Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Psychometrics > Final Exam > Flashcards

Final Exam Flashcards

(116 cards)

Start Studying
1
Q

A depression test has been shown to have strong association with current levels of anxiety. This is an example of which form of validity?

A

Concurrent Validity

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

Correlation coefficients range from ____ to _____

A

-1 to +1

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

The use of meta-analysis to assess evidence of a test’s adequacy and appropriateness for use in multiple situations and settings is call a _____ study

A

Validity Generalization

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

The ____ the error, the _____ the reliability

A

Higher, lower

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

The variance between two measurements is attributable to the method used for measurement is:

A

Shared method variance

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

Ratio Scales have a _____ while Interval Scales do not

A

True Zero

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Estimates of reliability such as the alternate forms method and the split-half method require that the tests be ____

A

Parallel

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

A test in which all items are keyed in the same direction is most vulnerable to _____

A

Acquiescence bias

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

A test that compares a test-taker to a reference sample is what kind of test?

A

Norm-referenced

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

A depression test has been shown to predict future life satisfaction in multiple research studies. This is an example of what kind of validity?

A

Predictive validity

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

According to _____ validity, the structure of a test should match the theory behind that construct

A

Structural validity

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

______ reliability is most appropriate to assess reliability in a test that that measures traits that are not expected to change from one testing to the next

A

Test-retest reliability

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

Other factors that are irrelevant to the construct and are affecting the results of it

A

Construct irrelevant variance AKA construct contamination

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

Universal design in test construction refers to facilitating _____ for all test takers in the population

A

Accessibility

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

This type of validity gives you the ability to make future predictions from the resulting measurement

A

Criterion Validity

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

This type of validity ensure that you are adequately measuring the construct you intend to measure

A

Construct validity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

A range of values containing the true score

A

Confidence interval

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

Correlation values are _____ whereas covariance values are not

A

Standardized

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Data that consists of categories

A

Nominal Data

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

Data that can be placed in a specific order

A

Ordinal Data

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

Rewording items to be more neutral and providing a distraction-free testing environment can help reduce _____ bias

A

Social desirability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Correlation between two measures that is consistent with theory/expectations (a form of associative validity)

A

Convergent validity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

The three measures of central tendency

A

Mean, median, mode

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Agreement across observers or coders shows ______

A

Inter-rater reliability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Findings that can be generalized beyond the study and sample have _______ validity
External validity
26
The process of quantifying variables for the purpose of measuring their occurrence, strength, and frequency
Operationalization
27
Fairness in testing is an issue of (reliability or validity?)
Validity
28
A number between -1 and +1 representing the linear association between two variables
Correlation coefficient
29
The majority of constructs psychologists study have a (relative or absolute) zero
Relative Zero
30
The degree to which individual scores remain consistent over administrations of the same test (or alternate versions of the test)
Reliability
31
Concurrent Validity
Theory-consistent correlations at the SAME testing
32
Difference between ratio scales and interval scales
ratio has true zero value while interval does not
33
Criterion-referenced test
uses a cutoff score to sort people into groups
34
Universal Design
intentional about how the construct is operationalized, taken, and measured so it’s accessible to the most amount of people
35
Social Desirability Bias
changing responses in order to appear more socially desirable
36
Inter-rater reliability
same reliability across raters
37
Construct underrepresentation AKA construct deficiency
A test does not fully measure a construct; missing important pieces
38
Construct irrelevant variance AKA construct contamination
A test includes irrelevant factors in the items
39
Structural Validity
Test structure should match the theory
40
Factor analysis
uses statistics to identify clusters
41
Unidimensional
All items correlate (ex: measure of depression)
42
Multidimensional
All items do not correlate (ex: measure of bipolar)
43
Response process
Match between the intended process and the process respondents use when completing the measure
44
Predictive Validity
Theory-consistent correlations at a FUTURE testing
45
Convergent Validity
A construct's correlation with other constructs
46
Discriminant Validity
A construct's lack of correlation with other constructs
47
Consequential Validity
Correlation between the intended consequences of the test use, and actual consequences of the test use
48
Correlation coefficient
an estimate of association/consistency between constructs OR parts of a test
49
Three ways to evaluate the correlation between two variables
Pearson Correlation coefficient, Spearman's rho, Kendall tau
50
Used for measuring internal consistency
Chronbach's alpha
51
Alternate forms method
estimate reliability based on consistency of scores across two versions of a test
52
Test-retest
estimate reliability based on consistency of scores across two separate testings
53
Used for estimating agreement between two or more measures
Intraclass correlations coefficient
54
Content validity
How well a test measures a representative sample of subject matter being investigated
55
Criterion validity
How well a test correlates with an established standard of comparison
56
Three types of criterion validity
Predictive, concurrent, and retrospective
57
Construct validity
How well a test measures what it intends to measure
58
Convergent validity
How well responses on a test relate to responses on a similar test
59
Two aspects of construct validity
Convergent validity and discriminant validity
60
Discriminant validity
The degree to which a test diverges from another test that is conceptually unrelated
61
Reflects how well an assessment instrument predicts an indicator of a given concept
validity coefficient
62
Multitrait-Multimethod Matrix (MTMM)
Shows correlations among two or more measurement techniques
63
Four types of measurement scales
Nominal, ordinal, interval, ratio
64
Continuous variables
has an infinite number of possibles values
65
Discrete variable
limited number of possible values
66
Bounded variables
measurement scales with a mathematical boundary
67
Nominal
used for qualitative variables
68
Ordinal
rank-order quantitative variables
69
Interval
consistent intervals w/o true zero
70
Ratio
consistent intervals w/ true zero
71
Mean
average
72
Median
Middle number
73
Mode
Most frequently occuring
74
Range
Span
75
Reliability
consistency/stability of test scores over time
76
Classical Test Theory
Observed score = true score + error
77
Test-Retest Reliability
Estimates reliability based on consistency of scores across two separate testings
78
Alternate forms reliability
Both forms of a test (1) measure the same true amount of a construct and (2) have equal error variance
79
Conceptual relation between reliability and validity
A test must be reliable in order to be valid, but it does not need to be valid in order to be reliable
80
Parallel tests
Different forms of a test; used to control for memory/practice effects
81
Internal Consistency Method
one test at one point in time; each part is treated as a different form
82
Split-half estimates
– Comparing results from one half of the test to results of the other half
83
Cronbach’s alpha
– The most widely used method for estimating reliability in psychology Each item on a test is treated as it’s own test
84
Validity
– interpretation and intended use of test scores
85
Content Validity
– How well a test measures a representative sample of a subject matter being investigated
86
Validity generalization
– A test’s adequacy and appropriateness for use in multiple situations and settings
87
Construct underrepresentation (aka construct deficiency)
– A test does not fully measure a construct; important pieces are missing
88
Construct irrelevant variance (aka construct contamination)
– A test includes irrelevant factors in the items
89
Structural Validity
– Test structure should match the theory
90
Response Process
– Match between the intended process and the process respondents use when completing the measure
91
Multi-trait-Multimethod Matrix
– shows correlations among two or more measurement techniques
92
Concurrent validity
– theory-consistent correlations at the same testing
93
Predictive validity
– theory-consistent correlations at a future testing
94
Discriminant Validity
– a construct’s lack of correlations with other constructs
95
Convergent Validity
– a constructs’ correlation with other constructs
96
Consequential validity
– correlation between the intended consequences and actual consequences of the test use (pros/cons balance)
97
Relation between bias and validity
more bias = less validity
98
Malingering
– “faking bad”; appearing more impaired, distressed, challenged, or disturbed
99
Extremity bias
– picking answers in the extreme rather than the middle
100
Acquiescence Bias
– agreeing/disagreeing without considering meaning Especially problematic when all items are keyed in the same direction May artificially inflate/deflate correlations
101
Social Desirability Bias
– changing responses in order to appear more socially desirable May artificially inflate correlations
102
Prevention-oriented Strategies (Managing Test Context)
Offering anonymity Limiting the demands of testing Leading participants to believe bias can be detected
103
Detection- and Intervention-oriented Strategies (Managing Test Content)
Items/scales embedded in the measure | Useful for desirability tests, extremity tests, acquiescence tests
104
Prevention-oriented and Effects-oriented Strategies (Specialized Tests)
``` Keep it simple Frame items neutrally Use forced-choice formats Introduce random/unsystematic measurement Use balanced scales Introduce guessing penalties ```
105
Construct Bias (AKA Measurement Bias)
– Teat may have different meaning for different groups
106
Predictive Bias
– Test may have a different predictive value for different groups
107
Reliability analysis
– internal consistency coefficients
108
Rank order
– item difficulty ranking
109
Item discrimination index
– item-level discrimination
110
Factor analysis
– internal structure of the test
111
Differential item functioning analysis
– probability of answering a certain way based on trait levels
112
Universal design
– test is operationalized, taken, and measures to it is accessible to the most amount of people
113
Norm-referenced tests
– compares a test taker to a reference sample
114
Criterion-referenced tests
– uses a cutoff score to sort test takers into groups
115
Correlation vs. Covariance
– correlation values are standardized, covariance values are not
116
Trait vs. state variables
– state variables change very quickly, while trait variables are more stable over time

Psychometrics (1 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions