Summary Factor Analysis

Question 1

What is factor analysis used for?

Answer 1

Summarizing a number of variables by sorting them on a smaller number of factors

Examples include grouping variables like fear, worry, stress into a subset like anxiety.

Question 2

What are the two types of factor analysis?

Answer 2

• Exploratory factor analysis
• Confirmatory factor analysis

Exploratory analysis emerges from the data, while confirmatory tests a prior division.

Question 3

What is the main focus of the document regarding factor analysis?

Answer 3

Component analysis

It is simpler than other types of factor analysis.

Question 4

In component analysis, what is the starting point?

Answer 4

A set of scores from a sample of subjects on several variables

The factors are summarized into a smaller set of new variables.

Question 5

What is the term for variables that load highly on a particular factor?

Answer 5

High correlation with that factor

These variables are said to load highly on that factor.

Question 6

What does Multiple Group Method (MGM) aim to verify?

Answer 6

Whether a proposed grouping of the variables is supported by a data set

It sums scores of standardized variables thought to belong to a particular group.

Question 7

What does Principal Component Analysis (PCA) aim to find?

Answer 7

Factors that explain as much of the variance in the data as possible

It identifies the most significant factors affecting the data.

Question 8

What is the purpose of rotating the loading matrix in PCA?

Answer 8

To enhance interpretability of the loadings

Rotation can be orthogonal (uncorrelated) or oblique (correlated).

Question 9

What is Variance Accounted For (VAF)?

Answer 9

The proportion of explained variance

VAF increases by adding more factors to the analysis.

Question 10

What is the difference between item-test correlation and item-rest correlation?

Answer 10

• Item-test correlation: correlation with itself
• Item-rest correlation: correlation with the sum of remaining variables

Item-rest correlation is always lower than item-test correlation.

Question 11

What is the goal of methods like Varimax in PCA?

Answer 11

To create a simple pattern in the data

These methods help in making the rotation process more automatic and accurate.

Question 12

What is one purpose of test construction in relation to PCA?

Answer 12

Detection of deviant items

Items with extremely low variance should be inspected carefully.

Question 13

True or false: The measurement scale of combined items is less sensitive than that of an individual item.

Answer 13

FALSE

The combined items’ scale is more sensitive, reducing the influence of measurement error.