2. Factor analysis

Question 1

Why is FA necessary? (FA Methods)

Answer 1

Identifies an underlying structure (factors) from a large set of correlated variables

Question 2

What is a factor? (FA Methods)

Answer 2

A cluster of items that all measure the same idea

Question 3

What assumptions are required for FA? (FA Methods)

Answer 3

• Normally distributed data
• Worst offenders
• Correlation of items

Question 4

How do you identify for normally distributed data? (FA Methods)

Answer 4

SD between .5 and 1.5

Question 5

How do you identify the worst offenders in a data set? (FA Methods)

Answer 5

z-scores

Question 6

What is expected of correlation of items? (FA Methods)

Answer 6

• Coefficients more than .3 & less than .9

- Determinant to be above .00001

Question 7

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

Answer 7

• Sampling adequacy
• Values range from 0 (not adequate) to 1 (very adequate)
• If below .5, rewrite questionnaire

Question 8

What does a Bartletts test show? (FA Methods)

Answer 8

• If correlations are too small for FA

- P is significant, then FA is appropriate

Question 9

How do you report Kaiser-Mayer-Olkin results? (FA Methods)

Answer 9

KMO test is …

Question 10

How do you report Bartlett’s test? (FA Methods)

Answer 10

Bartletts test is non/significant [x²(df)=…, p=…]

Question 11

What is factor extraction? (FA Methods)

Answer 11

Deciding on how many factors best capture the data

Question 12

What is an eigenvalue? (FA Methods)

Answer 12

• The variance in all the variables accounted for by a particular factor
• Low values do not explain the data and should be removed

Question 13

What are the two rules for using an eigenvalue? (FA Methods)

Answer 13

1) When variables < 30 and all commonalities are > .7

2) When P’s > 250 and average commonality is ≥ .6

Question 14

What is a commonality? (FA Methods)

Answer 14

The percent of variance in a variable explained by all of the factors together (values after extraction)

Question 15

What should happen if the two criteria for eigenvalues is not met? (FA Methods)

Answer 15

Use a scree plot

Question 16

When using a scree plot, what are you looking for? (FA Methods)

Answer 16

• Inflection point (where the slope changes)

- Keep everything to the left of this point

Question 17

What does rotation do? (FA Methods)

Answer 17

Optimises how the items load onto a factor (spreads variance evenly along factors)

Question 18

What are the two types of rotation and when are they used? (FA Methods)

Answer 18

• Orthogonal (uncorrelated factors)

- Oblique (inter-correlated factors)