Lecture 4 Flashcards

1
Q

What is the key difference between principal components and factor analysis?

A
  • PCA: finds optimal linear transformations
  • FA: assumes latent factors that are not directly oberved
  • there is no model in PCA, but there is a model (can test fit) in EFA
  • PCA is simply a weighted sum of variables
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How does PCA work?

A
  • graphically, finds new axes for your data

- new components are chosen one by one, to maximise variance not yet accounted for

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

How many components can you make with N variables?

A

N components.
BUT if you use less than N, then there are a smaller no. of components, then there is freedom in the final solution
- also if you use less than N, you can rotate to get a simplet solution

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

Why is PCA simple?

A
  • they are not correlated, even if the original variables are
  • first component explains the most variance > thus you know which components are the most important
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How do you determine how many components/factors to extract?

A
  • SPSS default is no. of eigenvalues > 1 (DO NOT USE), called Kaiser-Guttman
  • use Screen plot (where it turns)
  • use parallel
  • use MAP
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Explain the parallel test

A
  • uses random data (with same dimensions as your dataset) as a baseline
  • if eigenvalue is higher than random (noise) data, then it must be signal
  • where “raw data”
  • the “pcntile” is the 95th percentile
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Explain the MAP test

A
  • plots squared partial correlations and gets MINIMUM
  • as more components are extacted, more are partialled out of correlation matrix, SPCs approach 0
  • but then at some point ‘noise’ components get partialled out, and the SPCs increase again
  • therefore, want the minimum
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What does a -ve or high component/factor loading mean?

A
  • negative: you get a high score on that item, you get a low score on the component/factor
  • negative loading similar to reverse scoring
  • high: higher score on that item, higher score on factor/component
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Why rotate components?

A
  • simpler structure

- easier to interpret

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

What are the 2 types of rotation?

A
  • orthogonal: remain uncorrelated

- oblique: correlated

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

What are the specific SPSS rotations?

A
  • orthogonal: varimax, equamax, quartimax

- oblique: direct oblimin, promax

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

What do you interpret after rotation?

A
  • oblique: pattern matrix
  • factor correlations
    (structure matrix = product of pattern and factor correlation matrix)
  • orthogonal: rotated
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What does EFA assume?

A

that there are some underlying latent factor that cannot be directly observed > searches for these

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

What is ui? What is k?

A
  • u: the specific factor (noise/error)

- k: the common factor

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

What are the assumptions of EFA?

A
  • common factors standardised (variance = 1)
  • common factors uncorrelated
  • specific factors uncorrelated
  • common factors uncorrelated with specific factors
  • multivariate normality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the underlying rationale of EFA?

A
  • partial correlations
  • correlation b/w item 1 and item 2, WHEN HOLDING CONSTANT a latent variable is…
  • if PC is 0, then correlation b/w the items is fully explained by the factor > want it as close to 0 as possible
  • aim to find a latent variable that accounts for observed correlation (i.e. make it as close to 0 as possible)
  • if we can find these correlations/mimic the covariance matrix, then we have found the latent factors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the communality?

A
  • the variance due to the common factors

- want HIGH communalities

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What are the rules/guidelines about sample size for EFA? What is the problem will small sample size?

A
  • 150+
  • absolute sample size + communalities are more important
  • ratio > variables:sample size NOT important
  • if loadings are high, then you can have a lower sample size
  • less generalisable if too small
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are the 3 things you want for EFA?

A
  • high communalities (>.8 ideal, but reality is .4-.7) > can drop things if they have low communality (but be careful)
  • few cross-loadings (>.032)
  • more than 3 strongly loaded items per factor

^^^ need a larger sample if these are not met

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is the issue with high communalities? How do you fix this?

A
  • you only know them after you find the factor loadings

- so… use prior diagnostics!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What are the prior diagnostics?

A
  • correlations (low - low loadings)
  • Bartlett: want >.05 (usually always is)
  • anti-image: diagonal (MSA) close to 1, off-diagonal (anti-image correlations) close to 0
  • Kaiser: want high, >.9 great
22
Q

Why is ML good?

A

has a goodness of fit test

23
Q

What is the issue with chi-square?

A
  • very sensitive test!! (wan >.05)

- use RMSEA instead. Want less than 0.06

24
Q

What are Heywood cases? How do you find and fix them?

A
  • technical problems
  • values of .999
  • look for in un-rotated factor matrix
  • problem > maybe too many factors extracted
  • increase interations from 25 to 250
25
Q

What are the 3 ways of estimating factor scores using congeneric tests?

A
  • regression model
  • Bartlett (probs best)
  • Anderson-Rubin

AR assumes uncorrelated, so don’t use with oblique solutions

26
Q

What is an eigenvalue?

A
  • the variance of the first component extracted (variance of Y1)
  • derived from correlation matrix of variables (NOT covariance, variables are standardised before analysis)
27
Q

Is PCA statistics? Why is this good?

A

No, just a mathematical technique

  • no error terms, no prediction
  • there is no model!
  • this is why there are no assumptions or requirements, it always works
28
Q

Is PCA a type of EFA?

A

Nope

29
Q

What is plotted in a scree plot?

A

eigenvalues vs. components

30
Q

What do you do if parallel and MAP tests disagree?

A

make a decision!
can cite someone who says one is better
or choose the more interpretable one

31
Q

How can you write out the component loadings to equal the component? In matrix form?

A
  • Y1 (component 1) = loadingXitem + loadingXitem….. etc.

- Y = aX

32
Q

Why should you only use oblique rotation?

A
  • more realistic

- more statistically sound

33
Q

Which 2 factor methods are recommended by Schmitt?

A
  • maximum likelihood

- principal axis factoring

34
Q

How do you calculate RMSEA?

A

square root: (X2 - df) / ( (N-1)df)

if df > X2, then treat as zero! (amazing fit)

35
Q

What are the 4 key components of factor scores created by sum?

A
  • assumes equal weight of each item (tau-equivalent)
  • underpins test theory for reliability
  • basis for coefficient alpha reliability
  • if not true: alpha a serious underestimate
36
Q

What are congeneric tests?

A

assumption of varying factor loadings

37
Q

What happens is a factor/component or an item is added in PCA vs. EFA?

A
  • PCA: adding item may change component; adding component will not change loadings
  • EFA: adding item should not change others; adding factor will change factor loadings
38
Q

Why are there differences in PCA vs. EFA in terms of adding/removing items/factors? What does each method aim to do?

A

Issues with diagonal elements of correlation matrix

  • PCA: value of 1 used
  • aim to explain all variance of variable
  • reproduce whole variance-covariance matrix
  • EFA: diagonal is the communality
  • aim to explain only common variance of an element
  • reproduce only off-diagonal parts of variance-covariance matrix
39
Q

Widaman’s conclusion

A
  • rarely, or never, do a component analysis of empirical data if your goal is to interpret patterns of observed covariations among variables as arising from latent variables or factors
40
Q

What do PCA and EFA actually do?

A
  • use associations among variables to condense into a smaller, simpler number of variables
41
Q

What is the trade-off in PCA? What do we do to help this?

A
  • trade-off b/w getting a simpler structure (less components) and explaining a higher proporiton of variance
  • scree, MAP and parallel help us decide this trade-off?
42
Q

What do quartimax, varimax and equamax actually do? And oblimin and promax?

A
  • quartimax: simplifies variable pattern of loadings
  • varimax: simplifies the factor patterns of loading
  • equamax: compromise of above 2
  • oblimin: change delta -0.8 to 0.8
  • promax: change kappa, from 1 upwards
43
Q

What is important for factor in when you are deciding what to call your factors/component?

A
  • the direction (+ve or -ve) of items
44
Q

What are the anti-image and image correlations? What is image analysis?

A
  • image: correlations due to common parts
  • anti-image: correlations due to unique parts
  • image analysis: partitioning variance of observed variable into common and unique parts
45
Q

Why would you ever use PCA instead of EFA?

A
  • historically, PCA was simpler and faster

- PCA can be a fallback if you have a smaller sample or other technical issues that mean you cannot do EFA

46
Q

Do PCA and EFA have similar outputs?

A
  • yes
  • but not for all datasets
  • Widaman: only if there are high communalities
47
Q

What is the common factor model?

A
  • k common factors that explain observations on the variables
  • Xi = (labmda)i1F1 + (labmda)i2F2 … +(labmda)ikFk + ui
  • u is the specific factor
48
Q

How do you calculate the variance of observed variable X?

A
  • sum of factor loadings^2 (common variance) and variance of u (unique variance)
  • common variance = communality! (denoted h2)
  • covariance = multiply factor loadings
49
Q

What happens when you sum the squared coefficients (loadings) in PCA?

A

it always equals 1

- because this is the communality and PCA assumes communalities of 1!!

50
Q

What is the key difference between PCA/EFA and cluster analysis/MDS?

A
  • FA uses correlations as associations

- cluster and MDS use proximities (distances)

51
Q

What are the variables and components/factors in PCA v. EFA?

A
  • PCA: components are just weighted sums, so are also observed variables
  • EFA: factors are superordinate to observed variables (cause correlations b/w variables)