Human resource safety practices and employee injuries.

Workplace safety continues to be a high operational priority facing many organizations across all types of industries. And, as Ruth (2004) notes, injuries oftentimes result from managerial issues, rather than more notable safety issues. Thus, identifying potential steps employers can take in their managerial practices is critical to managing organizational costs, improving effectiveness of public policy and, most important, protecting employees. A recent trend in human resources (HR) research may provide some clarity to the safety research. A number of studies have established the effectiveness of HR practices (Huselid, 1995; McEvoy and Cascio, 1985), establishing an overall consensus that certain "good" HR practices lead to positive organizational outcomes (Delaney and Huselid, 1996; Becker and Gerhart, 1996). Organizational involvement has also been linked to improved safety outcomes (Oliver et al., 2002), but these have not been broken down into specific practices. Therefore, the purpose of this study is to explore the connection between organizational-level HR practices and employee injuries.

Background/Proposed Associations

Since this study is taking an exploratory look at how HR practices will impact employee injuries at the organizational-level, the study focuses on the more easily measurable aspects of the four most basic HR practices-selection, training, employee evaluations, and compensation. These are the practices that have been studied most frequently and have consistently been tied by research to other positive organizational outcomes.

Selection. Past research has indicated organizations may improve their safety outcomes through two primary selection processes: (1) identifying and eliminating individuals unsuited to certain types of work and (2) by hiring for positions that require a very high degree of skill where the cost of accidents as well as the ratio of applicants to positions open is high. Beyond these processes, selection has not been found to be as useful as once hoped in the safety literature (Hale and Hale, 1972). However, empirical evidence continues to indicate that well-designed selection procedures improve overall organizational performance (Terpstra and Rozell, 1993). Thus, selecting applicants for safety (e.g., hiring employees with increased knowledge of safety, based on their past safety performance; asking questions specifically regarding safety in the interview) should have a positive association with organizational performance tied to safety, such as wearing safety equipment and following safety behaviors. These behaviors, in turn, should reduce the number of employee injuries incurred at the organization. Socialization may also occur in the selection process (Anderson and Ostroff, 1997) by emphasizing the organization's safety values to new employees, in turn reducing employee injuries.

Hypothesis 1a: Selecting for safety has a positive association with reduced injuries.

Since selecting specifically for safety (e.g., asking direct questions and discussing safety in the interview process) is not highly practiced in organizations, an exploratory look at how other selection practices (e.g., pre-employment testing, screening for past work experience) may be associated with safety is also undertaken. Although numerous pre-employment tests/selection criteria exist, only criteria that tied with past safety research were explored in this study.

Prior work experience has been one of the longest used screening tools in the selection process. When screening for work experience, organizations typically seek experience in the industry as well as for a specific type of work. In Hansen's (1989) causal model of accidents, he found job experience was one of only two variables that were significant parameters of accident risk. Experience provides employees with knowledge of both general industrial hazards, as well as familiarity with individual machines and components (Hale and Hale, 1972), providing an expectation that increased work experience should be associated with reduced employee injuries.

Several researchers have reported significant correlations between certain aspects of employee personality and safety behaviors (Tiffin and McCormick, 1962; Hansen, 1988). Characteristics such as extroversion (Powell et al., 1971), general social maladjustment, neuroticism, impulsivity, and aggression (Hansen, 1988) have been linked specifically to industrial accidents. Intelligence, as measured by a general aptitude battery or an IQ test, has also been linked to safety outcomes (Tiffin and McCormick, 1962). Individuals involved in increased injuries were found to underestimate risk, overestimate control, and lack understanding of how injury is caused (factors potentially related to intelligence), leading to fewer precautions, less timely action, and increased injuries (Gottfredson, 2004).

Past research has established that alcohol is negatively related to many aspects of employee safety performance, including vision, psychomotor functions, judgment, reasoning and memory (Hale and Hale, 1972). Yet, the research on drug usage and workplace safety has yielded mixed results. Spicer, Miller, and Smith (2003) found the odds of injury among workers increased when an indicator of problem substance use was identified. Whereas Kaestner and Grossman (1998) found a negative link among males, but not among females, and Macdonald (1995) found that a connection with illicit drug use and in juries only held for males and the youngest age group of workers. Normand, Salyards, and Mahoney's (1990) study, looking specifically at the usefulness of pre-employment drug-testing (rather than ability tests while under the influence), did not reveal evidence of a negative link with injuries. Although past findings are mixed, given the pattern of results with ability tests in alcohol-related research, a link between usage of both alcohol and drug-testing and decreased injuries is expected.

Additionally, pre-employment physicals may prevent employees from performing duties beyond their physical capabilities, or identify areas where employees need additional strength/ conditioning training. Spence (1998) indicates that an individual assessment of abilities is important as each patient presents a different pattern of behavioral strengths and weaknesses, requiring a tailored approach. Further, some jobs are noted as needing minimum physical standards (Sothmann et al., 2005) to ensure job performance and potentially reduce injuries.

Hypothesis 1b: Using selection practices that screen for prior work experience, personality, intelligence, alcohol, drugs, and physical ability has a positive association with reduced injuries.

Training. Training is one of the HR practices most commonly discussed in the literature as a way to improve employee safety. This study examines training on both safe work procedures and the importance of safety. Training impacts employees by improving their skills and abilities, as well as by communicating what is important. Training is an "essential component" because organizations rely on frontline employee skill and initiative to identify and resolve problems and to initiate changes in work methods (Pfeffer and Veiga, 1999: 43). These aspects are important to safety, because organizations need employees to help identify and resolve potential safety issues. Training employees on attitudes and beliefs toward safety has been shown effective (DeJoy et al., 2000; Harvey et al., 2001). Training interventions on the supervisor and employee level have been associated with reduced lost-time injuries and injury costs (Harshbarger and Rose, 1991), a decrease in minor-injury rates, an increase in protective equipment usage, and improved safety climate ratings (Zohar, 2002). Training also provides communication. As injuries occur and organizations learn what behavior should be changed, these changes need to be communicated to all employees (Dougherty, 1997).

Hypothesis 2: Safety training has a positive association with reduced injuries.

Performance Evaluations. Performance evaluations may also establish organizational values. Performance evaluation involves identifying, measuring, revising, and developing human performance in organizations (Carroll and Schneier, 1982). This study examines whether or not safety is included as a part of overall employee evaluations, as well as whether feedback is provided to employees regarding their safety performance. By including safety on employee evaluations, management communicates the importance of safety to employees. McAfee and Winn's (1989) review found consistency across research regarding the importance of feedback on safety to employees. Research has also demonstrated that factors emphasized by performance evaluation influences the safety behaviors employees follow. In PateCornell's (1990) study of oil exploration companies, organizations with a high rate of injuries emphasized speed over safety on performance evaluation. Cooper (2001) suggests what gets measured and rewarded is what gets done well. Ingalls (1999) supports this viewpoint, indicating companies should identify measures reflecting organizational initiative and promoting well-being and safe performance.

Hypothesis 3: Evaluation of safety has a positive association with reduced injuries.