Mechanisms for Influence of Safety Climate and Psychological Climate on Safety Behaviour

A hypothetical model was tested, where several dimensions of Safety Climate and Psychological Climate were viewed as antecedents, the effects of which were mediated through five determinants that, in turn, predicted Safety Behaviour. The model was applied to questionnaire data from a large-scale construction project. Answers from 192 blue-collar workers were analysed using structural equation modelling technique. The results showed that both Safety Climate and Psychological Climate influenced safety behaviour. Furthermore, four of the five suggested determinants (Safety Motivation, Safety Knowledge, Felt Stress and Workplace Commitment.) seemed to mediate these relationships, and thus qualify as mechanisms for how these influences occurred.
Keywords: 
safety climate, psychological climate, safety behaviour, structural equation modelling, mediated model, mechanisms
Main Author: 
Anders
Pousette
Co-authors: 
Marianne
Törner
Susanna
Larsson

Pousette, Anders; Törner, Marianne; Larsson, Susanna

National Institute of Working Life West/ P.O. Box 8850/ SE-402 72 Göteborg, Sweden

46 31 50 16 33 / anders.pousette@arbetslivsinstitutet.se

ABSTRACT

A hypothetical model was tested, where several dimensions of Safety Climate and Psychological Climate were viewed as antecedents, the effects of which were mediated through five determinants that, in turn, predicted Safety Behaviour. The model was applied to questionnaire data from a large-scale construction project. Answers from 192 blue-collar workers were analysed using structural equation modelling technique. The results showed that both Safety Climate and Psychological Climate influenced safety behaviour. Furthermore, four of the five suggested determinants (Safety Motivation, Safety Knowledge, Felt Stress and Workplace Commitment.) seemed to mediate these relationships, and thus qualify as mechanisms for how these influences occurred.

Keywords

Safety Climate, Psychological Climate, Safety Behaviour, Structural Equation Modelling, Mediated Model, Mechanisms.

INTRODUCTION

Safety Climate is considered to play an important role for safety performance [8]. Psychological Climate is known to influence transitional states, such as felt stress and workplace commitment [5,9]. These transitional states are possible determinants for Safety Behaviour. Safety Climate is here treated as shared perceptions on certain dimensions related to safety at both organisational and work group levels. Psychological climate refers to how individuals in the organisation in general perceive their psychosocial conditions (psychosocial work environment).

Although it seems clear that Safety Climate is of importance for Safety Behaviour, and there is growing evidence that Psychological Climate also is of importance, the knowledge on how this influence take place is limited. In other words, what are the mechanisms? Some mechanisms are proposed in the literature. Safety Motivation and Safety Knowledge has been shown to mediate the relations between Safety Climate and Safety Behaviour [2,8]. Stress has been proposed as a mediator between individual factors and accident liability [6]. Furthermore, work related attitudes are proposed as mediators between Psychological Climate and Organisational Behaviour [9]. The aim of the present study was to investigate five possible mechanisms for influence of Safety Climate and Psychological Climate on Safety Behaviour.

In the study, a hypothetical model was tested, where Safety Climate (six dimensions) and Psychological Climate (eleven dimensions) were viewed as antecedents, the effects of which were mediated through determinants (five dimensions) that, in turn, predicted Safety Behaviour (three dimensions), Figure 1.

The five suggested determinants were: 1) Safety Motivation, 2) Safety Knowledge,

3) Felt Stress, 4) Job Satisfaction and 5) Workplace Commitment.

Psychological Climate

11 dimensions

Safety Climate

6 dimensions

Determinants

5 dimensions

Safety Behaviour

3 dimensions

Figure 1. Hypothesized relationship between Psychological Climate, Safety Climate, and Safety Behaviour, mediated through Determinants.


METHODS

The study was based on cross-sectional data from the third wave of a longitudinal research project. This research project studies safety issues within a major construction project where a road tunnel is being built under central parts of Göteborg, Sweden. Data were collected by means of a comprehensive questionnaire.

The questionnaire was directed to all personnel directly involved in the construction project. Response rate was 87%. For the purpose of the present study, answers from all 192 non-managerial construction workers were analysed. Mean age was42.6 years (SD= 11.7) and mean job tenure was 19.5 years (SD= 12.0). All were males.

Measures of 6 dimensions of Safety Climate, 11 dimensions of Psychological Climate, 2 dimensions of general work attitudes and 1 dimension of felt stress were collected. The safety climate factors were based on a model of safety climate developed by Cheyne et al. [2] in British and French manufacturing industry. The underlying questionnaire items were further developed and expanded with two more climate factors building on concepts acquired from Zohar [13]. The psychological climate factors were adapted from the Copenhagen Psychosocial Questionnaire [5]. Each factor was measured with scales composed of at least three items. Origin of each scale and reliability in the present sample are showed in Table 1. Reliability was sufficiently high for most factors.

Three different aspects of self-rated Safety Behaviour were measured. These scales were developed by the authors based on work by Cheyne et al [2]. The first dimension, Structural Safety Behaviour (SSB), concerned participation in organised safety activities (e.g. taking part in risk assessment). The second dimension, Interactional Safety Behaviour (ISB), concerned safety activities in the daily work in interaction with co-workers and management (e.g. discussed a safety issue with a fellow worker). The third dimension, Personal Safety Behaviour (PSB), comprised behaviour aimed at personal protection (e.g. use of all prescribed protective equipment, following safety rules). Although related to each other, the three safety

behaviour factors have been found to constitute different manifestations of how workers claim to behave in relation to safety in the workplace [11]. Reliability was good for all three scales, Table 1.

The hypothetical model was tested using structural equation modelling technique. For modelling, AMOS 4 was used [1]. To evaluate the fit of the model to

Table 1. Sources, number of items and reliability for scales in the study.


Factor Scale source No item Reliability (alpha)


Safety climate: Organisational perspective

Management safety priority

[2] 1)

4

.87

Safety management

[2] 1)

16

.93

Safety communication

[2] 1)

8

.83

Safety incentives

[7,13]

3

.71

Workgroup safety involvement

Safety climate: Workgroup perspective


[2] 1) 8 .78

Safety social status [7,13] 3 .60


Psychological climate

Quantitative demands

[5] 1)

4

.61

Cognitive demands

[5]

4

.78

Role-clarity

[5]

4

.71

Role-conflicts

[5] 1)

4

.78

Influence at work

[5]

5

.68

Possibilities for development

[5]

4

.61

Predictability (information)

[5] 1)

3

.57

Sense of community

[5]

3

.83

Social support

[5]

4

.77

Feedback at work

[5] 1)

3

.75

Quality of leadership

[5]

4

.87

Commitment to the workplace

Psychological determinants(mediators)


[5] 3 .63

Job satisfaction [10,12] 3 .84

Felt stress [4] 6 .86


Safety determinants(mediators)

Safety motivation

[2] 1)

7

.76

Safety knowledge

2)

4

.77


Safety behaviour (outcome variables)

Structural safety behaviour (SSB)

Interactional safety behaviour (ISB)


Personal safety behaviour (PSB)

[2] 1) 8 .79

[2] 1) 5 .76

2) 6 .85

Note 1) Scale sources: Further developed within the present project; 2) Developed within the present project.

sample data, several fit indices were used, as recommended in literature [1,3].

RESULTS

The overall fit of the model was good: chi-square (61,N= 192) = 159.6, p

Papers relacionados

ORP 2015
MARIA EUGENIA
LONDOÑO LONDOÑO
Centro de Investigación en Comportamiento Organizacional Cincel S.A.S.
Colombia
ORP 2015
Raúl
García Bercedo
Universidad del País Vasco
España
ORP 2015
JOSÉ CARLOS
ZANELLI
Faculdade Meridional
Brasil
ORP 2015
Maria Erley
Orjuela Ramirez
Universidad Nacional de Colombia
Colombia