Unlike exploratory factor analysis (EFA) where there are no specific a priori
restrictions governing which indicators should measure the effects of which latent
variables (or indeed how many latent variables there might be), with the
confirmatory (CFA) method of factor analysis used in SEM analysis, the role of
theory is central. Specifically, with CFA researchers specify, or, hypothesise
based on theory, which latent variables are to be measured with which specific
indicators. The CFA method then confirms (hence its name) or refutes these
hypotheses.
The remainder of this chapter elaborates the rationale for the inclusion of each of
the latent factors and identifies / develops, from theory or from prior empirical
research findings, a set of indicator variables (items) for each of the latent
factors. As described above using the example from the MLQ, in the survey
instrument, each indicator variable takes the form of a statement: the item
statements are generated according to the advice of Hinkin (1998: 107-108) to
(a) keep statements as short as possible and to (b) ensure that the statement
wordings are understandable to the target population.
5.1 Dispositional factors / individual values
Work Values
The rationale for including a factor relating to individuals’ values is grounded in
the work of McGregor (1966) who described the Theory X / Theory Y model of
managers’ / employers’ assumptions of subordinate employees’ work orientation
(see Section 2.3.2). Theory X management tends to view employees as typically
self-serving and not being inclined to enjoy work – this perspective suggests that
initiating structure leadership behaviours may be more effective. Theory Y on the
other hand describes management assumptions which tend towards viewing
employees as enjoying work and seeking challenges and responsibility at work –
an orientation that is suited more to consideration styles of leadership.
Setting aside the question of how management assumptions inform leadership
behaviours, this research seeks to evaluate what effect, if any, employees’ work
orientations / work values have on the attitudinal outcomes of Organisational
Commitment, Job Satisfaction and Work Meaning. The hypothesised causal flow
(illustrated in Figure 4 - 5 ) from Work Values → Employee Attitudes → Individual ,
group and organisational level outcomes is drawn from the
values→attitudes→behaviour (V→A→B) hierarchy described by Homer and Kahle