%0 Journal Article
%T Scale construction utilising the Rasch unidimensional measurement model: A measurement of adolescent attitudes towards abortion
%A Jacqueline Hendriks
%A Sue Fyfe
%A Irene Styles
%A S. Rachel Skinner
%J Australasian Medical Journal
%D 2012
%I The Australasian Medical Journal pty ltd.
%X BackgroundMeasurement scales seeking to quantify latent traits likeattitudes, are often developed using traditionalpsychometric approaches. Application of the Raschunidimensional measurement model may complement orreplace these techniques, as the model can be used toconstruct scales and check their psychometric properties. Ifdata fit the model, then a scale with invariant measurementproperties, including interval-level scores, will have beendeveloped.AimsThis paper highlights the unique properties of the Raschmodel. Items developed to measure adolescent attitudestowards abortion are used to exemplify the process.MethodTen attitude and intention items relating to abortion wereanswered by 406 adolescents aged 12 to 19 years, as part ofthe ˇ°Teen Relationships Studyˇ±. The sampling frameworkcaptured a range of sexual and pregnancy experiences.Items were assessed for fit to the Rasch model includingchecks for Differential Item Functioning (DIF) by gender,sexual experience or pregnancy experience.ResultsRasch analysis of the original dataset initially demonstratedthat some items did not fit the model. Rescoring of one item(B5) and removal of another (L31) resulted in fit, as shownby a non-significant item-trait interaction total chi-squareand a mean log residual fit statistic for items of -0.05(SD=1.43). No DIF existed for the revised scale. However,items did not distinguish as well amongst persons with themost intense attitudes as they did for other persons. Aperson separation index of 0.82 indicated good reliability.ConclusionApplication of the Rasch model produced a valid andreliable scale measuring adolescent attitudes towardsabortion, with stable measurement properties. The Raschprocess provided an extensive range of diagnosticinformation concerning item and person fit, enablingchanges to be made to scale items. This example shows thevalue of the Rasch model in developing scales for bothsocial science and health disciplines.
%K Rasch unidimensional measurement model
%K adolescent
%K abortion
%K attitudes
%K attitude scale
%U http://www.amj.net.au/index.php?journal=AMJ&page=article&op=viewFile&path%5B%5D=952&path%5B%5D=929