Chapter_12_Factor Analysis, Path Analysis, and Structural Equation Modeling

Question 1

factor analysis

Answer 1

• examine the relationships among a set of INTERCORRELATED variables
• one or more underlying DIMENSIONS or factors
• each item reflects an ASPECT of the construct being measured

Question 2

Exploratory factor analysis

Answer 2

• identify subsets of those variables
• uncorrelated with the variables in the other subsets

Question 3

Uses of Exploratory Factor Analysis

Answer 3

1. Data Reduction
2. Scale Development

Question 4

Data Reduction

Answer 4

Principal components analysis
- be easier to understand
- empirical

Question 5

Scale Development

Answer 5

a scale should represent only one construct or be composed of subscales
- the items intercorrelate in the way they theoretically should
- examine the STRUCTURAL validity of a measure

Question 6

Considerations in EFA

Answer 6

1. Number of Research PARTICIPANTS
   - 200 to 400
2. Quality of the DATA
3. Factor EXTRACTION and Rotation
4. Number of FACTORS
5. Interpreting the Factors: factor LOADINGS
6. Factor SCORES

Question 7

Quality of the Data

Answer 7

• items are representative
• items are relevant

Question 8

extraction

Answer 8

determine the number of factors underlying a set of correlations
- Factors are extracted in order of importance
- first factor accounts for the most variance

Question 9

rotation

Answer 9

clarify the factors once they are extracted
- simplifies the results of a factor analysis by minimizing these multiple loadings
1. Orthogonal
- forces factors to be uncorrelated with one another
2. Oblique
- allows factors to be correlated

Question 10

Number of Factors

Answer 10

1. eigenvalues
   - the percentage of variance in the variables being analyzed that can be accounted for by that factor
2. scree plot
   - plotting the eigenvalue of each factor against its order of extraction
3. parallel analysis
   - create a random data set with the same number of observations and variables

Question 11

factor loadings

Answer 11

the correlation of each item with its underlying factor

Question 12

Factor Scores

Answer 12

combined Z scores for each factor

Question 13

Confirmatory factor analysis

Answer 13

researchers propose hypotheses about the dimensions a set of items will load on

Question 14

2 Purposes of CFA

Answer 14

1. hypothesis testing
2. measure validation

Question 15

Hypothesis Testing

Answer 15

different patterns of relationship among the variables ­encompassed by those theories

Question 16

Measure Validation

Answer 16

1. Structural Validity
2. Generalizability

Question 17

Structural Validity

Answer 17

the dimensionality of a measure reflects the dimensionality of the construct it measures

Question 18

Generalizability

Answer 18

a measure provides similar results across time, research settings, and populations
- differential validity
- testing the invariance

Question 19

Evaluating Goodness-of-Fit

Answer 19

tests a hypothesized factor structure against the structure that exists in a data set

Question 20

Testing Mediational Hypotheses

Answer 20

The Causal Steps Strategy
multiple regression analyses
1. test a and b
2. test ab

Question 21

Path analysis

Answer 21

sets of multiple regression analyses to estimate the strength of the relationship between an independent variable and a dependent variable controlling for the hypothesized mediating variables

Question 22

Structural Equation Modeling

Answer 22

combines path analysis with confirmatory factor analysis

Question 23

Limitations on Interpretation

Answer 23

1. Causality
2. Completeness of the Model
3. Alternative Models