Chapter 5 Flashcards
Predictive Study
Test is adminisistered, there is an interviewing time interval, and then the criterion is measured.
*Validity study
Concurrent Studies
Test is administered and criterion is measured at about the same time.
- Validity Study
Test Criterion Evidence
AKA
(predictive validity) (Criterion Validity) (Criterion-Related Validity)
Two types used
- Predictive studies: test admin, interval, then criterion measured
- Concurrent Studies: test admin then criterion measured same time.
Standard Error of estimate
Describes the amount of prediction error due to the imperfect validity of interpretation of a test score.
Decision Theory Model
Helps the test user determine how much information a predictor test can contribute when making classification decisions.
Multitrait-multimethod matrix
combines convergent and divergent strategies.
Validity argument
integration of numerous lines of evidence into coherent commentary.
Validity
appropriatenss and accuray of the interpretation of test scores.
Validity is a
unitary concept existing on a continuum, it’s not all or none.
Page 159 at bottom
Content Coverage
Experts look at the overall test and rate the degree to which the items cover specific domain
Item Relevance
Experts examine each individual test item and determine whether it reflects essential content in the specified domain.
Face Validity
NOT technically a FORM of Validity.
A test appearing to measure what it is designed to measure.
Threats to validity
2 threats
- test measures less or more than construct its supposed to measure.
- Construct irrelevant variance- content measures are unrelated content or characteristics of test
Reliability and Validity
without reliability there can be no validity.
Reliability is a necessary but insufficient condition for validity. Only true score variance is reliable. If reliability is equal to zero than true score variance has to be equal to zero.
Content Valdity
How adequate the test samples the content area of the construct.
Appropriateness of the test.
EX: Employers use content validity to classify employees before hired.
Criterion-related Validity
relationship between the test and external variables that are thought to be the direct measures of the construct. uses correlation regression analysis.
Construct Validity
integration of evidence that relates to the meaning or interpretation of test scores.
Sensitivity
Classify individual or characteristic
the ability of a test or instrument to detect disorder when present.
Specificity
classify individuals or characteristic
ability of the instrument to detect the absence of the disorder.