Factor Analysis Flashcards

1
Q

What is a factor analysis?

A
  • Identify psychological constructs from a scale
    Are they driven by the same underlying factor?
  • Reduce the number of variables in a data set
    Representations of themes or constructs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the different types of factor analysis?

A
  • Principal component analysis (focus)
    New measurement
  • Exploratory factor analysis
    Questions that tap into the same factor or factors, ask if they have the same underlying concept
  • Confirmatory factor analysis
    Adapting to new context or new language
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is needed to do before factor analysis can begin?

A
  • Check correlation
    Needs to be reasonable between items
  • Multicollinearity
    If any item is over r=+.90
    -Singularity
    Perfect correlation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What tests need to be included in factor analysis?

A
  • Kaiser-Meyer-Olkin
    Compact association, distinct and reliable factor
    Should be .50, pref over .80
    How well suited the items are for factor analysis
  • Bartlett’s test of sphericity
    Do we have good correlation among scale items?
    Should be significant= decent correlation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Factor Analysis - SPSS output

A

Table 1
- Correlation between scale items
Table 2
- KMO and Bartlett’s Test
Table 3
- Information about proportion of common variance in a variable
- 0 = has a unique variance
- 1 = share variance with others
- Extraction= to make conclusions, % overlap with overall variation of the other items
Table 4
- Decide how many factors our scale is composed of (extraction)
- Eigenvalue = variation explained by a factor
- Retain all factors greater than 1
Table 5
- Scree plot; how many factors we should retain
- Find point inflextion, drastic change in slope
- Retain factors on the left side of point inflextion
Table 7
- Info of which items makes up a factor
- Conventional cut off for factor loadings .30
Table 8
- Info of correlation between factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Factor Analysis - APA style

A
  • What we performed
  • How many items
  • How many factors
    Eigenvalues
  • Explained variation from factors
  • How many items in each factor, their loading range
  • Inter-item reliability for each factor
  • Conclusion
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is item-scale correlations?

A

How different each scale is to the factor

How well did you know this?
1
Not at all
2
3
4
5
Perfectly