8 - Exploratory Factor Analysis II

Flashcards based on lecture notes about Factor Analysis and PCA

What do factor loadings represent?

Correlation coefficients between factors and original variables

What is the primary goal of PCA?

To extract the maximum variance from the data and ensure orthogonality.

What is the solution to the problem of unclear factor interpretation in PCA?

Rotate the coordinate system.

What are two different rotation methods in factor analysis?

Orthogonal and Oblique

What does Varimax rotation aim to maximize?

The variance of the factor loadings for each factor.

What is a key characteristic of Oblimin/Promax rotation?

Factors lose their orthogonality, allowing correlations to achieve simple structure.

What is the purpose of factor rotation?

To make factor loadings more distinct and achieve a simple structure.

What are the desired characteristics of factor loadings after rotation?

High absolute loadings on one factor and near-zero loadings on others.

What is a Bartlett test of sphericity used for?

To determine if the covariance matrix is suitable for factor analysis.

What does the Kaiser-Meyer-Olkin (KMO) test measure?

Sampling adequacy.

What is a general recommendable range for the ratio of participants to items in EFA?

Between 5:1 and 10:1.

What does communality (h2) represent?

The amount of variance of an original variable explained by all extracted factors.

What are cross-loadings?

Factor loadings of variables that have significant loadings on more than one factor.

What does Cronbach's alpha assess?

Internal reliability of items loading on the same factor.

What is the purpose of factor scores?

To provide individual values on the factors.

Define Eigenvectors in the context of factor extraction

Rotate the raw data into the new coordinates/factors therefore could also be called [initial] ‘rotation matrix’

Define Eigenvalues.

Report the Fof variance explained by each of the new factors