Sum Notes Flashcards
What is MMPI (Minessota Multiphasic Personality Inventory)?
Self-report inventory
Whats the purpose of MMPI?
Differentiate between normal and abnormal groups
By its validity indexes and criterion-keying approach
What inventories did MMPI use at first and what did it use later on?
It used content-based inventories at first, and then moved on to the criterion-keying using 10 different scales
Which are referred as ‘‘basic scales’’ in MMPI-2?
The validity indexes and the clinical scales
What are the 6 major categories of MMPI-2?
Validity indexes, clinical scales, content scales, supplementary scales, critical items and code types
What do high scores in one or more of the validity indexes mean?
It means that the actual scores in the clinical scales should not be trusted
L scale
Favorable light
F scale
Wanna appear as bad
K scale
'’Defensiveness scale’’, underreporting
Clinical scales
Are criterion-keyed scales
with 50 items per scale.
(Together with validity indexes=Basic scales)
Content scales
15 content scales per total
How were the content scales developed?
They were developed by a rational analysis of items
This analysis grouped the items together which measure the same construct and have a high item-total correlation
What are the 2 things that the analysis of the content scales do?
The analysis
1) groups together the items measuring the same construct
2) also groups together the items that have high item-total correlation
Critical items
Reports the responses with much significance (i feel like cutting my wrists)
Woodworth Personal Data Sheet
Content model, simple and straight-forward
Advantages and disadvantages of woodworth personal data sheet
Good face validity, bad because response sets occur
What is the difference between norm referenced test interpretation and criterion referenced test interpretation?
In criterion referenced: There is a prior criterion with which the scores will be interpreted
In norm-referenced: Scores will be interpreted depending on a norm
What 2 things are needed for interpreting a test?
Reliability coefficient and standard error of measurement
What is the SEM? Standard error of measurement?
Its the standard deviation of hypothetical infinite scores around a person’s true score
What is standard error of measurement directly related to?
Its directly related to reliability of a test.
What does high SEM mean?
The higher the SEM, the lower the test reliability
Related to the SEM, when is the test more reliabiable?
When the SEM is small. Small standard deviation
What if test reliability = 0?
The standard error of measurement will equal the standard deviation of the observed true scores
What if test reliability=1.00?
Then the SEM will be 0
Appropriate units and SEM
SEM has to be expressed in the scored units that are used for the interpretation
What happens if the interpretation of a test employs normed scores?
Then the SEM raw scores have to be converted
When do we use Spermans-Brown formula?
1) if we have a long test
2) if the length of the test has changed and we want to see if the reliability has increased
What is Spearmans-brown formula?
It is a measure to test reliability when the test length has been changed
What is the basic premise for Spearmans-Brown?
We see what happens to reliability (does it increase or decrease) when we change the test
What is k in the Spearmans-Brown formula?
Its the division between the number of items in the original test and the number of items in the new test
If original> 20
And new test>10
Then k=2
rkk in the spearmans brown
reliability of a test ‘‘k’’ times as long as the original test
r11 in the spearmans brown
reliability of original test (Cronbachs alpha)
WAIS (Weslchers Adult Intelligence Test)
Converts raw scores to standard scores
What is the score system of WAIS
M=10 S=3
How does an index score come in WAIS validity?
It comes from 2-3 subset scores and
is an outgrowth of factor analysis of the substests
What does general ability index measure in WAIS?
Its based on Verbal Comprehension and Perceptual Reasoning scores
Which statistical techniques does factor analysis include?
Principal component analysis, various rotation procedure, and stopping rules
What does factor analysis do?
It indentifies the common dimensios that underlie performance on many different measures
How does the factor analysis begin?
With raw data, with a correlation matrix
How are results from factor analysis presented and what does it show?
They are presented in a factor matrix and they show the loadings
Loadings= The weight that each original variable has on a new established factor
Why do we do the rotation thing in factor analysis? (Varimax rotation)
To aid the interpretation
What happens if there is a high correlation between A and B but not between A and C?
If A and B variables are highly correlated, we are considering them 1 dimension.
If A and C are not highly correlated (0.20) then we just say we have 2 dimensions (A,B & C)
What happens if there is a high correlation between A and B but not between A and C?
If A and B variables are highly correlated, we are considering them 1 dimension.
If A and C are not highly correlated (0.20) then we just say we have 2 underlying dimensions (A,B & C)
Which loadings are considered noteworthy?
Above 0.30
What is the difference between factor analysis and principal component analysis?
In factor analysis, we are already assuming that there are latent factors in our data. In principal component analysis we seek to identify variables that are composites of the observed variables
What is a loading matrix in PCA?
When we correlate every item with facotrs
How are factors that correlate with each other well called?
Oblique factors
How are the factors that dont correlate well with each other called?
Orthogonal
What happens to VAF in regards to factors?
If we make more factors, the VAF will increase
What is the difference between PCA and MGM?
In PCA, we find q number of factors that together explain the maximum amount of variance
How does PCA find the q number of factors that explain for the maximum amount of variance?
One by one. It finds one factor with maximum VAF. This is called the first principal component.
Then> computes residues
Then>find ideal weights for next factor,creates 2nd factor
What is the first principal component?
The factor with maximum VAF
Why do we use factor analysis?
To demonstrate internal structure validity
