Psychological factors as predictors of employees' attitudes towards work place safety in a public electricity company in South Africa

Abstract

English: Research on the relationship between individual differences and work place incidents/accidents, including driver accident causation, originates from as early as 1917. In the 21st century this phenomenon continues to be a critical agenda for research internationally. In addition, a wide range of individual (e.g. intelligence, personality, stress) and organisational (e.g. leadership support, safety culture, job satisfaction) variables have been investigated as latent or observable predictors of causes of why employees engage in atrisk behaviour. Attitude towards work place safety is a well researched phenomenon in parts of the world other than South Africa. In addition, no research in South Africa has suggested a process depicting a combination of psychological factors that are related to employees' attitude towards work place safety. This study explored the influence of psychological factors on attitude towards work place safety. The aim of this study was two-fold: firstly to determine whether a statistically significant relationship exists between the predictor and criterion variables, and secondly through partial least square path modelling (PLS), to determine whether employees' attitude towards work place safety could be influenced by individual psychological factors. The study included intelligence, personality, burnout, work engagement, and sense of coherence as the predictor variables, and safety consciousness (safety control, risk avoidance, stress tolerance), driver attitude and quality orientation as the criterion variables measuring attitude towards work place safety. Based on the literature, the study suggested a model depicting a sequential process of interrelationship amongst the psychological factors and their relationship with safety consciousness, driver attitude, and quality orientation. To test the validity of the theoretical model, the current study used a sample of individuals that have been involved in a work place incident/accident or vehicle accident between 2006 and 2008. A sample of 279 employees from a public electricity company in South Africa participated in the study. The study applied both survey and statistical modelling methodologies to guide the research. Standardised questionnaires were used to measure the different psychological and safety attitudinal variables. The scoring and reliability analysis of the measurement instruments for intelligence and personality variables were conducted by service providers. Exploratory factor analysis was performed on the burnout, work engagement, sense of coherence, and safety attitudinal constructs to determine the applicability of the factor structures to the current sample. The suggested factor structures were confirmed through confirmatory factor analyses with acceptable levels of fit. Hypothesis I was supported through correlation analysis, confirming that statistically significant path coefficients exist between the different variables. Furthermore, the results of the regression analysis suggested that 9 psychological factors explained 67% of the variance on safety consciousness, 5 psychological factors explained 53% on driver attitude, and 7 psychological factors explained 35% on quality orientation. The Partial Least Squares Modelling approach was used to evaluate the reliability of the psychological and safety attitudinal factors included in the measurement model (outer model), as well as the evaluation of the theoretical model (inner model) of the current study. Acceptable levels of fit were obtained which guided the process for the evaluation of the proposed model. In the evaluation of the pathways between latent constructs statistically significant path coefficients were found between (1) intelligence and driver attitude, (2) personality and burnout, (3) personality and sense of coherence, (4) personality and safety consciousness, (5) sense of coherence and burnout, (6) sense of coherence and work engagement, (7) sense of coherence and quality orientation, (8) driver attitude and safety consciousness, and (9) quality orientation and safety consciousness. Of importance is that only intelligence, personality and sense of coherence were found to significantly influence participant attitude towards workplace safety. Personality and sense of coherence moderated the effect of burnout experience, while sense of coherence contributed positively to experiences of work engagement. The path relationship between the safety attitudinal constructs with burnout and work engagement was non significant. The study suggested that the theoretical model requires further refinement as well as the testing of the model on a larger sample.

Afrikaans: Die verhouding tussen individuele verskille en die oorsaak van werksplek insidente/ongelukke insluitend voertuigongelukke word reeds vanaf die jaar 1917 nagevors. Internasionaal word hierdie verskynsel steeds as 'n kritiese agenda vir navorsing in die 21ste eeu beskou. Benewens dit, is daar reeds ondersoek na 'n wye reeks latente of waarneembare individuele (bv. intelligensie, persoonlikheid, stres) en organisatoriese (bv. leierskapondersteuning, veiligheidskultuur, werkstevredenheid) veranderlikes as voorspellers vir die oorsake waarom werknemers by riskogedrag betrokke raak gedoen. Houdings teenoor werksplekveiligheid is in ander dele van die wêreld 'n weldeurdagte navorsingsverskynsel. Daarbenewens, kon geen vorige navorsing in Suid-Afrika wat 'n proses van 'n kombinasie van psigologiese faktore wat verband hou met werknemers se houding teenoor werksplekveiligheid gevind word nie. Die studie het die invloed van psigologiese faktore op houdings teenoor werksplekveiligheid verken. Die doel van die studie was tweeledig, eerstens om vas te stelof 'n statistiese betekenisvolle verband tussen die voorspeller- en kriterium veranderlikes bestaan, en tweedens deur middel van gedeeltelike minste vierkantige modellering "Path Least Square Modelling" te bepaal of 'n werknemer se houding teenoor werksplekveiligheid deur psigologiese faktore beinvloed kan word. Intelligensie, persoonlikheid, uitbranding, werksbegeestering, en koherensiesin was as die voorspellerveranderlikes en veiligheid van bewussyn (veiligheidbeheer, risikovermyding, strestoleransie), voertuigbestuurder houding, en oriëntasie teenoor kwaliteit wat houdings teenoor werksplekveiligehid beinvloed, was as die kriterium veranderlikes in die studie gebruik. Uit die literatuur, het die studie 'n logiese volgorde wat die interverwantskap tussen die psigologiese faktore en hul verwanskap met veiligheid van bewussyn, voertuigbestuurder houding, en oriëntasie teenoor kwaliteit voorgestel. Om die geldigheid van die proses te bepaal, het die studie van 'n groep proefpersone wat tussen 2006 en 2008 in 'n werksplek insident/ongeluk of voertuig ongeluk betrokke was gebruik gemaak. 'n Groep van 279 werknemers in diens van 'n 'n publieke elektrisiteitsorganisasie in Suid Afrika, het aan die studie deelgeneem. Beide opname- en statistiese modelleringsmetodiek was gebruik. Gestandardiseerde vraelyste was gebruik om die verskillende psigologiese en veiligheidshoudings veranderlikes te meet. Die berekenings en geldigheidsontleding van die meetinstrumente wat intelligensie en persoonlikheid meet was deur die diensverskaffers bereken. Die studie het van verklarende faktorontleding gebruik gemaak om die toepaslikheid van die faktorstruktuur van die uitbranding, werksbegeestering, koherensiesin, en veiligheidhoudings konstrukte op die steekproef te bepaal. Die voorge telde faktorstruktuur van die meetinstrumente was deur bevestigende faktor ontleding bevestig met aanvaarbare passing statistiek. Hipotese I was deur korrela ie ontleding bevestig wat aangedui het dat daar wel 'n statistiese betekenisvolle verband tu en die verskillende veranderlikes bestaan. Verder het die resultate van die regressieontleding aangedui dat 9 psigologiese faktore 67% van die variansie op veiligheid van bewussyn voorspel het, 5 psigologiese faktore 53% van die variansie op voertuigbestuurder houding voorspel het, en 7 psigologiese faktore 35% van die variansie op oriëntasie teenoor kwaliteit voorspel het. Daar was van die minste vierkantige modellering "Path Least Square Modelling" benadering gebruik gemaak om die betroubaarheid van die psigologiese en veiligheidshoudingsfaktore wat in die meetingsmodel (buite model) ingesluit was te bereken, asook met die evaluering van die teoretiese model (binne model). Aanvaarbare passingsstatistiek was verkry wat die rigting aangedui het vir die evaluering van die teoretiese model. Tydens die evaluering van die verwantskappe tussen die latente konstrukte van die teoretiese model was die volgende statistiese betekenisvolle padkoeffisiënte verkry (I) intelligensie en voertuigbestuurder houding, (2) persoonlikheid en uitbranding, (3) persoonlikheid en koherensiesin, (4) persoonlikheid en veiligheid van bewussyn, (5) koherensiesin en uitbranding, (6) koherensiesin en werk begeestering, (7) koherensiesin en oriëntasie teenoor kwaliteit, (8) voertuigbestuurder houding en veiligheid van bewussyn, en (9) oriëntasie teenoor kwaliteit en veiligheid van bewussyn. Van groot waarde was dat slegs intelligensie, persoonlikheid en koherensiesin die proefpersone e houding teenoor werksplekveiliheid betekenisvol beinvloed het. Per oonlikheid en 'n sterker koheresiesin het die effek op uitbranding gemodereer, terwyl kohere iesin 'n positiewe uitwerking op werksbegeestering getoon het. Die padverwantskap tussen die veiligheidhoudingskonstrukte met uitbranding en werksbegeestering was nie betekenisvol nie. Die studie het voorgestel dat die teoretiese model verder verfyn moet word asook die toetsing van die modelop 'n groter steekproef.