Page 211 - Quantitative Data Analysis

Quantitative Data Analysis
Simply Explained Using SPSS

support the possibility that they are a measure of your hypothesized
construct.

In other words, FA and PCA are data aggregation procedures that
are "clustering" your items together based on the mathematical
relationships of the items (obtained from similarities in participants'
responses). The FA or PCA results provide evidence of whether your
groups of items are correlated as you hypothesized. Again, it does
not provide direct evidence that your groups of items are measuring
the operational definition you created.

Difference between PCA and EFA

Researcher sometimes mixes-up these two terms and that is
somehow understandable because both PCA and FA are variable
reduction techniques and there are some important similarities
between these two procedures however there are some key
conceptual differences between PCA and FA. The most important
difference between PCA and EFA deals with the assumption of an
underlying casual structure. It is the fundamental distinct between
PCA and EFA.

PCA EFA

V1 V2 V3 V1 V2 V3

In principal component each component is defined as,

The Theory and Applications of Statistical Inferences 195