Implications of turnover and trust for safety attitudes and behaviour in work teams

Abstract

Turnover potentially leads to a new individual being selected into a work team. This study investigated the safety-specific trust which team members place in their organisation’s selection and induction processes, and related this to the perceived risk from new employees. The research was conducted with teams working in forest harvesting, an occupation which has high-turnover, high risk and a high accident rate. Results indicate that trust in induction processes was negatively correlated with perceived risk from a new employee. Team members also engaged in a number of safety ensuring behaviours when a new individual joined the team, and these were related to the level of perceived risk, and how much they cared about their team members’ safety. It is argued that trust in the safety-specific characteristics of an organisation’s selection and induction process may have negative consequences for safety.

Introduction

A number of studies have reported a relationship between accidents and employee turnover (e.g., Bell and Grushecky, 2006, Kincaid, 1996). Furthermore, team instability from absenteeism has been associated with occupational accidents (Goodman and Garber, 1988). A possible interpretation of these relationships is that new employees (those replacing turned-over staff or filling in for absentees) pose a safety risk. Team members’ perceptions of this risk could be decreased if team members trust that their organisation will deal with turnover and absenteeism by selecting a new worker that will work safely. The safety-specific trust issues associated with organisational selection and induction, and team members’ perceptions of, and safety behaviour towards, new team members are the focus of this paper.

The research was conducted in the forestry industry where individuals work in small teams harvesting trees. Turnover is often very high in the forestry industry (Kirk et al., 1997), and forestry work is internationally recognised as risky (Chapman and Husberg, 2008, Driscoll et al., 1995, Lilley et al., 2002, Östberg, 1980, Slappendel et al., 1993). In a review of factors affecting forestry worker safety, Slappendel et al. (1993) cited a number of studies that have reported that the inexperienced may be more at risk of an injury. We were unable to find any studies which have proposed a theoretical explanation for the relationship between turnover and accidents. However, the theoretical relationship between absenteeism and accidents proposed by Goodman and Garber (1988) in their study of coal miners appears to be applicable to the turnover situation. They argued that new workers are likely to have a lack of familiarity with the unique characteristic of particular machinery, specific work environments, work methods and people which could increase accidents, particularly if existing workers do not make compensatory changes. Bentley et al. (2005) have also argued that changes in team personnel can put team safety at risk because the new team member may not be familiar with the team’s practices, procedures or equipment. Furthermore, Bentley et al. (2002) reported that 44% of injuries on logging skid sites occur within the workers’ first year on the job, with 32% within the first 6 months of employment.

The ‘compensatory change’ component of Goodman and Garber’s (1988) premise appears to have received little research attention. We argue that an issue associated with the level of ‘compensatory change’ is the degree of trust that team members place in their organisation’s selection and induction processes. The selection process can include measures that attempt to predict applicants’ safety potential. Furthermore, once an individual has been offered the job, pre-start induction processes should introduce the new employee to the organisation’s safety processes and policy. How much team members trust these processes to have a positive ‘safety ensuring’ outcome, may determine how they respond when a new employee joins the team.

More specifically, and in line with recent arguments put forward by Conchie and Donald (2008), we argue that in the case of turnover (and indeed absenteeism) a degree of distrust is advantageous for safety. McEvily et al. (2003) have also argued that trust can reduce individuals’ inclination to monitor and safeguard. While there is a developing literature outlining the safety benefits of trust (e.g., Jeffcott et al., 2006, Reason, 1997, Watson et al., 2005, Zacharatoes et al., 2005), very little has been written on the value of distrust (but see Burns et al., 2006). Furthermore, there is some debate as to whether trust and distrust are opposite ends of the same continuum (e.g., Hosmer, 1995, Meyer et al., 1995, Rotter, 1971), or are separate constructs (e.g., Deutsch, 1960, Lewicki et al., 1998, Sitkin and Roth, 1993). Because our research is not addressing this debate, we will simply argue that trust in the context of employee turnover is negative for safety. In relation to turnover, employees should show a degree of caution in working with a new team member, and not trust their organisation’s ability (via selection and pre-start induction processes) to ensure the new employee will work safely, has the attitudes to act safely in general, and has the relevant job context knowledge (familiarity).

While Reason (1997) argues that trust at the organisational level is likely to promote open communication about safety, complete trust in the organisation’s ability to manage safety via selection and induction process could have negative consequences. Similarly, trust at the interpersonal (team) level has been linked with positive safety outcomes (e.g., Donald and Young, 1996, Hofmann and Stetzer, 1998, Watson et al., 2005), but we would argue that it is essential that this trust be earned by new team members rather than given as a consequence of a set of assumptions about selection and induction processes which may be incorrect. Safety may well be enhanced by adopting what Hale (2000) referred to as “creative mistrust”: being positively wary about safety systems and safety management.

Why might team members trust an organisation’s safety-specific selection processes? Several researchers have found significant correlations between personality and safety behaviours (e.g., Hansen, 1989, Tiffin and McCormick, 1962), while others have found that companies with low accident rates had more elaborate selection systems (e.g., Cohen, 1977, Smith et al., 1978). However, while many researchers would agree that it is hard to accurately predict safety attitudes and behaviour, this may not be the view held by members of a work team. Team members may recall being asked questions about safety when they were interviewed for their job, and not realise that interview data may show very little relationship with post employment behaviour. Similarly, workers may hold incorrect assumptions about the effectiveness of pre-start induction processes. While training has been shown to improve safety attitudes (e.g., DeJoy et al., 2000, Harvey et al., 2001) and decrease lost-time accidents (e.g., Harshbarger and Rose, 1991), inadequate or inappropriate safety training has been given as a reason for accidents in the forestry industry (Crowe, 1985, Holman et al., 1987, MacFarlane, 1979). Furthermore, Bell and Grushecky (2006) have noted that safety training programs in the forestry industry are rarely evaluated for effectiveness. If team members trust their organisation’s selection processes and/or their safety induction/pre-start training processes to have a positive impact, and this trust is misplaced, they may face risks from the behaviour of the new employee which they are not anticipating. Thus Hypothesis 1 is that trust in the safety-specific aspects of selection and induction will be negatively correlated with perceived risk from new employees: as trust increases perceived risk decreases.

Trust in selection and trust in induction effectively place the responsibility for safety with the organisation. Team members that do not trust these processes may attempt to mitigate their own (and their co-workers) risk by engaging in safety ensuring behaviours when a new member joins the team. Safety ensuring behaviours could include informally attempting to assess the employee’s attitudes to safety, watching out for their safety, offering them assistance and information, familiarising them with the specifics of the team’s operation, and being wary of their actions. Thus Hypothesis 2 predicted that safety ensuring behaviours would be positively associated with perceived risk from new team members: as the perceived risk increased safety ensuring behaviours would increase. Another factor which may influence an employees’ reaction to a new team member, is the level of concern the individual holds for their existing team members’ safety. Thus Hypothesis 3 predicted that employees that have a high level of concern for their existing co-workers safety are likely to be particularly careful with a new team member and engage in more safety ensuring behaviours when a new member joins the team.