Thesis (MComm)--Stellenbosch University, 2012. / ENGLISH ABSTRACT: Pay satisfaction is an essential aim in any company’s reward system, since various studies
indicate that it affects employee job performance. With limited resources and restricted capacity
to increase pay, firms require a better understanding of how employee pay satisfaction develops
and how employee work-related behaviour is affected by pay satisfaction. The purpose of this
study was threefold: First, to investigate the relationship between actual pay and pay satisfaction,
and second, to explore the effect that pay motivation dimensions derived from expectancy theory
(Vroom, 1964) — pay valence and pay instrumentality — have on this relationship. Lastly, the
study aimed to explore the relative effects of these objective (pay) and subjective (pay
perceptions) variables on employee job performance.
Self-report survey data of managerial employees (N = 177) from a South African retail
organisation were collected, along with supervisor ratings of job performance and objective
actual pay data, for each participant. The Pay Satisfaction Questionnaire (PSQ) and a selfdeveloped
Pay Valence and Instrumentality Questionnaire (PVIQ) were administered through an
intranet-based survey.
Data were analysed with a variety of statistical techniques. Descriptive statistics were used to
assess underlying assumptions of the multivariate analysis techniques used to analyse the
research data, and to describe the central tendency and dispersion of study variables. Pearson
product moment correlations were calculated to assess bivariate relationships. Standard multiple
regression analysis was utilised to assess the joint relationship between IVs and DVs, and
relative weights analysis (RWA) to assess the relative importance of IVs within respective
regression models. Moderated multiple regression (MMR) analysis was used to examine possible
moderator effects. To determine whether mediating (indirect) influences were present, normal
theory and bootstrap estimates of indirect effects were obtained.
In this research, selected antecedents and consequences of pay satisfaction were investigated.
Regarding antecedents of pay satisfaction, the results suggested that actual pay had a small but
significant effect on pay satisfaction, but not pay valence, nor pay instrumentality. Pay
motivation perceptions (pay valence and pay instrumentality) neither moderated, nor mediated, the relationship between pay and pay satisfaction. The consequences of pay perceptions, in
conjunction with actual pay, were also assessed. Pay satisfaction had no statistically significant
relationship with job performance (p > .05), although actual pay had a moderate association with
job performance. Although pay perceptions slightly incremented the validity of predicting
performance from actual pay itself, this increment was not statistically significant (p < .05).
In summary, the present research highlights the central role of actual pay in influencing pay
satisfaction and job performance, but questions still remain about the way in which performance
results from pay, since pay satisfaction did not mediate this relationship. The implications for
future research are discussed and recommendations for research are made. / AFRIKAANSE OPSOMMING: Vergoedingstevredenheid is noodsaaklike doelwit in maatskappy se vergoedingsisteem en
verskeie studies toon hoe dit werksprestasie van werknemers positief beïnvloed. Aangesien
maatskappye oor beperkte hulpbronne en kapasiteit beskik om vergoeding te verhoog, word ‘n
beter begrip vereis oor hoe werknemers se gedrag deur vergoedingstevredenheid beïnvloed word.
Die doel van hierdie studie was drieledig: eerstens, om die verband tussen werklike vergoeding
en vergoedingstevredenheid te ondersoek, en tweedens, om die effek van
vergoedingsmotiveringsdimensies — afgelei van die verwagtingsteorie (Vroom, 1964), naamlik
betalingsvalensie en betalingsinstrumentaliteit — en die verband te ondersoek. Derdens, om die
invloed van hierdie objektiewe (vergoeding) en subjektiewe (vergoedingspersepsies)
veranderlikes op werknemerprestasie te ondersoek.
Die data is in Suid-Afrikaanse kleinhandelsonderneming versamel. Die volgende inligting is bekom van deelnemers: selfrapporteringsdata vanaf werknemers in bestuursposte (N = 177),
werksprestasiebeoordelings deur hul onderskeie toesighouers, asook objektiewe werklike
vergoedingsvlakke vir elke respondent. Die Vergoedingstevredenheidsvraelys en die
selfontwikkelde Betalingsvalensie- en Instrumentaliteitvraelys is deur intranetopname
geadministreer.
Die data is met behulp van verskeidenheid statistiese tegnieke ontleed. Beskrywende statistiek
is gebruik om die onderliggende aannames van meerveranderlike analise in die navorsingsdata te
evalueer, en die sentrale neiging en verspreiding van die studieveranderlikes te beskryf. Pearsonprodukkorrelasiestatistiek
is gebruik om die beduidendheid van die hipoteses rakende tweeveranderlike korrelasies te evalueer. Standaard meervoudige regressie-analise is gebruik om die gesamentlike verwantskap tussen onafhanklike veranderlikes en afhanklike veranderlikes te evalueer. Relatiewegewigte-analise (RWA) was gebruik om die relatiewe belangrikheid van
onafhanklike veranderlikes, binne die onderskeie modelle, te evalueer. Gemodereerde
meervoudige regressie-(MMR)-analise is gebruik om die moontlike bemiddelende uitwerking van veranderlikes te ondersoek. Normale teorie en “bootstrap”-raming van die indirekte invloede
van veranderlikes is verkry om die moontlike mediëring van verbandskappe te evalueer.
In hierdie navorsing is die voorafgaande faktore, sowel as die gevolge van,
vergoedingstevredenheid ondersoek. Wat die oorsake van vergoedingstevredenheid betref, het
die resultate aangedui dat werklike vergoedingsvlakke klein, maar wel statisties beduidende
uitwerking het op vergoedingstevredenheid, terwyl betalingsvalensie en
betalingsinstrumentaliteit geen beduidende uitwerking getoon het nie. Verder toon die resultate
dat die verband tussen betaling en vergoedingstevredenheid word nie deur
vergoedingsmotiveringspersepsies (betalingsvalensie en betalingsinstrumentaliteit) bemiddel of
medieër nie. Die nagevolge van vergoedingspersepsies is onderling met werklike betaling
ondersoek. Vergoedingstevredenheid het geen statistiese beduidende verband met werksprestasie
getoon nie, tog het werklike betaling matige verband met werksprestasie getoon. Alhoewel
betalingspersepsies die geldigheid van die voorspelling van werksprestasie vanaf werklike
betaling inkrementeer, toon dit nie beduidende effek (p < .05) nie.
Die huidige navorsing beklemtoon die sentrale rol wat werklike betaling steeds in die
beïnvloeding van vergoedingstevredenheid en werksprestasie speel, hoewel die rede vir die
verband tussen werklike betaling en prestasie steeds nie ten volle begryp word nie, aangesien
vergoedingstevredenheid nie hierdie verband medieër nie. Beperkinge van die navorsing, asook
die implikasies vir vergoedingspraktyk en toekomstige navorsing word bespreek.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/20410 |
| Date | 03 1900 |
| Creators | Badenhorst, Marizanne |
| Contributors | De Kock, Francois, Stellenbosch University. Faculty of Economic and Management Sciences. Dept. of Industrial Psychology. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 119 p. |
| Rights | Stellenbosch University |
Page generated in 0.0129 seconds