Premium
Improving Measurement in Couple and Family Therapy: An Item Response Theory Primer
Author(s) -
Anderson Shayne R.,
Miller Richard B.
Publication year - 2020
Publication title -
journal of marital and family therapy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.868
H-Index - 68
eISSN - 1752-0606
pISSN - 0194-472X
DOI - 10.1111/jmft.12448
Subject(s) - item response theory , classical test theory , field (mathematics) , psychology , marital therapy , quality (philosophy) , computer science , scale (ratio) , test theory , construct (python library) , psychometrics , cognitive psychology , clinical psychology , mathematics , epistemology , philosophy , physics , quantum mechanics , pure mathematics , programming language
The valid and reliable assessment of individual and relational functioning relies on high‐quality assessment tools. Most assessments used in the field of couple and family therapy were developed using Classical Test Theory (CTT). An alternate theory for guiding the creation, evaluation, and scoring of assessments is Item Response Theory (IRT). IRT has several advantages over CTT and can be used to improve measurement in the field. For example, the IRT approach creates measures that are significantly more precise, yet uses fewer items, than scales created using CTT. This manuscript provides an overview of how IRT differs from CTT and describes the fundamental concepts and assumptions of IRT. Following this summary, we provide a step‐by‐step example of how IRT can be used to reduce the length of the Marital Disaffection Scale from 21 to 6 items without losing a significant amount of information about the construct.