The aim of the study was to translate 'The COPD self-efficacy scale' (CSES) into Danish and to evaluate the psychometric properties of the Danish version (CSES-DK). CSES enables assessment of self-efficacy in individuals with chronic obstructive pulmonary disease (COPD). The scale consists of 34 items, describing situations which may cause dyspnoea in patients with COPD. The CSES was translated into Danish using a standard forward-backward translation procedure. To estimate the reliability, measurements of internal consistency and repeatability were applied. The validity of the Danish version was evaluated by examining the associations between the CSES-DK score and socio-demographic variables (age, gender, education, disease severity and self-rated health). Factor analysis was conducted to compare the internal structure of the Danish version and the American source version. The study included 151 patients with COPD, recruited from three outpatient clinics. Estimates of reliability were in accordance with the original version of CSES (Cronbach's α = 0.97, test-retest r = 0.82, p < 0.001). Significant correlations were obtained between the CSES-DK total score and vocational training and education (r = 0.27, p = 0.001), disease severity (r = -0.27, p = 0.001) and self-rated health (r = -0.41, p < 0.001), indicating construct validity. Five factors were extracted from both versions of CSES. However, in the CSES-DK, only one factor concerns emotions, whereas two factors describing emotions were obtained for the original scale. Furthermore, important discrepancies exist with respect to the direction of the scoring of CSES. In some studies, a high score indicates high self-efficacy, whereas it indicates low self-efficacy in other studies, which complicates the comparison of studies. The Danish version of CSES showed acceptable measurements of reliability and validity. Potential limitations of the scale were identified, and discrepancies exist between the factor structure of the original and Danish version. Consequently, more studies of the factor structure should be conducted on both the original CSES and the translated versions of the instrument.