Managers’ attitudes towards safety and accident prevention

Abstract

The aims of the present paper are to determine attitudes towards safety and accident prevention among presidents, vice-presidents and managers in the industrial company Norsk Hydro and to analyse the associations between attitudes, behavioural intentions and behaviour. The sample consisted of 210 respondents and the data was collected in 1997 and 1998 among participants at the Hydro Management Safety Training Workshops, which is a safety course for the managers employed by the company. Managers’ attitudes are interesting because they may affect behavioural intentions and the managers’ behaviour related to the achievement of safe working practices. Eight attitudinal dimensions explained up to nearly 40% of the variance in behaviour. The study shows that safety attitudes may be an important causal factor for managers’ behavioural intentions as well as behaviour. High management commitment, low fatalism, high safety priority, and high risk awareness seemed to be particularly important attitudes for managers.

Introduction

There are a great number of studies showing that attitudes and behaviour are significantly associated. Studies based on the theory of reasoned action and planned behaviour (Ajzen and Fishbein, 1980, Ajzen, 1988) show that there is a significant association between health attitudes and risk behaviour. The theory of reasoned action was developed by the mean of explaining health-related beliefs, attitudes and behaviour. Health-related beliefs are similar to safety beliefs since they might affect ones health. A major assumption in this theory is that people behave sensible, they deliberately employ information from their surroundings and consider the implications of their actions. Two central variables predict peoples intentions to engage in healthy or unhealthy behaviours. These are attitudes towards behaviour and subjective norms, e.g. people's beliefs about social expectations that one will behave in a specific way. These two components combine to determine people's subjective estimates of how likely it is that they will engage in a specified behaviour (the behavioural intention). The theory of planned behaviour is an extended model that in addition to attitudes and subjective norms also includes the individual's control over his or her actions. Behavioural control relates to how easy or difficult it is to perform a behaviour (Ajzen, 1988). In these models attitudes are conceived to be “causal” factors influencing behavioural intentions as well as behaviour.

Attitudes may affect top and middle management decisions, which also exert influence on the conditions under which employee individual decision making take place. They may affect the priorities as well as company policy about safety. In addition, they may directly as well as indirectly affect employee attitudes and behaviour. The main aims of the present paper are to determine attitudes towards safety and accident prevention among managers and to analyse the associations between attitudes, behavioural intentions and behaviour.

The present study is restricted to analysing the relations between managers’ safety attitudes, behavioural intentions and their self-reported behaviour. It is hypothesised that attitudes affects behavioural intentions and that behavioural intentions have affects behaviour. However, it may also be that behaviour and behavioural intentions affect attitudes. LISREL-modelling of the relation between safety attitudes and behaviour among a sample of seventy-four managers showed that a model where attitude was the “causal” factor fitted the data, while a model where attitude was the dependent variable had a bad fit (Rundmo, 1998a). Attitude was found to be one of the strongest predictor of shop floor workers’ risk behaviour (Rundmo, 1992, Rundmo, 1998b). The great majority of studies on safety and risk issues focus lay people and employee judgements. According to Holmes et al. (1997) there is a lack of studies measuring managers’ subjective evaluations. Consequently, the present study focuses on managers’ safety attitudes as well as subjective evaluations of the working conditions.

A very pragmatic point to start for answering the question of which attitudes are ideal could be to discuss what behaviour in general is ideal for managers to show in controlling hazards. The managers have (1) to detect hazards, (2) find ways to control them, (3) priorities them, (4) chose good solutions, (5) implement them and then (6) monitor and learn. This means that they have to have or mobilise the knowledge to carry out each of the steps and they have to give the tasks resources (time, money, competence and equipment). They also have to give the tasks priority and not try to shuffle off the responsibility onto others. In addition, they have to know with whom to collaborate and also be prepared to do so. Then they have to apply effective management control methods to ensure that the tasks are carried out to schedule and have been successful, i.e. a management and control function (Hale et al., 1997, Hale et al., 1998, Hale, 1998). Based on these principles it is also possible to define what is ideal as well as non-ideal behaviour (see also Hale and Glendon, 1987). Thereafter, we should be able to relate the behaviour to attitudes.

An ideal safety attitude is an attitude contributing to enhancing the safety, i.e. safe behaviour and lowering the frequency of accidents and near-accidents. A non-ideal attitude is one that contributes to the opposite. Safe behaviour is behaviour that leads to a reduced frequency of accidents. However, it may not be easy to decide on what really is an ideal attitude contributing to safety. In safety research there are very few studies related to how different type of behaviours amongst managers and other management routines predicts accidents and near misses. “Ideal” safety attitudes among managers are those attitudes that affect employee behaviour and the number of accidents. Studies comparing high and low accident plants have shown management commitment and involvement in accident prevention and safety promotion to be important for the safety status as well as risk behaviour and the number of accidents and near misses (Cohen, 1975, Cohen et al., 1975, Simonds and Shafai-Sahrai, 1977, Smith et al. 1978; see also Hale et al., 1997). Consequently, managers should be committed to and involved in safety promotion. An ideal attitude is that they agree on indicators showing that they really are involved and committed.

Managers’ fatalism concerning the need for and the usefulness of accident prevention relate negatively to safety commitment and involvement. Such an attitude has been shown to be one of the most significant direct predictors of employee risk behaviour (Rundmo, 1996). Risk behaviour is the extent to which personnel ignore safety regulations in order to get a job done, carry out activities which are forbidden, perform their job duties correctly etc. It has to be distinguished from risk-taking behaviour (Zuckerman, 1979). Risk-taking behaviour refers to whether people like to take risks and are sensation seekers, i.e. a personality variable.

Studies have also shown that an attitude amongst employees accepting rule violation is the strongest predictor of risk behaviour (Rundmo, 1998a, Rundmo, 1998b). Managers’ attitude and behaviour may influence employee attitudes as well as their behaviour. Several companies give management safety talks and risk communication priority, i.e. during a walk-observe-talk safety inspection. To talk about safety focuses safe behaviour and so does communication about hazards and risks. To agree on statements admitting it is difficult and embarrassing to talk to the employees about safety shows a non-ideal attitude.

Rundmo, 1996, Rundmo, 1998b found that there was a significant positive association between risk perception and occupational risk behaviour. The more risk employees perceived the more risky was also their behaviour. However, the association was completely insignificant when we entered other predictor variables related to organisational and social factors relevant to safety into a model aimed at predicting risk behaviour (commitment and involvement in safety work, attitudes towards safety and accident prevention, satisfaction with the status of safety and contingency measures, job stress and train). One reason for studying risk perception is the assumption that perception affects behaviour. However, there are several hypotheses about the association between risk perception (i.e. subjective assessments of risk) and behaviour (see Rundmo, in press). It is often assumed that accident happen because there is something “wrong” with individuals’ perception of risk. “Misjudgement” of risk may cause inappropriate decisions as well as unsafe behaviour and “human error”. Accordingly, several studies have been aimed at classifying such errors (Mashour, 1974, Miller and Swain, 1987, Rasmussen, 1981, Reason, 1994, Swain and Guttmann, 1983, Taylor and Lucas, 1991). If the “real” risk is “misperceived” it should be corrected (Marek et al., 1985) and if the demands placed on the individual exceed some assumed “limits” in human “information processing capacity” (e.g. Shiffrin et al., 1973) the factors causing the problems should be removed. Therefore, to avoid accidents tasks have to be fitted to “human limitations” and taken care of in the design phase, e.g. of an industrial plant. However, can misjudgements and human error be the core causal factor of accidents? The relation between perceived risk and behaviour is that the affective reactions may be seen as reactions to stress, i.e. a strain factor, which may reduce a person's ability to cope with risk and therefore cause enhanced risk behaviour (Glendon and McKenna, 1995: 203–242). “Stress” at work may cause “strain” and affect the capacity to cope. In addition, employees may know when he or she is exposed to a risk source, judge the probability of experiencing an accident to be enhanced, and hence he or she feels unsafe and worried. The risk may cause affective reactions and strain (Ulleberg and Rundmo, 1997), which may enhance risk taking behaviour and the probability of experiencing accidents and near misses (Rundmo, 1992). Consequently, when an employee is exposed to a risk, an accident may happen because the perceived risk corresponds to the “real” risk. To perceive accidents to be probable and to be worried when thinking about the risks is a negative judgement. The less risk they experience, the better. However, risk awareness is also an ideal and the fact that a potentially hazardous risk source is recognised as so, have to be considered obligatory. Managers should therefore agree that accidents are serious events and that near misses are a sign that also more accidents could happen. Both these statements indicate priority of activity related to safety matters, which may give safety promotion and accident prevention priority and, hence, contribute to safety. Managers should feel nauseous and afraid when thinking about what consequence an accident and a serious injury could cause to a victim.

Several studies have shown that powerlessness is negatively associated with problem solving behaviour. People avoid seeking help because they believe it implies incompetence and dependence, and therefore is related to powerlessness (Lee, 1997). Powerlessness also has been found to be positively correlated with dissatisfaction (Begin et al., 1997), mistrust and decreased knowledge (Ross and Reynolds, 1996), anxiety (Fiske et al., 1996), lack of job control coping (Armstrong-Stassen, 1994), and risk behaviour (van Wesenbeeck et al., 1994).

Management priorities of safety versus production goals is important for the safety status. It is the strongest predictor of acceptability of rule violations, which also may influence employee risk behaviour (Rundmo, 1996). Therefore, managers should agree on statements giving priority to safety. Managers should ideally feel that they master their work situation and their duties. Black and Frauenknecht (1997) found that mastery was positively associated with problem solving and stress management. Thoits (1994) found that mastery served to select individuals into and out of stressful circumstances. Vinokur et al. (1995) found that a sense of mastery among recent job losers benefited reemployment and mental health outcomes. It may be ideal to report mastery relating to the job duties and to agree e.g. that one has an influence, can achieve what one really sets ones minds to do, and has a lot to gain from personal involvement in safety. Furthermore, it may be non-ideal to see and lay weight on hindrances for involvement in accident prevention, i.e. that employees admit involvement in accident prevention to be time consuming and that it makes it difficult to reach production goals.

Some dimensions of safety attitudes are intended to measure variance in characteristics supposed to have a positive effect on accident prevention and safety promotion, i.e. management commitment and involvement in accident prevention and safety promotion. Other dimensions, as fatalism, powerlessness and perceived hindrances, measure aspects which inhibit managers’ behaviour related to safety promotion and accident prevention. An ideal attitude related to these factors will not by necessity have the same effect on behavioural intentions and behaviour as an ideal answer on dimensions intended primarily to measure facilitating attitudes.