Factor analysis Flashcards

1
Q

Why use factor analysis?

A
  • effective for investigating relationships for complex concepts
  • investigate concepts not easily measured
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Test dimensionality

A

how many dimensions does the test have?

are the dimensions correlated?

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

Unidimensional

A

reflect only one psychological dimension

dimensionally has implications for scoring, evaluation and use

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

Multidimensional tests with correlated dimensions

A
  • tests with higher order factors
  • items reflect more than one psychological attribute
  • correlated dimensions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Two types of factor analysis

A
  • exploratory factor analysis (EFA)

- confirmatory factor analysis (CFA)

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

Steps of EFA

A
  • choose extraction method
  • identify number of factors
  • factor rotation
  • examine factor loadings
  • examine inter-factor loadings (where approriate)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

PAF

A

principal axis factoring

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

PCA

A

principle components analysis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Elgenvalues

A

examine elgenvalue sizes, location has implications for number of dimensions
- egenvalue > 1.0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Factor rotation

A
  • Used with multidimensional scales

- to clarify psychological meaning of factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Two general types of factor analysis

A
  • orthogonal rotation (uncorrelated)

- oblique rotation (either correlated or uncorrelated)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

“Factor loading”

A
  • range between -1 and +1

- influenced by type of rotation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

orthogonal

A

loadings can be seen as correlations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

oblique

A

several kinds of loadings

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

pattern coefficients

A

unique associations - unique association between item and factor

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

structure coefficients

A

correlations between responses and total

17
Q

size of loadings

A

> .30 or .40 = strong

>.70 or .80 = very strong

18
Q

positive loading

A

high score on item can have high level of factor

19
Q

negative loading

A

high score on item have low level of factor

20
Q

Examining associations among factors

A
  • oblique rotations
  • correlation for each pair of factors
  • higher-order associations
21
Q

primary objectives of EFA

A

determine number of common factors and strength of relationships

22
Q

Confirmatory factor analysis (CFA)

A
  • confirmatory procedure
  • used when there are clear indication of test dimensionality
  • testing specific hypotheses about underlying dimensions
23
Q

Primary objective of CFA

A

determine the ability of a predefined factor model to fit an observed set of data

24
Q

Factors affecting responses to test items

A
  • respondent trait level
  • item difficulty
  • item discrimination
25
Q

IRT measurement models

A
  • one parameter logistic model
  • two parameter logistic model
  • three parameter model
26
Q

One parameter logistic model (Rasch model)

A
  • simplest IRT model
  • responses to binary items = determined by item difficulty
  • trade off between trait level and item difficulty
  • applies only to binary items
27
Q

Two parameter logistic model

A
  • more complex
  • includes two item parameters
  • responses to binary items (trait level, item difficulty, and item discrimination)
28
Q

Three parameter model

A
  • incorporates respondent guessing

- as trait increases as does chance of guessing correct

29
Q

Graded response model

A

Many tests include more than two response options.

  • specific order, or ranking of responses
  • allows us to estimate how well test questions measure latent trait
30
Q

Initial parameter estimates obtained through two-step process

A
  • determine proportion of items each respondent answered correctly
  • compare response probabilities to actual responses
31
Q

Model fit

A
  • ‘fit includes’ - reflect model compatibility with actual responses
  • suggest incompatibility - exercise caution
  • suggest good fit - proceed with interpretation
32
Q

Item and test information

A
  • IRT provides information about items and tests

- item characteristics are used to evaluate items and maximise test quality

33
Q

Item characteristic curves

A
  • reflects relationship between latent ability and performance on test items
  • each item has a curve
34
Q

Applications of IRT

A
  • test development and improvement
  • differential item functioning
  • person fit
  • computerised adaptive testing