Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Psychometrics > Final Exam > Flashcards

Final Exam Flashcards

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
Q

Findings that can be generalized beyond the study and sample have _______ validity

A

External validity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

The process of quantifying variables for the purpose of measuring their occurrence, strength, and frequency

A

Operationalization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

Fairness in testing is an issue of (reliability or validity?)

A

Validity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

A number between -1 and +1 representing the linear association between two variables

A

Correlation coefficient

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

The majority of constructs psychologists study have a (relative or absolute) zero

A

Relative Zero

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

The degree to which individual scores remain consistent over administrations of the same test (or alternate versions of the test)

A

Reliability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

Concurrent Validity

A

Theory-consistent correlations at the SAME testing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

Difference between ratio scales and interval scales

A

ratio has true zero value while interval does not

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

Criterion-referenced test

A

uses a cutoff score to sort people into groups

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

Universal Design

A

intentional about how the construct is operationalized, taken, and measured so it’s accessible to the most amount of people

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

Social Desirability Bias

A

changing responses in order to appear more socially desirable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

Inter-rater reliability

A

same reliability across raters

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

Construct underrepresentation AKA construct deficiency

A

A test does not fully measure a construct; missing important pieces

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

Construct irrelevant variance AKA construct contamination

A

A test includes irrelevant factors in the items

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

Structural Validity

A

Test structure should match the theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

Factor analysis

A

uses statistics to identify clusters

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

Unidimensional

A

All items correlate (ex: measure of depression)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

Multidimensional

A

All items do not correlate (ex: measure of bipolar)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

Response process

A

Match between the intended process and the process respondents use when completing the measure

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

Predictive Validity

A

Theory-consistent correlations at a FUTURE testing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

Convergent Validity

A

A construct’s correlation with other constructs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

Discriminant Validity

A

A construct’s lack of correlation with other constructs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q

Consequential Validity

A

Correlation between the intended consequences of the test use, and actual consequences of the test use

48
Q

Correlation coefficient

A

an estimate of association/consistency between constructs OR parts of a test

49
Q

Three ways to evaluate the correlation between two variables

A

Pearson Correlation coefficient, Spearman’s rho, Kendall tau

50
Q

Used for measuring internal consistency

A

Chronbach’s alpha

51
Q

Alternate forms method

A

estimate reliability based on consistency of scores across two versions of a test

52
Q

Test-retest

A

estimate reliability based on consistency of scores across two separate testings

53
Q

Used for estimating agreement between two or more measures

A

Intraclass correlations coefficient

54
Q

Content validity

A

How well a test measures a representative sample of subject matter being investigated

55
Q

Criterion validity

A

How well a test correlates with an established standard of comparison

56
Q

Three types of criterion validity

A

Predictive, concurrent, and retrospective

57
Q

Construct validity

A

How well a test measures what it intends to measure

58
Q

Convergent validity

A

How well responses on a test relate to responses on a similar test

59
Q

Two aspects of construct validity

A

Convergent validity and discriminant validity

60
Q

Discriminant validity

A

The degree to which a test diverges from another test that is conceptually unrelated

61
Q

Reflects how well an assessment instrument predicts an indicator of a given concept

A

validity coefficient

62
Q

Multitrait-Multimethod Matrix (MTMM)

A

Shows correlations among two or more measurement techniques

63
Q

Four types of measurement scales

A

Nominal, ordinal, interval, ratio

64
Q

Continuous variables

A

has an infinite number of possibles values

65
Q

Discrete variable

A

limited number of possible values

66
Q

Bounded variables

A

measurement scales with a mathematical boundary

67
Q

Nominal

A

used for qualitative variables

68
Q

Ordinal

A

rank-order quantitative variables

69
Q

Interval

A

consistent intervals w/o true zero

70
Q

Ratio

A

consistent intervals w/ true zero

71
Q

Mean

72
Q

Median

A

Middle number

73
Q

Mode

A

Most frequently occuring

74
Q

Range

75
Q

Reliability

A

consistency/stability of test scores over time

76
Q

Classical Test Theory

A

Observed score = true score + error

77
Q

Test-Retest Reliability

A

Estimates reliability based on consistency of scores across two separate testings

78
Q

Alternate forms reliability

A

Both forms of a test (1) measure the same true amount of a construct and (2) have equal error variance

79
Q

Conceptual relation between reliability and validity

A

A test must be reliable in order to be valid, but it does not need to be valid in order to be reliable

80
Q

Parallel tests

A

Different forms of a test; used to control for memory/practice effects

81
Q

Internal Consistency Method

A

one test at one point in time; each part is treated as a different form

82
Q

Split-half estimates

A

– Comparing results from one half of the test to results of the other half

83
Q

Cronbach’s alpha

A

– The most widely used method for estimating reliability in psychology
Each item on a test is treated as it’s own test

84
Q

Validity

A

– interpretation and intended use of test scores

85
Q

Content Validity

A

– How well a test measures a representative sample of a subject matter being investigated

86
Q

Validity generalization

A

– A test’s adequacy and appropriateness for use in multiple situations and settings

87
Q

Construct underrepresentation (aka construct deficiency)

A

– A test does not fully measure a construct; important pieces are missing

88
Q

Construct irrelevant variance (aka construct contamination)

A

– A test includes irrelevant factors in the items

89
Q

Structural Validity

A

– Test structure should match the theory

90
Q

Response Process

A

– Match between the intended process and the process respondents use when completing the measure

91
Q

Multi-trait-Multimethod Matrix

A

– shows correlations among two or more measurement techniques

92
Q

Concurrent validity

A

– theory-consistent correlations at the same testing

93
Q

Predictive validity

A

– theory-consistent correlations at a future testing

94
Q

Discriminant Validity

A

– a construct’s lack of correlations with other constructs

95
Q

Convergent Validity

A

– a constructs’ correlation with other constructs

96
Q

Consequential validity

A

– correlation between the intended consequences and actual consequences of the test use (pros/cons balance)

97
Q

Relation between bias and validity

A

more bias = less validity

98
Q

Malingering

A

– “faking bad”; appearing more impaired, distressed, challenged, or disturbed

99
Q

Extremity bias

A

– picking answers in the extreme rather than the middle

100
Q

Acquiescence Bias

A

– agreeing/disagreeing without considering meaning
Especially problematic when all items are keyed in the same direction
May artificially inflate/deflate correlations

101
Q

Social Desirability Bias

A

– changing responses in order to appear more socially desirable
May artificially inflate correlations

102
Q

Prevention-oriented Strategies (Managing Test Context)

A

Offering anonymity
Limiting the demands of testing
Leading participants to believe bias can be detected

103
Q

Detection- and Intervention-oriented Strategies (Managing Test Content)

A

Items/scales embedded in the measure

Useful for desirability tests, extremity tests, acquiescence tests

104
Q

Prevention-oriented and Effects-oriented Strategies (Specialized Tests)

A 
Keep it simple
Frame items neutrally
Use forced-choice formats
Introduce random/unsystematic measurement
Use balanced scales
Introduce guessing penalties
105
Q

Construct Bias (AKA Measurement Bias)

A

– Teat may have different meaning for different groups

106
Q

Predictive Bias

A

– Test may have a different predictive value for different groups

107
Q

Reliability analysis

A

– internal consistency coefficients

108
Q

Rank order

A

– item difficulty ranking

109
Q

Item discrimination index

A

– item-level discrimination

110
Q

Factor analysis

A

– internal structure of the test

111
Q

Differential item functioning analysis

A

– probability of answering a certain way based on trait levels

112
Q

Universal design

A

– test is operationalized, taken, and measures to it is accessible to the most amount of people

113
Q

Norm-referenced tests

A

– compares a test taker to a reference sample

114
Q

Criterion-referenced tests

A

– uses a cutoff score to sort test takers into groups

115
Q

Correlation vs. Covariance

A

– correlation values are standardized, covariance values are not

116
Q

Trait vs. state variables

A

– 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