Use Confirmatory Factor Analysis (CFA) and Structural Equation Modelling (SEM) to construct a model relating leadership characteristics to bullying and mental health outcomes based on evidence from the research literature.
statistics
Description
Use Confirmatory Factor Analysis (CFA) and Structural
Equation Modelling (SEM) to construct a model relating leadership
characteristics to bullying and mental health outcomes based on evidence from
the research literature.
Results of CFA and SEM =1250 words
Brief discussion = 750 words
Structural Equation Modelling
7. State your approach to assessing model fit.
Example
from a recent paper for this question:
All confirmatory models were estimated
in LISREL 8.8 using maximum likelihood estimation. The suitability of each
model was assessed by firstly examining each of the modelled parameters, and
secondly, by assessing model fit. The use of absolute fit indices in model evaluation has recently been
questioned (Chen, Curran, Bollen, Kirby &
Paxton, 2008), and there is
no consensus on which indices should be used. We
rely partly on the simulations of Hu and Bentler (1998, 1999), which suggest
the utility of the standardized root mean square residual (SRMSR), the root
mean square error of approximation (RMSEA), and the non-normed fit index
(NNFI). We adopted cut-off points of £
.05 for the SRMSR (Spence, 1997), about .06 for the RMSEA (Browne & Cudeck,
1993), and ≥ .95 for the NNFI and CFI, which conform to recent recommendations
based on Monte Carlo simulation (Hu & Bentler, 1998, 1999). In order to
make direct comparisons between plausible models, we adopt the Akaike
Information Criterion (AIC; Akaike, 1974). The AIC can be used to compare two
models containing different numbers of parameters, with the lower values
indicating the better model (Kuha, 2004).
8. Secondly, describe the analyses
of the measurement model (table 5 below) and the major features of the
resulting model including assessment of fit.
Table 5. Final measurement model showing items,
standardized factor loadings and Squared Multiple Correlations (SMCs), and
average variance extracted (AVE)
