Picture of person typing on laptop with programming code visible on the laptop screen

World class computing and information science research at Strathclyde...

The Strathprints institutional repository is a digital archive of University of Strathclyde's Open Access research outputs. Strathprints provides access to thousands of Open Access research papers by University of Strathclyde researchers, including by researchers from the Department of Computer & Information Sciences involved in mathematically structured programming, similarity and metric search, computer security, software systems, combinatronics and digital health.

The Department also includes the iSchool Research Group, which performs leading research into socio-technical phenomena and topics such as information retrieval and information seeking behaviour.

Explore

Filling in the missing pieces: a comparative Rasch analysis of inpatient and outpatient SCL90R data

Elliott, Robert and Fox, Christine and Beltyukova, Svetlana A. and Moskowitz, Andrew (2009) Filling in the missing pieces: a comparative Rasch analysis of inpatient and outpatient SCL90R data. In: SPR (UK) Ravenscar research Conference 2009, 2009-03-14 - 2009-03-16. (Unpublished)

Full text not available in this repository. Request a copy from the Strathclyde author

Abstract

In our Rasch analysis of the Symptom Checklist-90-R (SCL90R) using an American outpatient sample (n=298), Elliott et al. (2006) noted among other things that there was a sampling gap in the upper extreme range of the measured underlying client distress dimension. In this paper we remedy this gap by replicating our earlier results with a Scottish inpatient sample (n= 135). When analyzed separately, a three-point rating scale again appeared to provide a better trade-off between reliability and simplicity than the standard 5-point scale. The inpatient sample showed comparable person separation to the outpatient sample (5.13 vs. 5.31), but as expected because of greater participant homogeneity, item separation was substantially lower (4.76 vs. 8.89), indicating less discrimination among items, but equivalent discrimination among persons. Interestingly, when we collapsed the rating scale into only two points ('not at all' vs. all other ratings), there was very little loss of information. Ironically, this suggests that the SCL90R could be used as a 'checklist' after all, in which symptoms are rated either present or absent.