyU 5103/07E). This pa- per forms part of the research project titled ‘Safety Climate and its Impacts on Safety Performance of Repair, Maintenance, Minor Alteration and Addition (RMAA) Works’, from which other delive- rables have been produced with different objectives/scope but sharing common background and methodology. The authors also wish to acknowledge the contributions of other team members including Prof. Francis Wong, Dr. Daniel Chan, Dr. Don Dingsdag and Prof. Herbert Biggs.
Appendix A. 22 Statements measuring RMAA safety climate
Factor 1 Management commitment to OHS and employee involvement
The company really cares about the health and safety of the people who work here
There are good communications here between management and workers about health and safety issues
The company encourages suggestions on how to improve health and safety
I am clear about what my responsibilities are for health and safety
I think management here does enough to follow up recommendations from safety inspection and accident investigation reports
All the people who work in my team are fully committed to health and safety
There is good preparedness for emergency here Accidents which happened here are always reported Most of the job-specific safety trainings I received are
effective I fully understand the health and safety risks associated Safety inspection here is helpful to improve the health and
safety of workers Staff are praised for working safely
Factor 2 Applicability of safety rules and work practices Some jobs here are difficult to do safely Not all the health and safety rules or procedures are strictly
followed here Some of the workforces pay little attention to health and
safety Some health and safety rules or procedures are difficult to
follow Supervisors sometimes turn a blind eye to people who are not
observing the health and safety procedures Sometimes it is necessary to take risks to get the job done
Factor 3 Responsibility for health and safety People are just unlucky when they suffer from an accident Accident investigations are mainly used to identify who
should be blamed Work health and safety is not my concern Little is done to prevent accidents until someone gets injured
Appendix B. Scales of safety participation and safety compliance
Please answer this section by circling the most appropriate numbers.
1. Safety participation (0 = Never; 1 = Yearly; 2 = Monthly; 3 = Weekly; 4 = Daily)
(a) How frequent do you put in extra effort to improve safety of the workplace (e.g. reminding coworkers about safety procedures at work)?
0 1 2 3 4
(b) How frequent do you voluntarily carry out tasks or activities that help to improve workplace safety (e.g. attending safety meeting, receiving safety training)?
0 1 2 3 4
2. Safety compliance Please circle on a scale of 0–100% the percentage of time:
(a) You follow all of the safety procedures for the jobs that you perform. 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
(b) Your coworkers follow all of the safety procedures for the jobs that they perform. 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
18 C.K.H. Hon et al. / Safety Science 65 (2014) 10–19
See Fig. C1.
See Fig. D1.
Begg, D.J., Langley, J.D., Williams, S.M., 1999. Validity of self-reported crashes and injuries in a longitudinal study of young adults. Injury Prev. 5, 142–144.
Behm, M., 2005. Linking construction fatalities to the design for construction safety concept. Saf. Sci. 43, 589–611.
Beus, J.M., Payne, S.C., Bergman, M.E., Arthur Jr., W., 2010. Safety climate and injuries: an examination of theoretical and empirical relationships. J. Appl. Psychol. 95 (4), 713–727.
Burke, M.J., Sarpy, S.A., Tesluk, P.E., Smith-Crowe, K., 2002. General safety performance: a test of a grounded theoretical model. Pers. Psychol. 55, 429– 457.
Byrne, B.M., 2009. Structural Equation Modeling with AMOS: Basic Concepts, Applications and Programming, second ed. Taylor and Francis Group, Routledge.
Census and Statistics Department, 2007. Report on the Quarterly Survey of Construction Output (4th Quarter 2006), C&SD, The HKSAR Government. <http://www.censtatd.gov.hk/products_and_services/products/publications/ statistical_report/commerce_and_industry/index_cd_B1090002_dt_back.jsp> (retrieved 21.06.11).
Census and Statistics Department, 2008. Report on the Quarterly Survey of Construction Output (4th Quarter 2007), C&SD, The HKSAR Government. <http://www.censtatd.gov.hk/products_and_services/products/publications/ statistical_report/commerce_and_industry/index_cd_B1090002_dt_back.jsp> (retrieved 21.06.11).
Census and Statistics Department, 2009. Report on the Quarterly Survey of Construction Output (4th Quarter 2008), C&SD, The HKSAR Government. <http://www.censtatd.gov.hk/products_and_services/products/publications/ statistical_report/commerce_and_industry/index_cd_B1090002_dt_back.jsp> (retrieved 21.06.11).
Census and Statistics Department, 2010. Report on the Quarterly Survey of Construction Output (4th Quarter 2009), C&SD, The HKSAR Government. <http://www.censtatd.gov.hk/products_and_services/products/publications/ statistical_report/commerce_and_industry/index_cd_B1090002_dt_back.jsp> (retrieved 21.06.11).
Census and Statistics Department, 2011. Report on the Quarterly Survey of Construction Output (4th Quarter 2010), C&SD, The HKSAR Government. <http://www.censtatd.gov.hk/products_and_services/products/publications/ statistical_report/commerce_and_industry/index_cd_B1090002_dt_back.jsp> (retrieved 21.06.11).
Chan, A.P.C., Wong, F.K.W., Yam, M.C.H., Chan, D.W.M., Ng, J.W.S., Tam, C.M., 2005. From Attitude to Culture – Effect of Safety Climate on Construction Safety. Research Monograph, Construction Safety Research Group, Research Centre for Construction and Real Estate Economics, Department of Building and Real Estate, The Hong Kong Polytechnic University.
Chan, A.P.C., Wong, F.K.W., Yam, M.C.H., Chan, D.W.M., Hon, C.K.H., Wang, Y., Dingsdag, D., Biggs, H., 2010. RMAA safety performance – how does it compare with Greenfield projects? In: Proceedings of the CIB World Congress 2010,
Building a Better World, The Lowry, Salford Quays, United Kingdom, 10–13 May 2010, Paper ID:457, p. 154.
Choudhry, R.M., Fang, D., Lingard, H., 2009. Measuring safety climate of a construction company. J. Construct. Eng. Manage. 135 (9), 890–899.
Christian, M.S., Bradley, J.C., Wallace, J.C., Burke, M.J., 2009. Workplace safety: a meta-analysis of the roles of person and situation factors. J. Appl. Psychol. 94 (5), 1103–1127.
Clarke, S., 2006. The relationship between safety climate and safety performance: a meta-analytic review. J. Occup. Health Psychol. 11 (4), 315–327.
Cooper, M.D., Phillips, R.A., 2004. Exploratory analysis of the safety climate and safety behavior relationship. J. Saf. Res. 35 (5), 497–512.
Dedobbeleer, N., Béland, F., 1991. A safety climate measure for construction sites. J. Saf. Res. 22 (2), 97–103.
Diamantopoulos, A., Siguaw, J.A., 2000. Introducing LISREL: A Guide for the Uninitiated. Sage Publications Ltd., Thousand Oaks, California.
Fang, D., Chen, Y., Wong, L., 2006. Safety climate in construction industry: a case study in Hong Kong. J. Construct. Eng. Manage. 132 (6), 573–584.
Flin, R., Mearns, K., O’Connor, P., Bryden, R., 2000. Measuring safety climate: identifying the common features. Saf. Sci. 34 (1–3), 177–192.
Gabbe, B.J., Finch, C.F., Bennel, K.L., Wajswelner, H., 2003. How valid is a self reported 12 month sports injury history? Br. J. Sports Med. 37 (6), 545– 547.
Gillen, M., Baltz, D., Gassel, M., Kirsch, L., Vaccaro, D., 2002. Perceived safety climate, job demands, and coworker support among union and non-union injured construction workers. J. Saf. Res. 33, 33–51.
Glendon, A.I., Litherland, D.K., 2001. Safety climate factors, group differences and safety behavior in road construction. Saf. Sci. 39 (3), 157–188.
Griffin, M.A., Neal, A., 2000. Perceptions of safety at work: a framework for linking safety climate to safety performance, knowledge, and motivation. J. Occup. Health Psychol. 5 (3), 347–358.
Hinze, J., 2006. Construction Safety, second ed. Alta Systems Inc, Gainesville, Florida.
Hofmann, D.A., Morgeson, F.P., 1999. Safety-related behavior as a social exchange: the role of perceived organizational support and leader-member exchange. J. Appl. Psychol. 84 (2), 286–296.
Hon, C.K.H., Chan, A.P.C., Wong, F.K.W., 2010. An empirical study on causes of accidents of repair, maintenance, minor alteration and addition works in Hong Kong. Saf. Sci. 48 (7), 894–901.
Hon, C.K.H., Chan, A.P.C., Yam, M.C.H., 2012. Empirical study to investigate the difficulties of implementing safety practices in the repair and maintenance sector of Hong Kong. J. Construct. Eng. Manage. 138 (7), 877–884.
Hon, C.K.H., Chan, A.P.C., Yam, M.C.H., 2013. Determining safety climate factors in the repair, maintenance, minor alteration, and addition sector of Hong Kong. J. Construct. Eng. Manage. 139 (5), 519–528.
Health and Safety Executive (HSE), 2012. Safety Culture on the Olympic Park. Research Report RR942. <http://www.hse.gov.uk/research/rrpdf/rr942.pdf> (retrieved 02.11.13).
Huang, Y.H., Ho, M., Smith, G.S., Chen, P.Y., 2006. Safety climate and self-reported injury: assessing the mediating role of employee safety control. Accid. Anal. Prev. 38 (3), 425–443.
James, L.R., 1982. Aggregation bias in estimates of perceptual agreement. J. Appl. Psychol. 67 (2), 219–229.
Jöreskog, K.G., Sörbom, D., 2006. LISREL 8.80 for Windows [Computer Software]. Scientific Software International Inc., Lincolnwood, IL.
Lamm, F., 1997. Small businesses and OH&S advisors. Saf. Sci. 25 (1–3), 153–161. Legislative Council, 2011a. Legislative Council Panel on Manpower – Hong Kong’s
Occupational Safety Performance in 2008 [LC Paper No. CB(2)2176/08-09(01)]. <http://www.legco.gov.hk/yr08-09/english/panels/mp/papers/mp0716cb2- 2176-1-e.pdf> (retrieved 07.07.11).
Legislative Council, 2011b. Legislative Council Panel on Manpower – Hong Kong’s Occupational Safety Performance in 2010 [LC Paper No. CB(2)2044/10-11(11)].
C.K.H. Hon et al. / Safety Science 65 (2014) 10–19 19
<http://www.legco.gov.hk/yr10-11/english/panels/mp/papers/mp0617cb2- 2044-11-e.pdf> (retrieved 07.07.11).
Legislative Council, 2011c. Occupational Safety Performance in Hong Kong. Background Brief Prepared by the Legislative Council Secretariat for the Meeting of Panel on Manpower on 17 June 2011. <http://www.legco.gov.hk/ yr10-11/english/panels/mp/papers/mp0617cb2-2044-12-e.pdf> (retrieved 08.05.13).
Lingard, H., Rowlinson, S., 2005. Occupational Health and Safety in Construction Project Management. Spon Press, London and New York.
Lingard, H., Cooke, T., Blismas, N., 2011. Coworkers’ response to occupational health and safety – an overlooked dimension of group-level safety climate in the construction industry? Eng. Construct. Archit. Manage. 18 (2), 159–175.
Mearns, K., Whitaker, S.M., Flin, R., 2001. Benchmarking safety climate in hazardous environments: a longitudinal, interorganizational approach. Risk Anal. 21 (4), 771–786.
Mearns, K., Whitaker, S.M., Flin, R., 2003. Safety climate, safety management practice and safety performance in offshore environments. Saf. Sci. 41 (8), 641– 680.
Mohamed, S., 2002. Safety climate in construction site environments. J. Construct. Eng. Manage. 128 (5), 375–384.
Neal, A., Griffin, M.A., 2004. Safety climate and safety at work. In: Barling, J., Frone, M.R. (Eds.), The Psychology of Workplace Safety. American Psychological Association, Washington, DC, pp. 15–34.
Neal, A., Griffin, M.A., 2006. A study of the lagged relationships among safety climate, safety motivation, safety behavior, and accidents at the individual and group levels. J. Appl. Psychol. 91 (4), 946–953.
Occupational Safety and Health Council (OSHC), 2008. Construction Industry Safety Climate Index Software. OSHC, Hong Kong.
Podsakoff, P.M., Organ, D.W., 1986. Self-reports in organizational research: problems and prospects. J. Manage. 12 (4), 531–544.
Pousette, A., Larsson, S., Törner, M., 2008. Safety climate cross-validation, strength and prediction of safety behavior. Saf. Sci. 46 (3), 398–404.
Reason, J., 1995. A systems approach to organizational error. Ergonomics 38 (8), 1708–1721.
Reason, J., 1997. Managing the Risk of Organizational Accidents. Ashgate Publishing Limited, Aldershot, England.
Satorra, A., Bentler, P.M., 2001. A scaled difference chi-square test statistic for moment structure analysis. Psychometrika 66, 507–514.
Siu, O.L., Phillips, D.R., Leung, T.W., 2004. Safety climate and safety performance among construction workers in Hong Kong: the role of psychological strains as mediators. Accid. Anal. Prev. 36, 359–366.
Ullman, J.B., 2006. Structural equation modeling: reviewing the basics and moving forward. J. Pers. Assess. 87 (1), 35–50.
Vroom, V.H., 1964. Work and Motivation. John Wiley, New York. Yiu, C.Y., 2007. Building depreciation and sustainable development. J. Build.
Appraisal 3 (2), 97–103. Zhou, Q., Fang, D., Mohamed, S., 2011. Safety climate improvement: case study in a
Chinese Construction Company. J. Construct. Eng. Manage. 137 (1), 86–95. Zohar, D., 1980. Safety climate in industrial organizations: theoretical and applied
implications. J. Appl. Psychol. 65 (1), 96–102. Zohar, D., 2002. Modifying supervisory practices to improve subunit safety: a
leadership-based intervention model. J. Appl. Psychol. 87 (1), 156–163. Zohar, D., 2003. Safety climate: conceptual and measurement issues. In: Quick, J.C.,
Tetrick, L.E. (Eds.), Handbook of Occupational Health Psychology. American Psychological Association, Washington, DC, pp. 123–142.
Zohar, D., 2010. Thirty years of safety climate research: reflections and future directions. Accid. Anal. Prev. 42 (5), 1517–1522.
- Relationships between safety climate and safety performance of building repair, maintenance, minor alteration, and addition (RMAA) works
- 1 Introduction
- 2 Safety climate
- 3 Safety performance
- 4 Relationships between safety climate and safety performance
- 4.1 Theoretical linkages
- 4.2 Empirical relationships
- 5 Research hypotheses
- 6 Research methods
- 6.1 Questionnaire design
- 6.1.1 RMAA safety climate
- 6.1.2 Self-reported near misses and injuries
- 6.1.3 Safety participation
- 6.1.4 Safety compliance
- 6.2 Participants and procedures
- 6.3 Data analysis
- 6.1 Questionnaire design
- 7 Results
- 8 Discussion and concluding remarks
- Appendix A 22 Statements measuring RMAA safety climate
- Appendix C
- Appendix D
- Appendix B Scales of safety participation and safety compliance