Researchers have studied affective constructs such as emotion and mood in the context of sport and exercise domains for many decades (e.g., Morgan, 1980; Terry, 1995; Lazarus, 2000; Lane, 2007), producing a substantial body of knowledge related to the theory and measurement of the mood construct (e.g., McNair et al., 1971, 1992; Lane and Terry, 2000; Terry et al., 2003a). Mood is defined as “a set of feelings, ephemeral in nature, varying in intensity and duration, and usually involving more than one emotion” (Lane and Terry, 2000, p. 17). Several researchers have identified important conceptual distinctions between mood and emotion (e.g., Ekman, 1999; Beedie et al., 2005). Emotions are typically described as more intense, quicker to rise and shorter-lived than moods, and associated with more identifiable sources than mood. Conversely, moods are generally described as lower intensity, diffuse and enduring, without obvious antecedent cause, and less subject to conscious monitoring and control (Beedie et al., 2005). Further, moods tend to impart a more global influence over feelings, thoughts and behaviors, whereas emotions tend to engender the development of specific responses to situations requiring immediate action (Lane, 2007).

Although the mood construct is sometimes conceptualized as a unified psychological state, Lane (2007) stressed that moods involve a complex combination of feelings and that the interaction among mood dimensions, rather than any one aspect of mood in isolation, exerts influence on the behavior and ultimately on the performance of individuals. The relationship between affective states and performance has been studied extensively and is well-established in a variety of performance domains (e.g., Hanin, 1997; Beedie et al., 2000). Pleasant moods are generally associated with good performance whereas unpleasant moods tend to be associated with poor performance, and it is not unusual for athletes to attribute unsatisfactory performance to an inability to get into the “right mood” (Lane and Terry, 2016). However, research has demonstrated that unpleasant moods can sometimes facilitate good performance and pleasant moods may be associated with poor performance (Hanin, 1997; Beedie et al., 2000). Lane and Terry developed and tested a conceptual model to explain how different dimensions of mood interact to influence performance in sport and academic settings (Lane and Terry, 2000, 2005). In this model, depression is proposed to play a central role, although it is important to note that the term depression or depressed mood is used to describe unhappiness, dissatisfaction or dysphoria rather than a clinical condition.

Lane and Terry's model was based on the dimensions of mood as described by the Profile of Mood States (POMS; McNair et al., 1971). The POMS, initially developed to assess mood in clinical populations and subsequently in college populations, was later claimed to be also valid for use in sport and exercise settings (McNair et al., 1992) and has been used in several hundred sport-related studies (Lane, 2007). A number of abbreviated versions of the POMS have been developed, some of which have been criticized for using small validation samples or for incomplete validation processes (Lane, 2007). To address these limitations, Terry et al. developed and validated a 24-item short version of the POMS, designed primarily for use in sport and exercise domains, referred to as the Brunel Mood Scale (BRUMS; Terry et al., 1999, 2003a). The BRUMS includes six subscales corresponding to the underlying factors of Anger, Confusion, Depression, Fatigue, Tension, and Vigor. To account for the ephemeral nature of the mood construct, the BRUMS asks respondents to rate how they feel right now rather than retrospectively over the past week including today, as in the original POMS (Lane and Terry, 2000). The BRUMS has demonstrated satisfactory predictive, concurrent, criterion, and factorial validity as well as appropriate test-retest reliability (Terry et al., 1999, 2003a). Tables of normative data are available for athletes and non-athletes, covering both adolescents and adults (Terry et al., 2003a; Terry and Lane, 2010).

In addition to being used extensively with athlete samples, the BRUMS has been used in medical screening protocols (Galambos et al., 2005), to assess the risk of post-traumatic stress disorder among combat troops (van Wijk et al., 2013), to assess the risk of youth suicide (Gould et al., 2005), and to monitor cardiac rehabilitation patients (Sties et al., 2014). Although these uses include screening protocols in clinical environments (van Wijk et al., 2013), Lane and Terry (2016) stressed that the BRUMS should not be used as a stand-alone diagnostic tool.

When applying theories and models across cultures and nations, it is necessary to translate instruments, accounting for linguistic and cultural differences (Moustaka et al., 2010). The majority of published measures in psychology are produced in the English language, which provides a significant challenge for professionals working in other contexts and aiming to use appropriate, linguistically-validated scales (Rohlfs et al., 2008; Terry et al., 2012). Although researchers have long stressed the need to extend sport and exercise psychology research to cross-cultural settings (Duda and Allison, 1990), the need to develop cross-culturally validated measurement scales has become more acute due to the global expansion of the field (Quartiroli et al., 2014).

To address the need for greater cross-cultural generalizability of research findings, the BRUMS has been translated and validated in several languages, including Afrikaans (Terry et al., 2003b), Brazilian Portuguese (Rohlfs et al., 2008), French (Rouveix et al., 2006), Chinese (Zhang et al., 2014), Farsi (Terry et al., 2012), Hungarian and Italian (Lane et al., 2007), and Malay (Lan et al., 2012). Generally, a self-report measure is considered valid if it demonstrates content, factorial, criterion, and construct validity (American Psychological Association, 2014). Although an Italian translation exists of a 32-item, 8-factor version of the BRUMS, its authors noted that the scale suffered from “marginal fit indices, inconsistent factor loadings and relationships between mood states” (Lane et al., 2007, p. 121). We therefore judged that an Italian version of the more widely-used 24-item, 6-factor version of the BRUMS (Terry et al., 1999, 2003a) should be developed using a rigorous translation and validation process. This new measure is referred to as the Italian Mood Scale (ITAMS).

The primary aim of the study was to evaluate the factor structure of the ITAMS against the hypothesized 6-factor measurement model. The secondary aim of the study was to evaluate the concurrent validity of the ITAMS using the PANAS (Watson et al., 1988) and the Depression, Anxiety, and Stress Scale (DASS-21: Henry and Crawford, 2005) as external references. These scales were used because they featured in previous BRUMS validations (Terry et al., 2003a) and because valid and reliable Italian-language versions were available. Based on the previous findings of Terry et al. (1999, 2003a), we hypothesized positive relationships between the negatively-valenced ITAMS scales, the Depression scale of the DASS-21, and the PANAS Negative Affect (NA) scale. Moreover, we hypothesized that the ITAMS Vigor scale would correlate negatively with the NA scale and the DASS-21 Depression scale, and positively with the PANAS Positive Affect (PA) scale.

We assessed the validity and internal consistency of the ITAMS by replicating the BRUMS validation process used by Terry et al. (1999, 2003a). Results indicated very good internal consistency of the six factors and excellent factorial validity. Factor inter-correlations were consistent with previous findings (Terry et al., 1999). As hypothesized, the ITAMS Depression scale was highly correlated with the other negative mood dimensions, while the ITAMS Vigor scale was inversely correlated with the five negative factors. ITAMS subscale means were comparable to those reported for adult athletes using the BRUMS (Terry and Lane, 2010). Concurrent validity of the ITAMS was supported via hypothesized relationships with the Italian versions of the PANAS (Terracciano et al., 2003) and the Depression scale of the DASS-21 (Bottesi et al., 2015). The five negatively-valenced ITAMS scales correlated with the NA and DASS-Depression scales, and were inversely correlated with the PA scale. The ITAMS Vigor scale was correlated with PA and inversely correlated with NA and DASS-Depression scales.

The measurement model of the ITAMS was confirmed via a multigroup analysis that confirmed factorial invariance across males and females. Fit statistics were very strong in the male, female, and combined samples. Although small gender differences in factor means were identified, they were not of sufficient magnitude to create model misfit, confirming that males and females conceptualize the mood construct similarly. Despite this conceptual consistency, there were differences in mood responses based on the gender and age of participants. Female athletes reported higher levels of fatigue, tension, depression (on the DASS-21), and lower levels of vigor and positive affect than their male counterparts. These results align with previous work by Terry et al. (2003b, 2012) and Rajkovic (2014) who found gender differences in mood responses among South African, Iranian, and Serbian samples. Moreover, older athletes reported lower confusion, depression, fatigue and tension scores, and higher vigor and PA scores than younger athletes, supporting previous findings (Zhang et al., 2014). Based on the comparability of responses to the BRUMS in paper-and-pencil and online versions (Terry and Lim, 2011), it is reasonable to assume that the ITAMS, although only administered online in our study, would show similar psychometric properties if administered in paper-and-pencil form. Future investigations should test this assumption.

Despite the demonstrated psychometric strength of the ITAMS, further exploration of its psychometric characteristics is necessary. First, although the ITAMS measurement model was supported, no analysis of its predictive validity was included in our study. Therefore, investigation of the relationship between mood and performance in Italian contexts is warranted to evaluate the predictive validity of the ITAMS. Second, test-retest reliability was not addressed in our study and hence further investigation is required to establish how well the ITAMS performs over multiple assessments. Given this lack of information about the psychometric consistency of the ITAMS over time, it is prudent to caution researchers and applied practitioners about drawing definitive conclusions from multiple assessments using the ITAMS until additional psychometric scrutiny has been completed. Third, due to the wide range of applications for mood scales, further investigation of the psychometric characteristics of the ITAMS among other populations of interest (e.g., youth, seniors, general population) and in other situational contexts (e.g., academic, business, medical, military) would be advantageous. Such research efforts can be informative in assessing the need for group-specific tables of normative data or whether the preliminary norms from the present study are equally relevant beyond the world of sport and exercise.

The ITAMS has a wide range of potential applications from both research and applied perspectives. From a research perspective, development of the ITAMS provides those researching in an Italian-language context with a mechanism for testing the central tenets of Lane and Terry's (2000) conceptual model of mood-performance relationships, and for replicating investigations of the predictive effective of mood assessments in sports such as karate (Terry and Slade, 1995) and swimming (Terry et al., 2004), The ITAMS also facilitates investigation of whether specific mood profiles, including the classic “iceberg” profile (Morgan, 1980) and more recently-identified profiles, such as the “inverse Everest,” “surface,” “submerged,” and “shark fin” profiles (Parsons-Smith et al., 2014), are evident among Italian speakers.

Other worthwhile lines of research include evaluating the efficacy of the ITAMS as part of medical screening protocols (Galambos et al., 2005), to assess risk of mental health issues (Gould et al., 2005; van Wijk et al., 2013), monitor cardiac rehabilitation patients (Sties et al., 2014), or as a quality of life index (Shin et al., 2016). Given the brevity of the ITAMS, it may serve a valuable role in mood research conducted in settings with limited time availability for data collection, such as prior to or between sport events. Further investigations of mood responses and mood regulation strategies among exercise participants, and comparisons between athletes and exercisers, would also appear to be salient.

From an applied perspective, the multiple uses for mood profiling in elite sport proposed by Terry (1995) are equally relevant in Italian-language contexts. Proposed uses for mood profiling include monitoring athlete responses to acclimatization and injury, as an indicator of general well-being, and as a catalyst for discussion. ITAMS assessments could also be used to gauge the effectiveness of mood regulation interventions (Lane et al., 2011) or as a self-monitoring tool to reduce risk of overtraining episodes (Lovell, 2004; Rohlfs et al., 2008). Given the global growth in the sport psychology profession (Quartiroli et al., 2014) and the associated need for practitioners to work with athletes from varied cultures (Vosloo and Quartiroli, 2014), cross-cultural validation of BRUMS translations such as the ITAMS help to support the delivery of sport psychology services across borders. The linguistic and cultural validation of the ITAMS is consistent with the quest for greater cultural sensitivity in applied sport psychology (Schinke and Hanrahan, 2009). Terry (2009), for example, described how belief systems engrained in cultural settings can play a significant role in how emotions are expressed and regulated. It is therefore important to further explore cultural influences that play a role in the experience, expression, regulation, and self-report of mood responses (Lan et al., 2012). Overall, our findings support the psychometric integrity of the ITAMS for use in Italian-language contexts.

AQ, PT, and GF: contributed equally to the development and completion of the manuscript. AQ and PT: co-led the project and the manuscript development. GF: led the statistical analysis and contributed to the manuscript development.