~AEA Public Library

AEA 2015 Session 1483: Examining How Rasch/IRT Can Increase Measurement Precision 

11-16-2015 16:08

Recent advances in measurement have important implications for the quantitative instruments (measures, surveys, etc.) used in conducting evaluations. This Master Teacher Series presentation illustrates how the advances in measurement theory, specifically item response theory (IRT) and particularly Rasch models, have distinct advantages over classical test theory (CTT) measurement models in terms of the precision of the measures (surveys) and the response options we attach to these measures (Likert scales) we use to determine the effectiveness of the programs and interventions we evaluate. The presentation provides an accessible and understandable overview of Rasch, other IRT models and their differences; as well as how they can be used in developing evaluation measures, selecting survey response options, and accessing the precision of data we have already collected. The advantages of using Rasch and IRT approaches are illustrated in two evaluation – 1) health provider effectiveness from a large data set of a commercial health plan, and 2) national evaluation of the effectiveness of treatment approaches for addiction.

#2015Conference

Statistics
0 Favorited
10 Views
1 Files
0 Shares
355 Downloads

Related Entries and Links

No Related Resource entered.

Tags and Keywords

Attachment(s)
pdf file
IRT Can Increase Measurement Precision   4.76 MB   1 version
Uploaded - 11-16-2015
Recent advances in measurement have important implications for the quantitative instruments (measures, surveys, etc.) used in conducting evaluations. This Mster Teacher Series presentation illustrates how the advances in measurement theory, specifically item response theory (IRT) and particularly Rasch models, have distinct advantages over classical test theory (CTT) measurement models in terms of the precision of the measures (surveys) and the response options we attach to these measures (Likert scales) we use to determine the effectiveness of the programs and interventions we evaluate. The presentation provides an accessible and understandable overview of Rasch, other IRT models and their differences; as well as how they can be used in developing evaluation measures, selecting survey response options, and accessing the precision of data we have already collected. The advantages of using Rasch and IRT approaches are illustrated in two evaluation – 1) health provider effectiveness from a large data set of a commercial health plan, and 2) national evaluation of the effectiveness of treatment approaches for addiction.