The current research draws upon and modifies the hospitality catering-centred
model established by Hancer and George (2003). For this study, Empowerment is
conceptualised primarily as Influence, which embodies Spreitzer’s original
conceptual domains of Self-Determination and Impact and is interpreted as ‘an
individual’s perceptions of (i) autonomy in decision making and (ii) ability to effect
process and outcomes at work’.
Including the Competence and Meaning factors in this study was not desirable for
the following reasons.
Firstly, regarding the Meaning factor, for this study Work Meaning is being
conceptualised as a discrete factor which is not located within the Empowerment
construct. The Work Meaning factor in this research is hypothesised as a
consequent of Empowerment based on the rationale that increased individual
perceptions of influence (self-determination and impact/personal control) may
enhance work meaningfulness. With respect to Spreitzer’s empowerment model,
the current conceptualisation and hypothesis linking Empowerment and Work
Meaning effectively repositions the Work Meaning factor as a consequent of the
Influence factor rather than as a covariate with it. Theoretical and empirical
support for this relationship can be found in organisational psychology literature
on job crafting (e.g. Wrzesniewski and Dutton 2001; Ghitulescu 2006; Berg et al.
2010) where evidence has been found linking employee autonomy in the
workplace to increased perceptions of work meaning (Wrzesniewski and Dutton
2001). Furthermore, the operationalisations of Meaning found in Spreitzer and in
Fulford and Enz/Hancer and George are found to be somewhat nebulous, leading
to a search for an operationalisation of the construct which embodies a more
specific and substantive interpretation of the factor. Details of this critique are
provided in Section 5.3 on Work Meaning below.
Secondly, regarding the Competence factor; while Hancer and George used
principal components analysis (a form of exploratory factor analysis), this study
uses confirmatory factor analysis (CFA). There are specific requirements for CFA
regarding the number of items (indicators) required to adequately measure each
latent factor – this is an aspect of SEM referred to as factor identification and is
described fully in Section 6.1. In short, a single factor (latent construct, or
variable) which is connected to other factors (latent variables) in an SEM model
needs to be measured by at least 2 significant indicator variables – starting off
with only two indicator variables (i.e. as per Hancer and George’s Competence