Once more : testing the job characteristics model

Jacobs, Charl Jacobus (2014-04)

Thesis (MComm)--Stellenbosch University, 2014.

Thesis

ENGLISH ABSTRACT: The Job Characteristics Model (JCM) is one of the most widely used and researched models in the field of Industrial Psychology. It has provided industry with useful solutions for its people-related business problems through the rearranging of the physical and psychological characteristics of jobs in order to address demotivation, dissatisfaction and marginal performance. The JCM has also endured a fair amount of criticism, however, specifically pertaining to the mediating role of the psychological state variables. Research findings on the model are divided into two camps. Some researchers argue that the model is empirically sound; while others believe the model should be discarded or adjusted. These studies were done circa 1990, however, when most of the advanced statistical analysis techniques utilised today were not available. Research related to the JCM has been decreasing steadily since then, and it seems that no final verdict was reached regarding the utility and validity of the model. The overarching objective of this study is to provide closure regarding this discourse by testing the three major theoretical postulations of the JCM in the South African context on a sample of 881 students with an ex post facto correlational research design. This was achieved by utilising structural equation modelling via LISREL. Three separate structural models were fitted and compared. The first model was a simplified version of the original model (Hackman & Oldham, 1980). The second model excluded the mediating psychological states proposed by Boonzaier, Ficker and Rust (2001). The final model had the same basic structure as the first model, but more causal paths were included between the job characteristics and the psychological states. The results show that more variance in the outcomes is explained with the inclusion of the psychological state variables. The psychological states are therefore a crucial component of the model. Although these findings corroborated the original model, the third model displayed superiority in terms of accounting for significant amounts of outcome variance in the dependent variables. These findings indicate that the job characteristics predict the psychological states in a more comprehensive manner than originally proposed in the literature. Job design interventions thus remain a useful tool and industry should utilise the suggested interventions. Furthermore, this study proposes preliminary equations (a Motivating Potential Score and resource allocation) that may be used to determine the relative importance attached to each job characteristic in the world of work.

AFRIKAANSE OPSOMMING: Die Taakeienskappe Model (Job Characteristics Model, JCM) is een van die Bedryfsielkunde-modelle wat die meeste gebruik en nagevors word. Dit het aan die bedryf bruikbare oplossings vir mensverwante besigheidsprobleme verskaf deur die herrangskikking van die fisiese en sielkundige eienskappe van werk om probleme soos demotivering, ontevredenheid en marginale prestasie aan te spreek. Die JCM is egter ook al baie gekritiseer, spesifiek rondom die bemiddelende rol van die sielkundige toestand veranderlikes. Navorsingsbevindinge oor die model word in twee groepe verdeel. Die een groep argumenteer dat die model empiries foutvry is, terwyl die ander groep glo dat dit weggedoen of aangepas moet word. Hierdie studies is egter in die 1990’s gedoen, toe die meeste van die gevorderde statistiese tegnieke wat vandag gebruik word, nie bestaan het nie. Navorsing oor die JCM het sedertdien stadig maar seker afgeneem, en geen finale besluit oor die bruikbaarheid en geldigheid van die model is al geneem nie. Die oorkoepelende doel van hierdie navorsing was om van die bogenoemde probleme te probeer oplos deur drie vername teoretiese uitgangspunte oor die JCM in die Suid-Afrikaanse konteks te toets deur middel van ‘n steekproef van 881 studente. Dit is met behulp van struktuurvergelykingsmodellering deur middel van LISREL gedoen. ‘n “Ex post facto” korrelasionele navorsings ontwerp is benut. Drie aparte strukturele modelle is gepas en vergelyk. Die eerste model was ’n vereenvoudigde weergawe van die oorspronklike een (Hackman & Oldham, 1980). Die tweede model het die bemiddelende sielkundige toestande uitgelaat wat deur Boonzaier, Ficker en Rust (2001) voorgestel is. Die finale model het dieselfde basiese struktuur as die eerste een gehad, maar nuwe oorsaaklike weë is tussen die werkseienskappe en sielkundige toestande ingesluit. Die resultate toon dat meer variansie in die uitkomstes verduidelik word wanneer die sielkundige toestand veranderlikes wel ingesluit word. Die sielkundige toestande is dus ’n kritieke komponent van die model. Hoewel hierdie bevindinge die oorspronklike model staaf, het die derde model die noemenswaardige variansie in uitkomstes van die afhanklike veranderlikes beter verklaar. Hierdie bevindinge dui daarop dat die werkseienskappe die sielkundige toestande meer omvattend voorspel as wat aanvanklik in die literatuur voorgestel is. Werksontwerp-intervensies is dus nog steeds ’n bruikbare hulpmiddel en die bedryf moet die voorgestelde intervensies gebruik. Hierdie studie stel ook voorlopige vergelykings voor (Motiverings Potensiaal Telling en hulpbrontoewysing) wat gebruik kan word om die relatiewe belangrikheid van elke werkskenmerk in die wêreld van werk te bepaal.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/86199
This item appears in the following collections: