A new scale measuring adaptive perceived control for people with Parkinson's: initial construction and further validation

Simpson, J, Chatzidamianos, Gerasimos, Fletcher, I, Perpetuo, L and Eccles, FJR (2018) A new scale measuring adaptive perceived control for people with Parkinson's: initial construction and further validation. Journal of the Neurological Sciences, 391. pp. 77-83. ISSN 0022-510X

Preview

Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (480kB) | Preview

Abstract

Introduction Perceived control is an important concept in understanding adjustment to chronic conditions such as Parkinson’s. While generic measures have been used to measure the construct in Parkinson’s, no Parkinson’s-specific scale currently exists. This study outlines the initial development and further validation of a free-to-use scale, the Parkinson’s UK Scale of Perceived Control (PUKSoPC). Method Focus groups were used to create items for the new scale. Potential items were then subject to screening for readability and coherence by people affected by the condition. This left 49 items that were then completed, along with other measures, by 231 people with Parkinson’s. Exploratory factor analysis then created a 15-item scale with five distinct subscales. This initial structure was then further tested using confirmatory factor analysis with 2032 people with Parkinson’s. Structural equation modelling confirmed the acceptability of the total scale and subscale structures. Results The final scale is concluded to be a psychometrically robust measure of perceived control. It has good face validity, evidence of convergent and criterion (concurrent and divergent) validity, good test-retest reliability and is internally coherent, with a demonstrably solid factor structure. While further testing would be useful to assess the scale’s predictive ability, it is currently considered robust enough for more widespread use. Conclusion The PUKSoPC is an appropriate scale to provide a more comprehensive measure of perceived control. It is preferable to single item, non-validated measures and can provide evidence of perceptions of control across a number of domains important in the measurement of the construct.