1. Introduction
Moods are pervasive to human functioning and deeply influence how individuals interact with the world around them. A mood has been defined as “a set of feelings, ephemeral in nature, varying in intensity and duration, and usually involving more than one emotion” [
1]. Moods are conceptualized as having a valence dimension that varies from positive (e.g., happy) to negative (e.g., depressed) and an arousal dimension that varies from activation (e.g., alert) to deactivation (e.g., tired) [
2,
3]. Moods differ from emotions in several ways; for example, compared to emotions, moods are more diffuse, of lesser intensity, longer duration, and not related to a specific cause [
4,
5]. Although moods often fluctuate from day to day, persistent and/or extreme negative moods represent increased risk of mental health disorders [
6]. The 2019 Global Burden of Disease Study estimated that 970 million people, or 12.6% of the global population, live with a mental health disorder, most commonly depression and anxiety [
7]. The very high global prevalence of mental health issues makes the study of risk indicators, such as mood, of great importance.
Moods have been shown to vary according to gender, with men tending to report higher vigor and lower anger, confusion, depression, fatigue, and tension than women [
8,
9]. There are many sociopsychological reasons why women would tend to report more negative moods than men, especially related to the disadvantage women face in many aspects of life, including education, family responsibilities, and careers [
10,
11]. Gender inequity is evident in Lithuania, which ranked 20 of 27 European Union (EU) countries on the 2021 Gender Equality Index [
12], providing a compelling rationale for examining gender differences in mood in our study.
Age has also been implicated in mood differences, wherein reported moods among adults tend to be more positive with increasing age [
7,
8]. Differential use of emotion-regulation strategies may partially explain age-related variations in mood. For example, compared to older adults, younger adults tend to engage more in maladaptive coping strategies, such as rumination, avoidance, and suppression [
13]. Similarly, mindfulness facilitates effective emotion regulation [
14] and psychological wellbeing [
15], and older adults are more likely to be classified into high mindfulness groups than their younger counterparts [
16]. Hence, it is appropriate to explore the influence of age on mood in the present study.
A process known as mood profiling, in which scores on a mood scale are plotted against normative scores to create a graphical profile, has a long history of use in identifying common patterns in mood states [
17,
18]. The Brunel Mood Scale (BRUMS), originally developed in England for use with male and female adolescents and adult athletes and nonathletes, is frequently used to assess mood in research and applied settings due to its short administration time (2–3 min.) and acceptable psychometric characteristics [
19,
20]. For example, during validation studies, the factorial, convergent and divergent validity of the BRUMS and the invariance of its measurement model was demonstrated among four samples of participants (
n = 2,549), independently and via multi-sample analysis [
20].
The 24-item BRUMS was developed as a shortened version of the 65-item Profile of Mood States (POMS) [
21,
22] to be suitable for adolescents, as well as adults. Hence, items for consideration were screened for comprehensibility by adolescents aged 12 and over, prior to inclusion in an initial list of 42 items, which was reduced to 24 items, following tests of factorial validity (see [
19,
20] for further details of the validation process). Compared to the POMS, the BRUMS has a shorter completion time (2–3 min vs. 7–10 min), is suitable for adolescents, as well as adults, and does not include mood descriptors, such as “blue” and “bushed”, that may not be understood in all cultural contexts [
19].
The POMS was originally developed for use with psychiatric outpatients [
21] and is, hence, oriented towards negative rather than positive moods, with six subscales representing the mood clusters of Anger-Hostility, Confusion-Bewilderment, Depression-Dejection, Fatigue-Inertia, Tension-Anxiety and Vigor-Activity [
21]. The POMS has been criticized for its negative orientation and for providing a limited assessment of mood rather than a comprehensive measure of the mood construct [
23], although it has been shown to have utility in screening for mental health issues [
24] and predicting outcomes in competitive domains, especially sport, where negative moods can debilitate performance [
18]. In terms of screening for mental health issues, negative mood is a risk factor for several psychopathologies [
6]. Morgan’s mental health model [
24], which is based on the POMS, posits that a mood profile of high Vigor combined with low Anger, Confusion, Depression, Fatigue, and Tension is indicative of positive mental health, whereas a mood profile of low Vigor combined with high Anger, Confusion, Depression, Fatigue, and Tension is associated with poor mental health outcomes.
In a mental health context, the BRUMS has been used to help manage performance anxiety and prevent injuries among adolescent ballet dancers in Japan [
25]; screen for risk of post-traumatic stress disorder among military personnel in South Africa [
26]; assess adolescents for elevated suicide risk in the USA [
27]; and evaluate population-level mental health and monitor the psychological wellbeing of cardiac rehabilitation patients in Brazil [
28,
29]. Related to use of mood profiling for predicting performance, Lane and Terry [
1] presented a conceptual model that emphasized the interactive effects of different mood dimensions on performance, particularly emphasizing how depressed mood interacts with anger and tension. For example, in the absence of any symptoms of depressed mood, anger has been shown to associate with good performance [
30,
31], whereas anger associates with poor performance when symptoms of depressed mood are evident [
18]. Evidence also exists that negative mood can have positive effects on some types of performance, such as increasing cognitive and emotional creativity via self-focused attention [
32].
Using mood profiling, several distinct patterns of mood scores have been identified. For example, the iceberg profile, which is characterized by a high Vigor score combined with low scores for Tension, Depression, Anger, Fatigue, and Confusion, typically associates with positive mental health and good athletic performance [
18,
24,
33,
34]. The iceberg profile was so named because the shape of the profile is reminiscent of an iceberg, in that most of the scores sit below the waterline (i.e., the mean standardized score of 50), with only the vigor score above the surface, in the same way that most of a real iceberg sits below the surface. The inverse iceberg profile, which is characterized by a below average Vigor score, combined with above average Tension, Depression, Anger, Fatigue, and Confusion scores, typically associates with increased risk of pathogenesis and underperformance [
35,
36]. Similarly, the inverse Everest profile, the most negative profile characterized by a low Vigor score, high scores for Tension and Fatigue, and very high scores for Depression, Anger, and Confusion indicates elevated risk of disorders, including post-traumatic stress disorder [
26].
The shark fin profile, which is characterized by below-average scores for Tension, Depression, Anger, Vigor, and Confusion, combined with very high Fatigue scores, may signal “an accident waiting to happen” and has been linked to athletic injury [
37] and poor adherence to safety procedures in high-risk vocations [
38]. The surface profile, which is characterized by average scores on all mood dimensions, can be considered to represent an average mood, and the submerged profile, which is characterized by below-average scores on all mood dimensions, may be beneficial in activities that place a premium on remaining calm and unemotional [
39]. These six mood profiles have been replicated in a variety of language and cultural contexts, including English [
8,
40,
41], Brazilian [
42], Chinese [
43], Italian [
44], and Singaporean populations [
45,
46].
Several validated translations of the BRUMS exist, including Afrikaans [
47], Brazilian Portuguese [
48], Chinese [
49], Czech [
50], French [
51], Hungarian [
52], Italian [
52,
53], Japanese [
54], Malay [
55,
56], Persian [
57], Serbian [
58], and Spanish [
59]. In most instances, the BRUMS measurement model was supported in translation [
47,
48,
50,
51,
52,
53,
54,
55,
56,
58], although in a few translations, some poor-fitting items were discarded [
49,
57,
59]. To date, no validated translation of the BRUMS into the Lithuanian language has been published. Preservation of the Lithuanian language and translation of English-language resources into Lithuanian are seen as significant imperatives for the country [
60]. Further, mental health issues are of particular concern in Lithuania. For example, subjective wellbeing is lower in Lithuanian society than the EU average [
61] and mental health resources, policies, and practices are perceived as being in urgent need of improvement in the country [
62,
63]. The availability of simple, quick, and effective risk indicators of mental health status, such as the BRUMS translated into the Lithuanian language, represents an important forward step.
The purpose of our study was to validate a Lithuanian-language version of the BRUMS to facilitate further mood research in a Lithuanian context. It was hypothesized that the 24-item, 6-factor measurement model of the BRUMS (Tension, Depression, Anger, Vigor, Fatigue, Confusion) would be supported in the translated version, as it has been in previous translations [
47,
48,
50,
51,
52,
53,
54,
55,
56,
58]. As well as evaluating the hypothesized measurement model of the translated scale (i.e., factorial validity), we also tested its convergent validity (i.e., whether subscale scores correlated highly with concurrent measures of similar constructs) and divergent validity (i.e., whether subscale scores showed minimal correlation with concurrent measures of dissimilar constructs) [
64]. More specifically, as reported previously, it was hypothesized that Tension, Depression, Anger, Fatigue, and Confusion scores would correlate highly with measures of Perceived Stress [
37,
65] and Neuroticism [
66,
67], the latter of which is described as the propensity to experience negative mood states [
67], and would correlate inversely, but to a lesser degree, with Agreeableness, which is described, in part, as being in control of anger and other negative emotions [
67]. It was also hypothesized that Vigor, characterized by being energetic and alert, would correlate, at least moderately, with Extraversion, characterized by talkativeness, sociability, and outgoingness, and show strong inverse correlations with Perceived Stress and Neuroticism [
66,
67]. Finally, it was hypothesized that mood scores would tend to show minimal correlation with measures of Conscientiousness, characterized by self-control, industriousness, responsibility, and reliability [
68], and with Openness, characterized by being imaginative, curious, and open-minded [
68]. Given that reported mood and the prevalence of mood disorders vary according to gender [
7,
8], we tested the invariance of the BRUMS measurement model of the Lithuanian translation across genders and compared the mood scores of men and women. We also assessed relationships between mood scores and age.
4. Discussion
The central purpose of our research was to validate the Lithuanian translation of the Brunel Mood Scale questionnaire [
19,
20], which we refer to as the BRUMS-LTU, among a sample of 746 participants who were fluent in Lithuanian. Results of our analyses supported the factorial validity of the measurement model and its configural, metric, scalar, and residual invariance across gender groups. It appears not only that the factor structure of the BRUMS-LTU is the same in men and women, but also that men and women understand the scale and the construct of mood in a similar way, and that the scale has similar error across genders. The convergent and divergent validity of the scale was adequately supported, as was the internal consistency of the six subscales. Overall, the psychometric properties of the BRUMS-LTU are very closely aligned to those of the original English-language BRUMS [
20]. Support for the psychometric integrity of the measure offers considerable potential for using the BRUMS-LTU in future research studies set in a Lithuanian context, including testing Morgan’s mental health model [
24] and Lane and Terry’s conceptual model of mood and performance [
1].
Another future research direction would be to interrogate Lithuanian data using cluster analysis for evidence of the six distinct mood profiles that have been identified previously in other language and cultural contexts; namely, the iceberg, inverse Everest, inverse iceberg, shark fin, submerged, and surface profiles [
38,
40,
41,
42,
43,
44,
45]. Given that the full range of possible scores (range = 0–16) was reported for all BRUMS-LTU subscales, with the single exception of Tension (range = 0–15), it is apparent that some participants reported extremely negative moods, which would indicate increased risk of mental health issues. More specifically, it would be anticipated that anyone reporting an inverse Everest profile, characterized by a low Vigor score, high scores for Tension and Fatigue, and very high scores for Depression, Anger, and Confusion, would be a candidate for follow-up assessment by a clinical psychologist or other health professional. Exploring use of the BRUMS-LTU as a mental health screening tool may be a fruitful avenue for future research in Lithuania, considering that the BRUMS has frequently been used for this purpose in English-speaking countries [
26,
27,
28,
37].
The benefits of regular physical exercise on psychological wellbeing, including the maintenance of positive mood and prevention of negative mood, is well documented [
93,
94] and has been shown to be especially pertinent in several countries during the COVID-19 pandemic [
95,
96]. Given that psychological wellbeing is below the EU average among the Lithuanian population [
62,
63], there is much scope for intervention studies that evaluate the impact of physical activity programs on wellbeing indicators, including mood, of participants. Further, several applications of mood profiling in the sport domain have been proposed, including to monitor emotional recovery from injury and responses to training load, to identify overtrained athletes, to predict athletic performance, and as a catalyst for discussion with a psychology professional (see [
97] for a review and [
98] for other applications). Prior to the current validation of the BRUMS-LTU, there was no validated mood scale available in the Lithuanian language, offering fertile ground for a wide variety of research questions and potential applications to be addressed in the sport and exercise domains.
Gender comparisons identified significant differences in mood scores in the present Lithuanian sample that are largely consistent with previous findings. The higher scores for Vigor and lower scores for Fatigue reported by men compared to women are consistent with the findings of several previous studies, although often, gender differences in mood scores extend to other mood dimensions, particularly Tension, Depression, and Confusion [
8,
9,
44,
45]. The relationship between mood scores and the age of participants was in the small effect range and, therefore, inconsistent with previous findings [
7,
8]. There is no obvious explanation for this lack of association and future research might investigate this relationship further in a Lithuanian context.
Although we provided preliminary evidence of acceptable psychometric characteristics of the BRUMS-LTU, further exploration of its psychometric integrity is recommended. The fit of the default measurement model was shown to be only just acceptable and was improved by adding covariance paths to the error terms for some items, as recommended by Byrne [
88], as a legitimate strategy to account for potential overlap in meaning between mood descriptors (e.g., anxious and worried). Further exploration of the measurement model is warranted, and additional psychometric properties of the BRUMS-LTU should be explored, including its test–retest reliability and predictive validity. Several other future research directions can also be identified. For example, to explore the lived experiences of Lithuanians during the COVID-19 pandemic, researchers could use the BRUMS-LTU together with retrospective pandemic experience questions (e.g., self-perceptions of how moods were affected by extended lockdowns or by personal experience of COVID-19, perhaps through contracting the disease personally or vicariously through relatives or close friends having caught the disease). Extending the validation of the BRUMS-LTU to youth participants would facilitate additional opportunities for impactful mood research, across many contexts, such as education and sport. Finally, once sufficient additional data are available from Lithuanian participants, tables of normative scores for the BRUMS-LTU should be generated to provide a reference point for interpreting data derived from the scale. Given the gender difference identified in the present study, separate norms for men and women may be required.
In common with all online questionnaire research, limitations were evident in our study. One limitation relates to our lack of knowledge of whether participants’ lived experiences during the COVID-19 pandemic influenced their moods, as has been shown in previous research [
9]. The observed lack of significant differences in mood scores during different phases of the pandemic suggests that mood did not fluctuate greatly during our data collection period. A second limitation concerns the requirement to access the questionnaire via the internet. Online questionnaires of any kind reduce access by marginalized and lower socio-economic groups. Thus, our BRUMS-LTU was validated on a predominantly university-educated sample, although it should be noted that 54% of the Lithuanian population hold a tertiary qualification, which is among the top 10 graduate rates in the world [
99]. A third limitation relates to the age of participants. Although the mean age of our sample (41.8 years) approximated the 2020 median age of 44.5 years for Lithuania as a whole [
100], we did not assess the moods of anyone under 17 years of age. A fourth limitation relates to the gender representation of our sample, which included 73.3% of participants who identified as women. Finally, although the translation process followed the recommended translation-back translation process [
101], this is seen as a minimum standard in good practice guidelines for test adaptation [
102], and the semantic and measurement equivalence of the translated BRUMS to the original English-language version needs further examination. Collectively, these limitations may restrict the ability to generalize our findings to the broader Lithuanian population.