Dissertation
An Investigation of the Relationship Between the Item-Content and Diagnostic Validity of an Autism Screening Instrument
Doctor of Philosophy (PhD), Washington State University
01/2016
Handle:
https://hdl.handle.net/2376/12040
Abstract
The present study investigated whether the diagnostic validity of a poorly-performing instrument (the Gilliam Autism Rating Scale – Second Edition, or the GARS-2) designed to screen for autism spectrum disorders could be improved through item content manipulation (via altering the weighting of items and including new items). In previous studies, the GARS-2 has been shown to have poor sensitivity and specificity. Item content manipulations were theory-based, and involved efforts to align item content with existing diagnostic criteria and with the item content of frequently utilized criterion instruments [e.g., the Autism Diagnostic Observation Schedule (ADOS)]. Consistent with expectations, results revealed that by reweighting items to mirror the ADOS and diagnostic criteria distributions, and by including items that assess executive functioning deficits, the sensitivity and specificity could be significantly improved. These results have implications for (a) improving the validity of a currently available instrument, and (b) developing instruments with high diagnostic validity in the future. These findings are timely, given that the recent advent of the DSM-5 (and the new autism diagnostic criteria it contains) necessitates the development of new screening instruments or the modification of existing ones.
Metrics
68 File views/ downloads
19 Record Views
Details
- Title
- An Investigation of the Relationship Between the Item-Content and Diagnostic Validity of an Autism Screening Instrument
- Creators
- Justin Samuel Hampton
- Contributors
- Paul Strand (Advisor)David Marcus (Committee Member)Maria Gartstein (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Psychology, Department of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Number of pages
- 83
- Identifiers
- 99900581530101842
- Language
- English
- Resource Type
- Dissertation