Reliability of the Swedish version of the Exercise Self-Efficacy Scale (S-ESES): a test-retest study in adults with neurological disease

Objective: To examine the test-retest reliability of the Swedish translated version of the Exercise Self-Efficacy Scale (S-ESES) in people with neurological disease and to examine internal consistency. Design: Test-retest study. Subjects: A total of 30 adults with neurological diseases including: Parkinson's disease; Multiple Sclerosis; Cervical Dystonia; and Charcot-Marie-Tooth disease. Method: The S-ESES was sent twice by surface mail. Completion interval mean was 16 days apart. Weighted kappa, intraclass correlation coefficient 2,1 [ICC (2,1)], standard error of measurement (SEM), also expressed as a percentage value (SEM%), and Cronbach's alpha were calculated. Results: The relative reliability of the test-retest results showed substantial agreement measured using weighted kappa (MD = 0.62) and a very high-reliability ICC (2,1) (0.92). Absolute reliability measured using SEM was 5.3 and SEM% was 20.7. Excellent internal consistency was shown, with an alpha coefficient of 0.91 (test 1) and 0.93 (test 2). Conclusion: The S-ESES is recommended for use in research and in clinical work for people with neurological diseases. The low-absolute reliability, however, indicates a limited ability to measure changes on an individual level.