BACKGROUND Imagery is an effective performance enhancement technique. Imagery has been described previously in a range of psychological domains. Measuring imagery is critical in research and practice in sport. Self-report questionnaires are the most regularly used method. The aim of the present study was to examine reliability and validity characteristics of the Imagination in Sport Questionnaire (Kwestionariusz Wyobrazni w Sporcie - KWS). PARTICIPANTS AND PROCEDURE Five and hundred eight (N = 326 - study I; N = 182 - study II) Polish athletes completed questionnaires (169 male, 156 female - study I; 139 male, 43 female - study II), aged between 12 and 57 years (M = 22.08, SD = 8.18 - study I; age 19- 24, M = 20.46, SD = 1.1 - study II), at different competitive levels and recruited from various sports disciplines. RESULTS Results indicated the maintained good stability and internal consistency over a 3-week period. Results of confirmatory factor analysis suggested that the 7-factor structure of the KWS resulted in acceptable model fit indices (NC = 2416.63, df = 1203, GFI = 0.944, AGFI = 0.944, CFI = 0.786, RMSEA = 0.056, p (RMSEA < 0.05) = 0.002 first study; NC = 2234.39, df = 1203, GFI = 0.673, AGFI = = 0.640, CFI = 0.691, RMSEA = 0.069, p (RMSEA < 0.05) = = 0.000 - second study). Concurrent validity was supported by examination of the relationships between the KWS subscales and the SIAM (Sport Imagery Ability Measure) in Polish adaptation. In addition, differences in athletes' imagery ability were examined across competitive levels, and in relation to both gender and age. CONCLUSIONS Overall, the results supported the reliability and construct validity of the KWS.