Reliability
Reliability is the consistency between independent measurements of the same thing (Aksayan & Gozum 2002). Cronbach’s a values, calculated to determine the internal consistency of Turkish version of the SF-MPQ, were found to be quite high, indicating excellent reliability (Cronbach’s a 0Æ88 for test and 0Æ91 for retest). Turkish version of SF-MPQ assures the reliability of its within-day measurements with coefficient was within the range from 0Æ83 to 0Æ92 for withinday measurements. Test–retest reliabilities coefficients of the administered tests were relatively similar to the previous studies (Dudgeon et al. 1993, Burckhardt & Bjelle 1994, Georgoudis et al. 2001, Yakut et al. 2007). In our study, for tool stability test–retest reliability for total, sensory, affective and ETPI scores of 0Æ85, 0Æ84, 0Æ82, and 0Æ70, respectively. High values (0Æ891–0Æ716) were also demonstrated by Yakut et al. (2007) in rheumatic patients using a test–retest design within the same day, while Burckhardt and Bjelle (1994) reported a value of 0Æ73 using a one-month interval between test and retest. Grafton et al. (2005) demonstrated excellent test–retest reliability (5 days apart) by ICC values ranging 0Æ96–0Æ88 for total, sensory and affective scores in osteoar- thritis patients awaiting hip- or knee-joint replacement and Georgoudis et al. (2001) also demonstrated excellent ICC values (0Æ91–0Æ87) in a mixed group of patients, although retest was performed after 15 days. The time intervals also differed, and when a test–retest reliability study is carried out over several days, pain is more likely to change than when it is performed during the course of one day, as demonstrated by Burckhardt and Bjelle (1994) and Georgoudis et al. (2001). When judging ICC values, all these factors should be taken into consideration. Although our values are somewhat lower than those obtained by Grafton et al. and Georgoudis et al., Turkish version of the tool still show excellent good reliability. These differences, which are characteristics of the culture of our country, may have affected the pain experiences of individuals. A complex interchange between cultural values, beliefs, religion and linguistic factors may be contributing to these observed differences.