Related Experiment Video
Updated: May 8, 2026
The Multiple Sclerosis Performance Test (MSPT): An iPad-Based Disability Assessment Tool
Published on: June 30, 2014
Assessing social isolation in motor neurone disease: a Rasch analysis of the MND Social Withdrawal Scale
Chris J Gibbons1, Everard W Thornton, John Ealing
1The Walton Centre for Neurology and Neurosurgery, Lower Lane, Liverpool, UK; NIHR Collaboration for Leadership in Applied Health Research and Care for Greater Manchester, Centre for Primary Care, University of Manchester, Manchester, UK.
Objective:
Social withdrawal is described as the condition in which an individual experiences a desire to make social contact, but is unable to satisfy that desire. It is an important issue for patients with motor neurone disease who are likely to experience severe physical impairment. This study aims to reassess the psychometric and scaling properties of the MND Social Withdrawal Scale (MND-SWS) domains and examine the feasibility of a summary scale, by applying scale data to the Rasch model.
Methods:
The MND Social Withdrawal Scale was administered to 298 patients with a diagnosis of MND, alongside the Hospital Anxiety and Depression Scale. The factor structure of the MND Social Withdrawal Scale was assessed using confirmatory factor analysis. Model fit, category threshold analysis, differential item functioning (DIF), dimensionality and local dependency were evaluated.
Results:
Factor analysis confirmed the suitability of the four-factor solution suggested by the original authors. Mokken scale analysis suggested the removal of item five. Rasch analysis removed a further three items; from the Community (one item) and Emotional (two items) withdrawal subscales. Following item reduction, each scale exhibited excellent fit to the Rasch model. A 14-item Summary scale was shown to fit the Rasch model after subtesting the items into three subtests corresponding to the Community, Family and Emotional subscales, indicating that items from these three subscales could be summed together to create a total measure for social withdrawal.
Conclusion:
Removal of four items from the Social Withdrawal Scale led to a four factor solution with a 14-item hierarchical Summary scale that were all unidimensional, free for DIF and well fitted to the Rasch model. The scale is reliable and allows clinicians and researchers to measure social withdrawal in MND along a unidimensional construct.
