JPASPEX (2013) 1(1), 74-77 http://www.jpaspex.com Factorial Validity and Internal Consistency of Iranian Adapted Brunel Mood Scale among Iranian Young Female Taekwondo Athletes Tahereh Bagherpour 1  Hairul Anuar Hashim 1 and Soumendra Saha 2 Sports Science Unit, School of Medical Sciences, Universiti Sains Malaysia,16150, Kubang kerian, Kelantan, Malaysia 1 , School of Health sciences, Universiti Sains Malaysia, 16150, Kubang kerian, Kelantan, Malaysia 2 Abstract The present study evaluated the factorial validity and internal consistency of the Iranian version of the BRUMS. BRUMS was administered to a total of 131 female Taekwondo athletes. Three models were tested, a 1-factor model, 2-factor model and a 6-factor model. CFA results revealed close model fit of the 2-factor model (x 2 = 482.70., df =251, p< 0.001; CFI=0.88, TLI=0.86, RMSEA=0.05, GFI=0.86) when compared to the 1-factor model and the 6-factor model. The results also showed a satisfactory level of internal consistency of the 2-factor model (α = .80 for positive mood and .82 for negative mood subscales). The findings suggested that BRUMS can be used for measurement of general mood states (positive and negative) among Iranian female Taekwondo athletes. However, there is need for future studies investigating the generalizability of the results with different samples of age groups, gender and types of sports. Key words: Confirmatory factor analysis, mood states, taekwondo athletes Introduction Mood is one of the most frequently researched areas in the field of sport and exercise psychology. This stems from the fact that moods have been found to influence individuals’ mental and physical wellbeing (Terry et al., 1999). Given its importance, there is a growing interest in measuring and profiling mood states across different populations. In sports, the assessment of mood is commonly done using the Profile of Mood States (POMS; McNair, Lorr, & Droppleman, 1971, 1992). POMS consists of 65 items representing six subscales. Mood-performance relationship has been proposed to exhibit a special pattern (Morgan, 1979). Specifically, Morgan (1979) proposes that performance is optimum when the mood exhibits an iceberg profile. According to Morgan (1980) performance was associated with above average vigour, coupled with below average anger, confusion, depression, fatigue, and tension, a profile that when plotted graphically resembled that of an iceberg. Despite being a preferred instrument to measure mood states among coaches and practitioners, it has been criticized for being time-consuming. Consequently, Terry and colleagues (Terry, Lane, Lane, & Keohane, 1999; Terry, Lane, & Fogarty, 2003) developed Profile of mood Scale for Adolescent (Terry et al., 1999) and it was later renamed Brunel Mood Scale which is also applicable for adult (Terry et al., 2003). BRUMS validity studies such as one conducted by Fazackerley, Lane and Mahoney (2003) provided support for the validity of the 6-factor model, with a Comparative Fit Index (CFI) of 0.90 and Root Mean Squared Error of Approximation of (RMSEA) 0.07. Furthermore, the internal consistency coefficients of the subscales were above the 0.70 criterion. Besides, BRUMS has also been adapted into several different languages such as Malaysian and studies of these adapted versions provide mixed results (Hashim, Zulkifli, & Yusof, 2010, Lan, Roy, Lane, & Azma , 2012). Despite evidence of its psychometric properties across studies involving various population, there are also some contradictory findings. For instance, Hashim et al. (2010) in their study among 355 adolescent athletes revealed adequate model fit in a 6-factor model, but with one of the items removed (item 24). Internal consistency of the questionnaire was marginally supported through alpha reliability method. Alpha coefficients of 0.72, 0.64, 0.73, 0.69, 0.65, and 0.58 were obtained for tension, depression, anger, and confusion, fatigue, and vigour subscales respectively. Closer inspection of items for confusion showed that the item 24 may be problematic. The alpha coefficient increased to 0.67 for this subscale when this item removed. In another study of the Persian version of BRUMS involving 405 Iranian university students (136 males, 269 females), the results provided support for six-factor measurement model but with a final 14-item (Terry et al., 2012). Given these contradictory findings, there seem to be a need for further analyses involving independent samples to confirm the generalizability of the psychometric properties of BRUMS, especially the adapted versions. Therefore, the present findings seek to validate the psychometric properties of Persian version of BRUMS among Iranian female taekwondo athletes. Research article 74