Validity and Item Analysis Chapter 4
Concerns what instrument measures and how well it does so Not something instrument “has” or “does not have” Informs counselor when it is appropriate to use instrument and what can be inferred from results Reliability is a prerequisite for validity Validity
Content-Related Criterion-Related Construct Traditional Categories of Validity
Evidence Based on Test Content Evidence Based on Response Processes Evidence Based on Internal Structure Evidence Based on Relations to Other Variables Evidence Based on Consequences of Testing Validation Evidence: Standards for Educational and Psychological Testing
Degree to which the evidence indicates that items, questions, or tasks adequately represent intended behavior domain Central focus is typically on how the instrument’s content was determined Content-related validation evidence should not be confused with face validity Evidence Based on Test Content
Concerns whether individuals respond in a manner consistent with construct measured May also examine information processing differences by subgroup Computer simulated performances Evidence Based on Response Processes
Examining internal structure using factor analysis Can also examine internal structure of instrument for different subgroups (differential item functioning) Evidence Based on Internal Structure
Correlational method Convergent and discriminant evidence Multitrait-multimethod matrix Prediction or instrument-criterion relationship Concurrent and predictive validity Regression Decision Theory Validity Generalization Evidence Based on Relations to Other Variables
Correlational Method Correlational method examines relationship between instrument and criterion or other pertinent variables Validity coefficient Multitrait-multimethod matrix examines relationships between instrument and traits that are theoretically related and unrelated to it
Concurrent validity Predictive validity Prediction/Instrument-Criterion Relationship
Regression a = y-intercept b = slope X = score
Standard Error of Estimate The margin of expected error in the individual’s predicted criterion score as a result of imperfect validity
Decision Theory: Expectancy Table
Expectancy Table (cont.)
Method of combining validation studies to determine if validity evidence can be generalized Must have a substantial number of studies Must use meta-analytic procedures Validity Generalization
Examples of social consequences: Group differences on tests used for employment selection Group differences in placement in special education Counselors should consider both validation evidence and social implications Evidence-Based Consequences of Testing
Gradual accumulation of evidence Counselor must evaluate the information to determine if it appropriate for which client under what circumstance Validation evidence should also be considered in informal assessments Conclusion on Validation Evidence
Examining and evaluating each item in instrument Item difficulty: p = # who answered item correctly total # of individuals Item discrimination: Extreme group method: d = upper % – lower % Correlational method Item Analysis
Focus is on each item and establishing items that measure ability or level of a latent trait Involves examining item characteristic curve and calibration of each individual item Parameters: Difficulty Slope (discrimination) Probability of guessing Item Response Theory
Item Characteristic Curve