Measures

All items were taken from existing measures to maximize the content validity of the questionnaire. Prior to adopting a multipronged approach to test the hypotheses, the questionnaires were translated from the English survey items via translations procedures (Brislin, 1970). To validate the survey translations, two HR managers proofread the Chinese surveys for readability and ease of comprehension. Any concerns about or discrepancies between the English and Chinese versions were addressed to ensure they were the same in terms of meaning. Each scale is rated from 1, indicating "strongly disagree", to 5, indicating

"strongly agree". All the items can be found in Appendix A.

HPWS. The study adopted a 35-item scale to measure bundled high-performance human resource practices (Chuang & Liao, 2010). The high-performance HR practices scale included eight dimensions: 'staffing', 'training', 'involvement and participation', 'performance

34

appraisals', 'compensation/rewards', and 'caring' (Chuang & Liao, 2010). Example items are

"Employees are often asked to participate in work-related decisions" and "Employee salaries and rewards are determined by their performance." All responses from the employees for a unit were then aggregated to create the unit-level variables of HPWS. The reliability of this scale was .94.

Individual employee perceptions of HPWS indeed exemplifies as a shared group property, thus, the appropriateness of within-group agreement and of between-group variability need to be justified. Empirically speaking, aggregating data to the mean level in the group can be justified by assessing the R_wg and interclass correlations. HPWS were with Rwg mean value of .98, ranging from .88 to 1.00. The ICC(1) estimate was .19 and the ICC(2) estimate was .56. Though ICC(2) value was lower than ideal owing to smaller group sizes (Klein & Kozlowski, 2000), the ICC(1) value exceeded the levels of reliability and agreement.

Overall, aggregating HPWS responses to the group level met statistical justification.

Job crafting. Job crafting was measured based on a scale developed by Petrou et al.

(2012) and included 13 items relating to general-level crafting from among the three dimensions of 'seeking resources', 'seeking challenges', and 'reducing demands'. Example items are "I ask my supervisor for advice.", "I ask for more responsibilities.", and "I try to simplify the complexity of my tasks at work." The Cronbach's alpha for this scale was .83.

Work engagement. Work engagement was assessed using nine items based on Bakker (2014). These items measure aspects of individuals' engagement. Example items are "At my work, I feel bursting with energy." and "I feel happy when I am working intensely." The Cronbach's alpha for this scale was .91.

Burnout. Burnout was assessed using nine items based on Bakker (2014). Example items are "There are days when I feel tired before I arrive at work." and "During my work, I often feel emotionally drained." The Cronbach's alpha for this scale was .85.

35

Affective commitment. Affective commitment was measured by five items from Meyer and Allen's Affective Commitment Scale (Meyer & Allen, 1997). Example items are "I feel a strong sense of belonging to my organization.", and "I would be happy to work at my organization until I retire." The Cronbach's alpha for this scale was .80.

Job satisfaction. Job satisfaction was measured by three items from Hackman and Oldham (1976). An example item is "All in all, I am satisfied with my job here at this organization." The Cronbach's alpha for this scale was .95.

Person-job fit. Person-job fit was measured by three items from Saks and Ashforth (1997). An example item is "To what extent does the job fulfill your needs?" The Cronbach's alpha for this scale was .92.

Intention to quit. Intention to quit was measured with the 3-item scale from Camman, Fichman, Jenkins, and Klesh (1979). An example item is "I often think of leaving the organization." The Cronbach's alpha for this scale was .80.

Work-family conflict. Work-family conflict was measured with the 4-item scale from Gutek, Searle, and Klepa (1991). An example item is "After work, I come home too tired to do some of the things I'd like to do." The Cronbach's alpha for this scale was .80.

Self-handicapping. Self-handicapping was measured with three items from Bakker (2014). An example item is "I tend to put things off until the last moment." The Cronbach's alpha for this scale was .84.

Control variables. Following other researchers (e.g., Jensen et al., 2013), age, gender, and education were included as control variabless. Gender is usually measured as a dichotomous variable and is coded such that 1 is female and 2 is male. In addition, this study controlled goal orientation because previous research has indicated that goal orientation can explain individuals' well-being, fatigue, job satisfaction (e.g., Van Yperen & Janssen, 2002), and job performance (e.g., Porath & Bateman, 2006). Goal orientations also reveal

36

work-related reinforcement associated with individuals' well-being and job outcomes in the work environment (e.g. Vansteenkiste, Duriez, Simons, & Soenens, 2006; Vansteenkiste, Neyrinck, Niemiec, Soenens, De Witte, & Van den Broeck, 2007). Following this, this study suggests that HPWS-espoused goals for employees may be interfered by employees' own goal orientations; thus, the inclusion of goal orientation in the analyses may decrease the likelihood that any observed relationships between HPWS and employee work outcomes are confounded by individual goal orientations. Goal orientations are composed of the following dimensions: learning (five items), performance (four items), and avoiding (four items) (VandeWalle, 1997). The sample items are "I am willing to select a challenging work assignment that I can learn a lot from.", "I try to figure out what it takes to prove my ability to others at work.", and "I prefer to avoid situations at work where I might perform poorly."

The Cronbach's alpha for this scale was .78. Recent research suggests that transformational leadership acts as a valuable leadership model in enhancing occupational safety and is worth being integrated within HPWS implementations (e.g., Zacharatos et al., 2005). In addition, transformational leadership nurtures employees with feelings of trust and commitment, thus enabling the implementation of HR practices (Ang et al., 2013). Therefore, this study also included transformational leadership as the control variable, which was measured with four items from McCollKennedy and Anderson (2002). A sample item is "Gives personal attention to each team member." The Cronbach's alpha for this scale was .72.