StatisticalMethods MCQ's E.F.A

?
Describe the purpose of Exploratory Factor Analysis...
To explain patterns of correlations in a smaller number of variables
1 of 37
Describe the purpose of Factor Analysis....
To explain the extent with which measured variables and their co-variability can be explained by a smaller number of variables
2 of 37
What is a 'Latent Variable' in the context of factor analysis?
A variable that cannot be directly measured
3 of 37
E.F.A aims to....
Understand the structure of a set of variables, to find a 'simple structure, thereby to identify relatively independent clusters of variables
4 of 37
What are the basic stages of the E.F.A analysis process?
Pre-Analysis Checks > Extraction > Rotation > Factor Naming
5 of 37
The ratio of participants to items in E.F.A should be....
5:1 or 10:1
6 of 37
The absolute minimum amount of participants required for E.F.A is...
100
7 of 37
The ratio of items to factors in E.F.A should be....
4:1
8 of 37
The ratio of participants to factors in E.F.A should be....
6:1
9 of 37
What x2 statistical checks should we undertake regarding our data and it's suitability for an E.F.A?
The KMO and R-Matrix or Identity Matrix
10 of 37
In our pre-analysis checks of E.F.A, the KMO score should be less than what?
0.5
11 of 37
Variability in E.F.A can be split into:
Unique Variability (specific to that variable), Common Variability (shared with other variables) and Error Variability (random variability)
12 of 37
Communality in E.F.A means...
The proportion of variance explained by the extracted factors - a measure of common variance
13 of 37
In E.F.A, if all communalities are more than 0.6, then the sample size must be...
More than 100
14 of 37
In E.F.A, if commonalities are less than or equal to 0.5, and there are only a few factors, then the sample size must be
Within 100 or 200
15 of 37
In E.F.A, if commonalities are much smaller than 0.5 then the sample size must be more than...
500
16 of 37
An eigenvalue represents what?
The variance explained by a factor
17 of 37
The K1 rule suggests what when extracting factors?
Select all factors with eigenvalues greater than 1
18 of 37
How does a scree plot work when extracting factors?
It plots eigenvalues against components, and provides a point of inflexion - any factor to the left of this is extracted
19 of 37
A researcher conducts a parallel analysis to aid in extracting factors from her E.F.A. Which factors should she extract?
Any factors that have observed eigenvalues greater than the randomly generated ones
20 of 37
Orthogonal rotation assumes what?
Factors are not correlated
21 of 37
Oblique rotation assumes what?
Factors are correlated
22 of 37
Give an example of a method of orthogonal rotation
Varimax
23 of 37
Gve an example of a method of oblique rotation
Direct Oblimin
24 of 37
What is 'cross-loading'?
Cross-loading is where eigenvalues are seen in more than one component
25 of 37
A psychologist finds that some of his eigenvalues have a difference of more than 0.3 across components. This suggests what?
Cross-loading
26 of 37
A psychologist finds that some of his eigenvalues have a difference of less than 0.2 across the same component. This is an example of what?
Nothing to worry about - no cross-loading
27 of 37
Define internal reliability.
Internal reliability relates to whether a measure is consistent within itself
28 of 37
What does the split half test do to examine the internal reliability of an E.F.A?
It splits the scale into x2 randomly selected sets of items. A score for each participant is then calculated on each half of the scale.
29 of 37
What is the problem with using the split half test to measure the internal reliability of an E.F.A?
There are numerous different methods to split the data
30 of 37
A researcher finds after conducting a split half test that their scores for each participant are the same or similar. This means
Their analysis has a high level of internal reliability
31 of 37
How does Cronbach's Alpha work?
It calculates two sets of items in every way possible and calculates the coefficient for each split
32 of 37
A psychologist finds that her Cronbach's Alpha score when measuring the internal reliability of her E.F.A comes out as 0.12. What does this mean?
This is a cause of concern - Cronbach's Alpha scores should be between 0.7 and 0.8
33 of 37
When utilising an E.F.A for dichotomous scores, the best method to assess internal reliability is what?
KR-20
34 of 37
A student finds that their internal reliability, according to a KR-20 test, comes out as 0.2. This means what?
There analysis has a high level of internal reliability - scores should be less than 0.9
35 of 37
Define external reliability.
External reliability asks whether a measure varies from one use to another.
36 of 37
What are some limitations of E.F.A?
Confirmatory Factor Analysis is now more commonly used, infinite rotations for the same variance are possible, GIGO
37 of 37

Other cards in this set

Card 2

Front

Describe the purpose of Factor Analysis....

Back

To explain the extent with which measured variables and their co-variability can be explained by a smaller number of variables

Card 3

Front

What is a 'Latent Variable' in the context of factor analysis?

Back

Preview of the front of card 3

Card 4

Front

E.F.A aims to....

Back

Preview of the front of card 4

Card 5

Front

What are the basic stages of the E.F.A analysis process?

Back

Preview of the front of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Visual System resources »