In the measurement model, the researcher specifies which variables are indicators
of each construct. The structural model specifies the causal relationship among the
constructs.
A raw data file was submitted to EQS 6.1 in order to obtain the Maximum
Likelihood (ML) estimation. Then, the program created the covariance matrix used for
analyses. It is recommended that the minimum sample size to ensure appropriate use of
ML (maximum likelihood) estimation is 100 to 150 (Ding, Velicer, and Harlow 1995).
Heise (1975) suggested a model with a sample size of 200 as being the critical sample
size. Given the total usable sample size of 585, the sample was considered to be of
acceptable size because it is more than the minimum sample size.