Psychometrics Overview Quiz

Questions and Answers

What can be done to increase the reliability of the scale?

Create more items to increase reliability.

What could very large alpha values indicate?

Low internal consistency

Common variance

Item redundancy (correct)

Low item-total correlation

What is a typical source of measurement error?

Inter-rater differences

Percent agreement can __________ inter-rater reliability.

overestimate Signup and view all the answers

The preferred method for assessing inter-rater reliability is __________.

Kappa Signup and view all the answers

The standard error of measurement (SEM) will be high if reliability is high.

False Signup and view all the answers

Confidence intervals tell us about what?

The range in which the true score is likely to fall Signup and view all the answers

Multiplying the standard error of measurement (SEM) by the reliability will give you what?

Standard error of estimate (SEE) Signup and view all the answers

What does differential validity refer to?

Higher validity for one subgroup than another. Signup and view all the answers

Criterion contamination happens when criterion assessment is independent of test results.

False Signup and view all the answers

What test is particularly for children between ages 14 and 18?

MMPI-A Signup and view all the answers

Which measure may indicate overreporting?

All of these are measures Signup and view all the answers

What is the likely result if the F scale is between 80-89 and VRIN > 79?

Random responding Signup and view all the answers

An MMPI-2 profile is invalid if more than how many items have been omitted?

30 Signup and view all the answers

Scores of _ and lower on _ suggest a high degree of self-criticism.

40; K Signup and view all the answers

Someone with the following scales: F < 80, L < 60, and K between 40-60 can be described as:

Open and non-defensive Signup and view all the answers

A person who minimizes problems would likely score at least a _ on the _ scale.

65; K Signup and view all the answers

A person wishing to appear virtuous would likely score at least a _ on the _ scale.

65; L Signup and view all the answers

Moderate elevations of the L scale are between what range?

60-64 Signup and view all the answers

T-score elevations of 65-69 can be described as:

Mild Signup and view all the answers

T-scores must be _____ or higher to indicate severe pathology.

75 Signup and view all the answers

A spike occurs when the highest scale is more than _____ higher than the second highest scale.

5T Signup and view all the answers

High scores on the AAS indicate what regarding substance abuse behavior?

Both A and C Signup and view all the answers

It is possible to take a non-normal distribution and normalize the scores.

True Signup and view all the answers

A test can have strong reliability without having strong validity.

True Signup and view all the answers

What is the formula for converting a z-score into a t-score?

T = 10z + 50 Signup and view all the answers

Given that M = 40 and SD = 6, what is the z-score for a raw score of 38?

-0.33 Signup and view all the answers

Percentage of total variance in a test that is due to random measurement is called:

Error Variance Signup and view all the answers

What is the correct hierarchy for psychological testing, from most comprehensive to most specific?

Batteries; Tests; Scales; Items Signup and view all the answers

Which of the following are examples of cognitive functioning tests, except?

Projective Tests Signup and view all the answers

What do personality tests measure?

States or traits Signup and view all the answers

If a test is evaluating an individual against a set of norms collected from a particular population, what type of reference is it?

Norm-referenced Signup and view all the answers

Tests that emphasize speed over difficulty are known as what?

Speed tests Signup and view all the answers

Positive skews indicate that a test has an inadequate what?

Ceiling Signup and view all the answers

What should be done to solve positive skews?

Replace easier items with harder items Signup and view all the answers

Negative skews indicate that a test has inadequate what?

Ceiling Signup and view all the answers

What is the solution for negative skews?

Replace easier items with harder items Signup and view all the answers

In both positive and negative skews, the median is always in the middle. In positive skews, the mode is what relative to the median?

Below Signup and view all the answers

On a normal distribution with M = 60 and SD = 6, what percentage of scores will fall between 54 and 66?

68% Signup and view all the answers

On a normal distribution with M = 60 and SD = 6, what percentage of scores will fall at or below 60?

50% Signup and view all the answers

When calculating the coefficient of determination, you do what with the correlation coefficient?

Square the correlation coefficient Signup and view all the answers

If the reliability is 0.95, then what is the error variance?

Error variance Signup and view all the answers

What does a Z-score of 0 indicate?

The raw score is the mean Signup and view all the answers

What allows us to predict the alternate form reliability from split-half reliability?

Spearman-Brown formula Signup and view all the answers

What should you do if you have a low alpha score?

Revise the test items Signup and view all the answers

Study Notes

Psychological Testing Hierarchy

The hierarchy for psychological testing from most comprehensive to most specific is Batteries, Tests, Scales, Items.

Cognitive Functioning Tests

Projective Tests are not considered cognitive functioning tests, whereas Aptitude, Intelligence, and Achievement Tests are.

Personality Tests

Personality tests measure states or traits, focusing on individual characteristics rather than intelligence or abilities.

Norm-Referenced vs. Criterion-Referenced Testing

Norm-referenced tests evaluate against population norms, while criterion-referenced tests measure against predetermined criteria.

Speed vs. Power Tests

Speed tests emphasize speed over difficulty, while power tests emphasize difficulty over speed.

Skewness in Test Scores

Positive skews indicate inadequate floor; tests may be too difficult for test-takers. Solutions include replacing harder items with easier ones.
Negative skews indicate inadequate ceiling; tests may be too easy for test-takers. Solutions include replacing easier items with harder ones.

Measures of Central Tendency

In positively skewed distributions, the mode falls below the median, whereas the mean is above the median.

Normal Distribution Properties

Tails of normal distributions are asymptotic, meaning they approach but never reach zero.

Normal Distribution Statistics

On a normal distribution with a mean (M) of 60 and a standard deviation (SD) of 6:
- 68% of scores fall between 54 and 66, while 50% will fall at or below 60, and 84% will fall at or below 66.

WIAT Test

The WIAT utilizes developmental norms for assessing psychological capabilities.

Within-Group Norms

Within-group norms compare individual scores directly against peers using Z-scores, percentile ranks, and standard scores.

Cumulative Frequency and Percentile Rank

Cumulative frequency indicates how many achieved a score equal to or lower than a given raw score; percentile rank derives from frequency distributions.

Scales and Measurement Error

Standard scores use mean and SD to create scores relative to peers, while a Z-score of 0 indicates a score at the mean.

Correlation Coefficient and Determination

A correlation coefficient (r) of 0 indicates no relationship between X and Y; the coefficient of determination (r²) is found by squaring the correlation coefficient.

Factor Analysis Principles

Factor analysis assumes variables correlate due to underlying themes; rotation examines how factors correlate.
Factor loading indicates the correlation between original variables and factors; eigenvalues signify variance associated with each factor.

Reliability and Measurement Error

Classical reliability theory expresses that obtained scores consist of true scores plus error.
Measurement error sources include time sampling, item sampling, and inter-rater differences.

Test-Retest Reliability

Test-retest reliability assesses stability across time; overestimating occurs with short intervals, while longer intervals may lead to underestimation of reliability.

Internal Consistency and Item Sampling

Internal consistency checks how well items measure the same construct; item sampling errors emerge from non-random item selection.

Reliability Coefficients

Reliability coefficients reflect the test’s consistency, while the standard error of measurement (SEM) estimates potential score error for individuals.

Confidence Intervals

Confidence intervals indicate the range within which an individual's true score is likely to fall, factoring in measurement error.

General Statistics

Criterion-referenced tests offer specific measurements against a predefined standard, while Spearman's Brown formula helps predict alternate form reliability through split-half reliability.### Confidence Intervals and Regression
Confidence intervals (CIs) give a range of values around an estimated score.
Regression towards the mean: estimated true scores are closer to the mean than observed scores.
Reliability influences the difference between estimated true scores and observed scores: lower reliability results in greater differences.

Validity Types and Concepts

Validity is based on empirical evidence, theory, and score interpretation; all factors are essential.
Content validity may be compromised by irrelevant material or exclusion of necessary constructs.
The validity coefficient reflects the correlation between the test and the criterion, influenced by factors like range restriction.

Specific Validity Issues

Face validity is determined by non-experts, while content validity is assessed by experts.
Different types of validity such as convergent, divergent, and differential validity measure various aspects of test performance and group differences.
Criterion contamination occurs when the assessment of the criterion is affected by the test, potentially inflating or deflating the validity coefficient.

Construct Validity and Testing

Construct validity assesses whether a test measures what it claims, building a nomological net over time.
Cross-validation examines the generalizability of a formula to new samples and estimates shrinkage.
Factorial validity evaluates construct validity through the test's factor structure against theoretical expectations.

Item Characteristics and Test Bias

Item characteristic curves illustrate item performance relative to total test scores; steeper slopes indicate better discriminability.
Test bias occurs when a test has different predictive validity for different groups, manifesting as slope or intercept bias.
Item bias arises when questions are not universally applicable across different subgroups.

Psychological Testing and Standard Scores

For young children, the WISC-V is not suitable; however, the WAIS-IV is appropriate for high schoolers.
A normative strength is considered 115 and above, while a weakness is anything 85 and below.
MMPI-2 validity scales assess response consistency and tendency to present oneself positively or negatively.

Reliability and Validity in MMPI Testing

MMPI-2 scores help identify self-criticism, defensiveness, and overall psychological profiles.
Elevated scores on validity scales indicate potential issues with response accuracy or exaggeration of problems.
A spike in T-scores indicates significant differences across scales and requires careful interpretation.

General Psychological Concepts and True/False

It's possible to normalize non-normal distributions.
Curvilinear relationships cannot be accurately described by Pearson's r.
Strong reliability does not guarantee strong validity, and the split-half method may not be applicable to speed tests.
Reliability is influenced by both test length and item characteristics.### True/False Statements
The mean of observed scores equals the mean of true scores.
Random error's average effect across respondents is zero.
Measuring random error is inherently impossible.
Low face validity in a test may cause test-takers to be cautious, introducing error.
If the criterion measure has low reliability, strong evidence for validity is hard to establish, even for a good construct measure.
Multicollinearity occurs when two variables measure the same construct and leads to larger shrinkage in results.
The Information subtest in the WISC's VCI scale is supplemental, while in the WAIS it is not.
Picture Span is unique to the WISC and does not appear in the WAIS.
A difference of 20 points or more between VCI and WMI renders FSIQ invalid.
A difference of 20 points or more between PRI and PSI also renders FSIQ invalid.

Score Conversion Formulas

To convert a z-score to a t-score: T = 10z + 50.
To convert a t-score to a z-score: Z = (T - 100) / 15.
To convert a standard score to a z-score: Z = (SS - 100) / 15.
To convert a z-score to a standard score: SS = 15z + 100.

Specific Score Calculations

The standard score corresponding to a t-score of 64 is 121.
To calculate z-score from individual raw score: Z = (x - M) / SD.
Given M = 40 and SD = 6, the z-score for a raw score of 38 is -0.33.
Given M = 40 and SD = 6, the standard score for a raw score of 48 is 120.

Variance Types

Common Variance: Shared variance with other tests in factor analysis.
Specific Variance: Variance unique to the test, not shared with others.
Error Variance: Variance due to random measurement errors.

Validity Scales for MMPI-2

L: Lie scale measuring inconsistency in responses.
Fp: Infrequency-Psychopathology indicating rare responses.
K: Correction scale for socially desirable responding.
VRIN/TRIN: Response consistency scales.
Fb: Infrequency, second half of the test.
S: Superlative scale assessing inflated self-reports.
F: Infrequency scale assessing uncommon responses.

Clinical Scales of MMPI-2

Scale 0: Social Introversion measuring isolation tendencies.
Scale 1: Hypochondriasis focusing on excessive bodily concerns.
Scale 2: Depression measuring depressive symptoms.
Scale 3: Hysteria correlating with psychological distress.
Scale 4: Psychopathic Deviate assessing rebelliousness.
Scale 5: Masculinity-Femininity measuring traditional gender traits.
Scale 6: Paranoia assessing suspiciousness.
Scale 7: Psychasthenia related to anxiety and self-doubts.
Scale 8: Schizophrenia relating to odd thoughts and beliefs.
Scale 9: Mania indicating overactivity and impulsivity.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

Test your knowledge of psychometrics with this overview quiz. It covers various aspects of psychological testing, including the hierarchy of tests and examples of cognitive functioning tests. Perfect for students and professionals in psychology and related fields.