Personality traits and select socio-demographic variables as predictors of military morale: longitudinal research in the Estonian defence forces

Military morale has been considered one of the most important contributors to improved performance in the military environment (Manning 1991; Ivey 2014; Fennell 2014; Britt and Dickinson 2006; Reed et al. 2011: 203). Throughout history, both practitioners and researchers have sought to define and conceptualize the phenomenon of military morale. However, what it is or how it is related to individual or collective performance is still not fully clear. Generally, the term “morale” has been used together with other constructs that have a similar meaning (see e.g. Hardy 2009: 24; Ingraham and Manning 1981; Peterson et al. 2008; Shamir et al. 2000).

The most widely used definition of military morale has been offered by Manning (1991: 455) as “the enthusiasm and persistence with which a member of a group engages in the prescribed activities of that group”. Britt and Dickinson (2006: 162) elaborate it further as “a service member's level of motivation and enthusiasm for accomplishing mission objectives”, arguing that their definition captures the elements of military morale better, especially in the case of individuals participating in military operations. Either Manning's definition has been used for military morale, as by van Boxmeer et al. (2007) and Jonsson et al. (2010), or Britt and Dickinson's definition has been used, e.g. by Dyches et al. (2017) and Britt et al. (2007).

The authors of this study propose a definition and model of military morale as a variable that gives a service member energy that drives him/her to perform qualitatively better in stressful conditions and that which is characterized by enthusiasm and persistence with which individual unit members engage in the tasks assigned to the unit as a whole/collective (Kasemaa and Säälik 2021). In this context, a unit could be a team, squad, platoon, company or any other military equivalent. Consequently, we approach military morale as both an individual-level and a collective (i.e. unit-level)-level construct, i.e. as a variable that describes the individuals within the social context and sees units as social entities.

Some previous studies have attempted to associate morale with certain psychological constructs, such as individual personality patterns (Herrera 2012), or with positive and negative affectivity of various aspects of morale (e.g. work engagement and burnout) (Kasemaa and Säälik 2021; Steger et al. 2013). This study seeks to answer whether directly measured morale could be predicted by individual characteristics, such as personality traits or socio-demographic factors, to provide a better understanding of why some service members are more likely to be enthusiastic about and more engaged in their work-related activities. Furthermore, it might contribute to the development of interventions to promote military morale and improve performance as a result.

Overall, the aim of this study is to explore the role of individual characteristics (i.e. personality traits and socio-demographic variables) in predicting military morale, at both the individual and collective levels. Consequently, the principal research question seeks to ascertain whether personality or socio-demographic characteristics contribute to the fluctuation of military morale and to what extent.

The definition and conceptualization of morale as an individual or collective psychological phenomenon has gone through several iterations over time, studied in both the military context and outside it (Gal 1986; Motowidlo and Borman 1977, 1978; Britt et al. 2007). Most commonly, morale has been defined by enthusiasm (Britt et al. 2007; Cotton and Hart 2003; Manning 1991; Shamir et al. 2000), general feelings of satisfaction (Motowidlo and Borman 1977; Manning 1994; Hardy 2009) and motivation to achieve group objectives (Motowidlo and Borman 1977; Britt and Dickinson 2006); or through well-being based on self- and group confidence (Gal 1986) or self- and/or collective efficacy (Hardy 2009); but also through the cognitive, emotional and motivational stance towards group goals and tasks (Peterson et al. 2008). Usually, it comprises such characteristics as confidence, optimism, enthusiasm and loyalty, as well as a sense of common purpose. Thus, almost all definitions of morale incorporate several self-determined psychological elements related to feelings or state of mind, e.g. energy and enthusiasm, but also factors related to goals and groups.

Britt and Dickinson (2006) examined military morale primarily as a unidimensional construct, while van Boxmeer et al. (2007) used a multidimensional conceptualization. Furthermore, van Boxmeer et al. (2007) argued that the elements of military morale, such as enthusiasm and persistence, might be very close to the key elements of work engagement (based on Maslach and Leiter's concept of work engagement, comprising energy and involvement) and its opposite – burnout, characterized primarily by exhaustion and cynicism (Maslach and Leiter 1997). This approach has subsequently been used in several empirical studies, e.g. by Chambel et al. (2015), Ivey et al. (2015) and van Boxmeer et al. (2011), and forms the basis for the conceptual model put forward by the authors of this study as well.

Generally, military morale has been found to be positively associated with several psychological and attitudinal variables that contribute to organizational success, such as well-being (Tucker et al. 2005), job satisfaction (Motowidlo and Borman 1978; Tucker et al. 2005), small unit cohesion (Griffith 1988) and collective efficacy (Jex and Bliese 1999), to name a few. Moreover, military morale has been shown to mitigate the effect of job-related stress on performance (Milgram et al. 1989; Britt et al. 2013). Furthermore, morale has been found to be negatively associated with the symptoms of post-traumatic stress disorder (Whitesell and Owens 2012; Dyches et al. 2017). Therefore, it is important to explore how the personal factors of each individual member of an organization affect the overall level of organizational morale.

Whether the phenomenon of morale is primarily an individual- or a group-level characteristic is still under debate. Some studies have found morale to be related to both individuals and groups (Ivey 2014: 9), while others have underlined the role of individuals within those groups (Peterson et al. 2008; Hardy 2009; Britt and Oliver 2013: 49). Considering our definition of military morale as something that gives service members energy and that is characterized by enthusiasm and persistence with which individual members engage in the tasks assigned to their unit, we might conclude that it is primarily an individual-level construct. However, when individual members believe that their unit is able to manage collectively in stressful situations, it could be a contributor to military morale. This assumption is supported by previous research (Shamir et al. 2000; Chen and Bliese 2002; Britt and Dickinson 2006: 172) and, therefore, it is warranted that some elements of morale should be considered at the unit level as well. One possible approach is to look at unit morale as an aggregation of the individual morale of each unit member. This approach is supported by such authors as Ivey (2014: 14) and Euwema et al. (2009).

An individual's personality can be defined as something that makes that person unique, setting them apart from others, ranging from Allport's definition as the psychophysical systems creating a person's characteristic patterns of behaviour, thoughts and feelings (1961: 28); to Child's construct as the more-or-less stable internal factors making one person's behaviour consistent from one time to another and different from the behaviour that other people might manifest in comparable situations (1968: 83); to Funder's suggestion as the individual differences in characteristic patterns of thinking, feeling and behaving, together with the psychological mechanisms (hidden or not) behind those patterns (2001: 2).

Most commonly in use is, however, McCrae and Costa's concept of personality as the way in which people perceive and interpret events, thoughts and feelings in their life (2008: 160). Drawing on those examples, the central idea of a uniquely characteristic pattern of thoughts, feelings and behaviours as someone's more-or-less constant and, therefore, predictable trait has been most recently formulated through perception and interpretation of life events, which leads us to assume that if we can identify measurable personality traits, we should also be able to predict other types of psychological perceptions of life events (e.g. perceived well-being, level of morale, and so on).

The five-factor model of personality, better known as the Big Five, with its acronym OCEAN, describes personality as the framework through which people perceive and interpret events, thoughts and feelings in their life, and it is made up of the following factors: Openness to experience (Ope), Conscientiousness (Con), Extraversion (Ext), Agreeableness (Agr) and Neuroticism (Neu) (Costa and McCrae 1995; McCrae and Costa 2008). Previous studies have found that personality dimensions are closely linked to a variety of organizational variables, such as job satisfaction (Judge et al. 2002), commitment (Erdheim et al. 2006) or counterproductive behaviour (Dalal 2005; Berry et al. 2007). Sinclair and Tucker (2006: 210–211) argue that the role of personality is more important for occupations involving higher intensity of interpersonal interactions, as in the case of the military.

Ope means that the person tends to be curious and creative and needs variety and independence; the person also tends to be quite open to feelings, actions, ideas and values (Costa and McCrae 1992; McCrae and Costa 2008). Ope has been found to have a strong correlation with contextual performance as compared to task performance (Chiaburu et al. 2011), as well as a positive correlation with work engagement (Akhtar et al. 2015; Morgan and de Bruin 2010), whereas it appeared to have a non-significant correlation with burnout (Kim et al. 2007).

Con as a personality trait means that the person is dutiful, self-disciplined, follows rules and ethical principles and tends to be focussed on job involvement (Costa and McCrae 1992; McCrae and Costa 2008). Con has been found to be correlated with, or even be able to predict, both task and contextual performances across most occupations (Sinclair and Tucker 2006; Chiaburu et al. 2011) as well as to work engagement (Bakker et al. 2012).

Ext refers to active, gregarious and excitement-seeking people who exhibit high levels of enthusiasm and optimism (Costa and McCrae 1992; McCrae and Costa 2008), with positive relationship to task and contextual performances (Chiaburu et al. 2011) and work engagement (Akhtar et al. 2015; Inceoglu and Warr 2011), while having a negative correlation with exhaustion (Gustafsson et al. 2009; Kim et al. 2007).

Agr means that the person is altruistic, trusting and trustworthy, sensitive towards the needs of others and therefore values a cooperative work environment or tends to develop positive relations with co-workers relatively easily (Costa and McCrae 1992; McCrae and Costa 2008). Agr has shown positive correlation with contextual performance (Ilies et al. 2009; Chiaburu et al. 2011) and with work engagement (Akhtar et al. 2015; Morgan and de Bruin 2010).

Neu is a personality trait that represents a tendency to experience negative emotions. As a result, these people tend to be anxious, depressed, fearful, impulsive and vulnerable (Costa and McCrae 1992; McCrae and Costa 2008). Neu has been shown to have a negative correlation with task and contextual performances (Chiaburu et al. 2011) and a positive one with exhaustion and cynicism (Bühler and Land 2003; Kim et al. 2007; Langelaan et al. 2006).

Thus, while most personality traits are associated with positive emotions and striving towards better performance, we can also see that neurotic people tend to focus on the negative aspects of situations. Moreover, they might also ignore or downplay the feelings of engagement due to their tendency to become easily frustrated and evaluate their personal competencies as well as future prospects negatively, as revealed by research conducted by Kim et al. (2009) and Bakker et al. (2002), who concluded that Neu is the primary personality trait predicting burnout at work.

Based on the aforementioned studies, it could be assumed that the Big Five personality traits are well suited for describing the level of morale. Based on previous findings, we posit the following hypotheses regarding the effect of the Big Five personality traits on military morale (H1): morale is positively correlated with Con, Ext, Agr and Ope (H1.1); and, morale is negatively correlated with Neu (H1.2).

In addition, previous studies have demonstrated that the Big Five personality traits are able to explain a significant part of subjective well-being (Ryff 1989). Therefore, the similarities between well-being and morale and the differences between individual and collective types of morale (i.e. the first addresses the level of personal morale, while the other portrays the morale of one's fellow unit members or unit-level morale) are considered. Additionally, we might also hypothesize that different personality traits are more strongly related with either individual or collective morale. For example, people who value social relationships (i.e. exhibit a high level of Agr) or those who are happy to express themselves in social contexts (i.e. a high level of Ext) might rate collective morale higher than individual morale, whereas individuals who tend to get anxious quite easily (a high level of Neu) or who value order and discipline (a high level of Con) might rate individual morale more highly. Therefore, we posit that Agr and Ext are more strongly correlated with collective morale (H2.1), while Con and Neu are more strongly correlated with individual morale (H2.2).

Only a few studies that explore military morale over a longer period of time exist, despite the fact that the concept of military morale and its fluctuation has captured the interest of both researchers and practitioners (Britt and Dickinson 2006). For example, Britt et al. (2007) conducted longitudinal research in Kosovo, focussing on long-term associations between morale and depression. Another example that suits our conceptualization of morale is the research conducted by Cremers (2018), studying morale in the Dutch army in the context of deployment. Cremers’ findings indicate that, in general, the positive components of morale, such as vigour and dedication, were higher before deployment as compared to during deployment, whereas the opposite was true for burnout (i.e. cynicism and exhaustion) (Cremers 2018).

Furthermore, the research on small unit cohesion suggests that during conscription, military morale may undergo a U-shaped progression. This hypothesis is supported by research conducted by Vaitkus and Griffith (1990), Siebold (1999) and Siebold (2006), who argue that it takes time for unit cohesion to develop, which tends to drop after a certain period of time, and that after a while, it starts increasing again but does not reach the same level as before the drop. Contrarily, Bartone and Adler (1999), as well as Maugen and Litz (2006), have found that the development of unit cohesion follows a reverse-U shape. The research conducted outside the military context seems to indicate that the perception of work engagement and burnout might be quite stable (see e.g. Xanthopoulou et al. 2009; Sonnenchein 2007: 14), although influenced by several organizational variables, such as supportive colleagues, self-efficacy, job autonomy, etc. (see e.g. Salanova et al. 2006; Xanthopoulou et al. 2009; Houkes et al. 2011). Therefore, the authors of this paper argue that military morale is not just an isolated and stable phenomenon but should instead be seen as a dynamic process that changes over time.

So far, there have not been many studies focussing on morale in the Estonian Defence Forces (EDF) and related organizations (Parmak 2010; Kasemaa and Säälik 2021). Beyond Estonia, the majority of research has focussed on professional armies and not on conscription-based ones (see e.g. Ivey et al. 2015; van Boxmeer et al. 2007; Britt et al. 2007, etc.), and therefore, we do not have much insights regarding the changes of military morale over longer periods of time in such compulsory service context. However, the specific context of the Estonian conscription service might lead to the assumption that morale is quite high in the beginning of the service, especially after soldiers finish their basic training. Drawing on research on work engagement and burnout (Sonnentag et al. 2010), we posit that there might be periods during which conscripts experience higher or lower morale. For example, conscripts probably feel quite motivated and efficient when they enter conscript service, and therefore, their engagement should be quite high and, despite the new stressful context, the possibility of burnout is not very likely due to the rather short time of military service. Further on, it could be that in the middle of the military service, the energy, motivation and enthusiasm will drop, because military life is not that exciting anymore, and, as a result, the conscripts become gradually bored. Finally, at the end of their service, the conscripts take part in the final exercise, which might mitigate the effects of burnout and boost positive emotions. This leads to our third hypothesis H3: military morale is high at the beginning, drops to its lowest level in the middle and returns to a higher level at the end of military service.

There have been studies analysing correlations between work engagement/burnout and socio-demographic variables (e.g. Suzuki et al. 2015; Ahola et al. 2006). Drawing on the studies that have conceptualized military morale through work engagement and burnout (e.g. van Boxmeer et al. 2007; Ivey et al. 2015), we estimate that there may also be an association between military morale and certain socio-demographic variables. Specifically, we are interested in exploring whether socio-demographic factors, such as age, position, education, mother tongue, place of residence and university student status, affect military morale and to what extent they might predict morale.

The differences among age groups in studies on morale show that younger employees tend to experience higher levels of burnout (Patrick and Lavery 2007; Garrosa et al. 2008) and register lower levels of work engagement (James et al. 2011). In addition, Maugen and Litz (2006), using a single-item approach, found a positive correlation between age and morale; their findings have subsequently been confirmed by Whitesell and Owens (2012), as well as Farley and Veitch (2003). Consequently, we deduce that older individuals who probably have better social skills and coping mechanisms might exhibit higher morale as well.

Second, some studies have found that work status (i.e. temporary versus permanent positions) has a significant impact on burnout (Garrosa et al. 2008). Moreover, Schaufeli and Bakker (2004) found that managers seem to experience higher work engagement than employees (Kim et al. 2009). Maugen and Litz (2006) found a positive correlation between military rank and military morale (measured by a single-item approach). Additionally, Halverson and Bliese (1996), whose research focussed on morale during peacekeeping operations, also found differences between ranks. In the context of the EDF, although we focus only on conscription service, we expect to encounter differences in morale across different positions.

Third, studies that have analysed the role of education have also found correlations with work engagement and burnout (Sharma et al. 2017). For example, Yildiz Durak and Saritepeci (2019) found that occupational burnout increased for people with higher levels of education; similarly, Demus (2018) also studied well-being among conscripts and obtained the same results. However, a few studies have found education to have no significant effect on work engagement (Geldenhuys and Henn 2017; Mahboubi et al. 2014) or attitudes towards conscription service (Müürsoo 2018). We might interpret these differing findings as follows: conscripts with higher levels of education are usually older and may have set their path of life as more stable already and, therefore, may not be that motivated to enter mandatory conscription service.

Fourth, as regards research on military morale in the EDF, one of the previous studies by Parmak (2010) reported differences between morale/burnout and ethnic background among conscripts, indicating that conscripts with mother tongue other than Estonian exhibited lower levels of morale at the beginning of conscript service; however, over time, their morale showed signs of increasing morale. Moreover, Müürsoo (2018) found significant differences in attitudes towards conscription service among different language groups in EDF.

In addition to the aforementioned socio-demographic characteristics (i.e. age, service position, education and ethnicity), we decided to also include place of residence and university student status into the analyses. In this context, although we found some pieces of information in previous studies explaining, for instance, the associations between place of residence and life satisfaction as reported by Switek (2012), in general, there is little information or insufficient reliable findings on the interrelationships between military morale and these additional socio-demographic variables.

This study examined the perceptions of individual and collective types of military morale among conscripts of the EDF and sought to ascertain how these different types of morale are affected by individual characteristics, such as the Big Five personality traits or socio-demographic variables. The interrelationships (especially with personality traits) established in the course of our research draw on the assumption that personality affects individual interpretations of and reactions to the environment, as has been suggested by several other authors (McCrae and Costa 2008: 162–163; Higgins and Scholer 2008: 184; Mischel and Shoda 2008: 209). In addition, the cognitive-affective theory of personality developed by Mischel and Shoda (2008) also posits that personality may influence how individuals encode and evaluate information coming from their environment.

Our hypotheses predicted that military morale is positively correlated with the following Big Five personality traits: Con, Ext, Agr and Ope (H1.1) and negatively correlated with Neu (H1.2). These hypotheses were confirmed partially, i.e. the predicted relationships were confirmed only for the first stage of conscript service, when all personality traits exhibited statistically significant relationships with the perception of individual morale. However, across all Big Five traits, only Con demonstrated significant correlation with individual morale throughout the whole conscription period, supported by Ext up to the middle stage of military service. On the other hand, the hypotheses posited with regard to collective morale yielded only low correlations, that too only with Ext and Agr (Time 1). During the middle stage of conscript service, both individual and collective types of morale showed significant correlations with Con and Ext, whereas Neu demonstrated correlation only with the collective morale. Con and Ext maintained their significant correlation with collective morale until the end of military service. Overall, this pattern of interrelationships was also supported by the results of logistic regression analysis, which demonstrated that personality traits such as Con, Ext and Neu predict individual morale statistically significantly in the beginning of the service, whereas only Agr seemed to bear an effect on collective morale.

Conscientious people, dutiful and self-disciplined (McCrae and Costa 2008), tend to rate their individual morale as very high. However, because conscript service is mandatory in Estonia, there are probably people whose self-discipline is actually lower and who are not that dutiful, which might explain why Conscientious people rated their own morale higher than that of their unit members. Surprisingly, Extraverted people's sense of enthusiasm and energy (Costa and McCrae 1992) to engage in unit activities seemed to change over time, from being more closely related to individual morale in the beginning to demonstrating a stronger correlation with collective morale in the middle stage and at the end of service. At the same time, conscripts who tend to experience anxiety, depression and other negative emotions (McCrae and Costa 2008) seemed to rate their individual morale less highly, especially at the beginning of military service. Perhaps, this could be explained by the fact that either conscripts are more prone to experiencing the negative aspects of military service primarily at the start and not so much at the end of their service or there are coping mechanisms that conscripts might have developed during the course of military training.

Another hypothesis – Agr and Ext are more strongly correlated with collective rather than individual morale (H2.1) – was confirmed by the absolute values of the correlations at different time points. However, Agr did not show any differences in correlations with either individual or collective morale and, thus, Agr does not support our hypothesis. Nevertheless, this is an interesting result, demonstrating that people with high Agr and, therefore, better suited for teamwork and cooperation (McCrae and Costa 2008), did not register a high level of collective morale. This result could be explained by the fact that the nature of military service does not support cooperation as such, and as a result, the potential positive effect is not amplified by the quality of the aspects that people with high Agr might value. On the other hand, Ext demonstrated mixed results: in the beginning of conscript service, Ext was more strongly correlated with individual morale than with collective morale, whereas in the middle stage of military service, the correlation was reversed. This could be explained by the fact that unit-level training is not in focus until the middle of service, and therefore Ext – as a coping mechanism – tends to support the feeling of individual morale and not collective morale. However, the results indicate that this tendency changes during the course of service. For example, in the middle stage of military service, Ext tends to support primarily collective, and not individual, morale.

Hypothesis H2.2 – Con and Neu are more strongly correlated with individual than collective morale – was proven in the case of Con throughout the whole service cycle, whereas Neu showed a stronger correlation with individual morale only at the beginning of military service. The latter might be explained by the change of focus to more unit-based activities during the course of the training cycle, as already mentioned before.

The hypothesis about military morale fluctuation over time was drawn from previous research on small-unit cohesion by Vaitkus and Griffith (1990) and Siebold (1999, 2006), as well as from research on work engagement and burnout by Sonnentag et al. (2010). This hypothesis was confirmed, and the change was found to be statistically significant. Thus, we can say that over the course of compulsory military service, conscripts’ individual morale as well as that of the unit was affected. For example, in the beginning of service, right after completing the basic course, the feeling of enthusiasm and persistence registered was quite high, but then reduced drastically. Previously, the EDF's annual conscript survey found that the middle stage of conscript service is often perceived as ‘a long and boring, a tedious and pointless period’ (e.g. Demus 2018). In the EDF, compulsory military service ends with a final practical exercise, which is believed to provide a renewed opportunity to experience something close to a real military situation, eliciting excitement, as well as an increased sense of cohesion and camaraderie within the unit.

The effect of socio-demographic factors (e.g. age, position, education, mother tongue, place of residence and status of university student) on military morale, as well as their predictive capacity, was analysed by binary logistic regression. Generally, socio-demographic variables did not seem to play a significant role in explaining the variance of either individual or collective military morale. Altogether, university student status, the level of education, age, position and place of residence were shown to be statistically significant for both individual and collective types of morale; however, the effect sizes were quite small (1%–2% for individual and 2%–4% for collective morale), demonstrating that an individual's perception of morale is not dependent on those variables. In order to show the similarity of these results with those from previous research, we refer to the concept of work engagement and burnout (Maslach and Leiter 1997) as an analogue, because morale has sometimes been explained using the components of these two factors (van Boxmeer et al. 2007; Kasemaa and Säälik 2021). For example, Haley et al. (2013) found that age may have a positive correlation with morale, whereas Aguayo et al. (2017), in their study of police officers, did not find age or sex to have any significant impact on burnout. Additionally, Akhtar et al. (2015) found that age is a statistically significant predictor of work engagement. Overall, there were no other equivalent reliable research results available on the issue of socio-demographic characteristics explaining the level of morale that could be compared to the findings of this study.

The quest to ascertain the factors that contribute to individual or collective performance (i.e. the factors that explain the secrets of group success, on the one hand, and those that mitigate the devastating effects of burnout, on the other hand) has led to the understanding that there are several interrelated psychological and organizational phenomena that describe and explain such processes, for example high level of morale contributes to success and satisfaction, while the opposite (i.e. low level of morale) has a detrimental effect.

This study focussed on the interrelationships among morale, the Big Five personality traits and socio-demographic factors in the military context. The research was carried out through measurement of direct morale and by administering the Big Five personality questionnaire. The results indicate that, in the context of military conscription, personality traits appear to correlate with, and may even predict, the level of perceived individual or collective morale. Of the Big Five personality traits, conscientiousness and extraversion, and to some extent – also neuroticism, were shown to exhibit a more persistent relationship to the conscripts’ perception of both individual and collective types of morale.

In addition, this longitudinal study also helped to highlight the fluctuating nature of military morale across the different stages of the conscription service cycle, i.e. the perceptions of morale are the highest at the start of military service, drop in the middle, and rise back up at the end of the service period. Furthermore, it seems that socio-demographic factors have a relatively modest effect on the conscripts’ perception of their individual morale or the collective morale of their unit.