Mathematical Model Efa And Cfa - 1474 Words

1) Mathematical model: EFA and CFA are similar in terms of their respective mathematical equation/model. While PCA omits error (b/c it assumes there is perfect reliability), the main distinction between CF and CFA is the fact that CF (EFA) does not constrain factor loadings and CFA does (a priori). 2) Measure selection: selection of measures can be haphazard or based on intuition in EFA, however it must be deliberate in CFA as the measures must be representative of the latent factors that were determined a priori by theory. The content validity of measures is much more important in CFA than EFA. 3) Number of Factors: the number of factors in EFA is determined by the communalities that arise in the factor loadings. Factors are not usually determined a priori (but they can be) and thus the number of factors in EFA is not a big concern. The number of factors in CFA is established a priori based on theory and is an important aspect of CFA. In CFA, constraints are implemented so that only a certain number of factors are possible. 4) Factor Pattern Matrix: the factor pattern matrix in EFA allows all items to load onto all factors. In CFA, the pattern contains both free (estimated) and fixed factor loadings so that only certain items load onto certain factors. 5) Interpretation of Factors: the communalities seen in the factor loadings in EFA determine the emergence of factors. Factors are determined by the high loading items clustered onto that respective factor and a lowShow MoreRelatedFactors Influencing The Factor Analysis2512 Words   |  11 Pages Osborne, 2005). Exploratory Factor Analysis (EFA) Costello Osborne, (2005) said that, factor analysis is used to uncover the structure of relatively large set of variables in the data(Costello Osborne, 2005). EFA identifies the underlying relationship between the variables or items in the data set. The method is particularly used in developing the scale and identifying the set of latent constructs as stated by (Bi, Lee, O Mahony, 2015). EFA is used when the researcher has no prior hypothesis