week 1

Question 1

What is factor analysis?

Answer 1

A technique for identifying latent constructs underlying a group of variables, used to categorise variables into groups based on common variance–What is the overlap between variables.

Question 2

What is an example of factor analysis

Answer 2

E.g. a scale that tests neuroticism and subscales of it (depression, anxiety and hostility eg), so how do we know what items assess subscales? Factor analysis

Question 3

What are the two types of extraction?

Answer 3

Maximum likelihood, principal axis factoring

Question 4

What are communalities?

Answer 4

Initial communalities are the proportion of variance in each item
accounted for by the rest of the items–higher the better! Smaller values indicate variables do not fit the factor solution.

Question 5

What is an Eigenvalue?

Answer 5

An eigenvalue for a given factor measures the variance in all of the
items that is accounted for by that factor
Think of the eigenvalues as providing the numeric “key” that allows the presence of factors to be decoded, we only use the largest as an indicator of number of factors to retain. We want factor solutions to be simple and straightforward.

Question 6

What are the two guides to factor retention?

Answer 6

Kaisers (1960) criterion: the number of eigenvalues > 1 reflects the number of factors that should be retained.

Cattells (1978) scree plot (useless).

Question 7

Does % of variance matter?

Answer 7

James says a factor that explains 10% of variance is useful, though the more variance explained the better.

Question 8

What are the two types of rotation/

Answer 8

Orthogonal rotations (e.g., varimax): These maximise the
difference between factors, and hence facilitate independent,
uncorrelated factors
• Oblique rotations (e.g., direct oblimin, promax): These are more
complex rotations that facilitate correlated factors