Open AccessReducing the length of mental health instruments through structurally incomplete designs
Author(s) -
Smits Niels,
Cuijpers Pim,
Beekman Aartjan T. F.,
Smit Johannes H.
Publication year - 2007
Publication title -
international journal of methods in psychiatric research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.275
H-Index - 73
eISSN - 1557-0657
pISSN - 1049-8931
DOI - 10.1002/mpr.223
Subject(s) - missing data , set (abstract data type) , mental health , test (biology) , depression (economics) , quarter (canadian coin) , statistics , scale (ratio) , data set , psychology , contrast (vision) , econometrics , computer science , mathematics , psychiatry , artificial intelligence , paleontology , history , physics , macroeconomics , archaeology , quantum mechanics , economics , biology , programming language
This paper presents structurally incomplete designs as an approach to reduce the length of mental health tests. In structurally incomplete test designs, respondents only fill out a subset of the total item set. The scores on the unadministered items are estimated using methods for missing data. As an illustration, structurally incomplete test designs recording, respectively, two thirds, one half, one third and one quarter of the complete item set were applied to item scores on the Centre of Epidemiological Studies‐Depression (CES‐D) scale of the respondents in the Longitudinal Aging Study Amsterdam (LASA). The resulting unobserved item scores were estimated with the missing data method Data Augmentation. The complete and reconstructed data yielded very similar total scores and depression classifications. In contrast, the diagnostic accuracy of the incomplete designs decreased as the designs had more unobserved item scores. The discussion addresses the strengths and limitations of the application of incomplete designs in mental health research. Copyright © 2007 John Wiley & Sons, Ltd.