Predicting army cadets’ performance: The role of character strengths, GPA and GMA

In a three-year bachelor's degree programme in military studies, the Norwegian Military Academy (NMA) aims to educate and develop highly skilled military officers so that they are able to cope with difficult and dangerous situations and execute leadership under extreme conditions. According to Cornum et al. (2011, p. 4), ‘modern warfare is characterised by demanding missions, extreme climates, sleep deprivation, cultural dissonance, physical fatigue, prolonged separation from family, and the ever-present threat of serious bodily injury or death’. One of the main challenges of military leadership is to be willing and able to make decisions that may result in the taking of lives and risking those of oneself and one's comrades in military operations. The Norwegian Chief of Defence states that military leadership is about doing those things that are uncomfortable and being able to cope with them, overcoming powerlessness, and avoiding emotional breakdown. Military leadership requires a certain degree of robustness that would facilitate clear and effective thinking and that would permit officers to cope with their feelings when facing complex and difficult situations (Forsvaret 2012).

What personal qualities characterise army cadets who succeed in their education and training to become military leaders well-suited for these demanding challenges?

Three different predictors are often examined in the studies of factors predicting academic achievement in undergraduate and graduate programmes: high-school grade point average (GPA; e.g. Allensworth and Clark 2020; Hiss and Franks 2014; Hodara and Cox 2016; Hodara and Lewis 2017; Wagerman and Funder 2007), general mental ability (GMA; e.g. Duckworth et al. 2019; Kobrin et al. 2008; Køber et al. 2017; Lounsbury et al. 2003; Roth et al. 2015; Schmidt and Hunter 1998; Strenze 2007) and Big Five personality traits – particularly conscientiousness (e.g. Chamorro-Premuzic and Furnham 2003; Conard 2006; Farsides and Woodfield 2003; Komarraju et al. 2009; Lounsbury et al. 2003; Poropat 2009).

This article will address two of these common predictors – GPA and GMA – and suggest another set of predictors that can be relevant for academic achievement, and particularly in military settings. Since the early 2000s, a growing number of studies indicate that certain character traits (e.g. character strengths, grit, hardiness, self-discipline and self-control) can also function as important predictors of both academic achievement and job performance (e.g. Duckworth and Seligman 2005; Harzer and Ruch 2014; Harzer et al. 2021; Lounsbury et al. 2009). Harzer et al. (2021) found, for example, that ‘perseverance, teamwork and leadership seemed to be especially relevant for numerous dimensions of job performance’, and that these three character strengths explained a substantial amount of variance in job performance even after controlling for GMA and personality. Lounsbury et al. (2009) found positive and significant correlations between college academic performance (AP) on the one hand, and Persistence, Open-Mindedness, Love of Learning, Self-Regulation and Prudence on the other.

Character strengths are suggested to be particularly relevant for performance in military settings, which often requires in extremis leadership (Kolditz 2010). A small, but growing number of studies have started examining the importance of character strengths for succeeding in military settings (e.g. Cosentino and Solano 2012; Duckworth et al. 2007; Gayton and Kehoe 2015a, 2015b; Gosnell et al. 2020; Kelly et al. 2014; Maddi et al. 2012; Matthews 2008; Matthews et al. 2006). Matthews (2008) claimed that ‘from the in extremis perspective, leadership under conditions of mortality salience, that is, conditions where risk of serious injury or death is present, may require different skills sets than leadership in the corporate world or on a sports team’. (p. 165). Building on these studies, we suggest that succeeding as a military officer requires something more than high GMA and good high school average grades. The extremely challenging conditions military officers face in live military operations including ‘sleep deprivation, high physical workload, stress, circadian disruption and fear of death or severe bodily harm’ (Laurence and Matthews 2012, p. 208) require strong character strengths that may not be reflected in GMA and GPA. Hence, we hypothesise that compared to GMA and GPA, character strengths will be able to function as better predictors for cadets’ AP and military performance (MP) during their three-year military training programme.

Character strengths can be defined as ‘positively valued trait-like individual differences with demonstrable generality across different situations and stability across time that manifest in the range of individuals’ thoughts, feelings, and behaviors’ (Harzer and Ruch 2014, p. 184). They are recognised and valued across cultures and religious traditions as morally valued behaviour patterns (Peterson and Seligman 2004).

There seems to be a lack of empirical studies examining the relationship between character strengths and cadet performance (academic and military) in military academies. One exception is the study by Cosentino and Castro Solano (2012), which found a positive association between a set of character strengths and AP or MP among cadets in the Argentinean military (Persistence, Leadership, Love of Learning, Perspective and Creativity), and a negative relationship for other character strengths (e.g. Teamwork and Fairness). Albeit significant, the correlations were in the low end of the scale, with no correlations higher that of r = 0.25.

The main purpose of this article is to examine to what extent certain character strengths predict MP and AP among army cadets in their bachelor's degree programme, and whether character strengths are better predictors compared to GPA and GMA.

Building on Peterson and Seligman's (2004) classification of 6 virtues and 24 character strengths (see Table 1), previous studies have suggested 12 character strengths as particularly important for succeeding as a military officer, and thus these are to be further developed in army cadets during their bachelor's degree programme (Boe 2016; Boe and Bang 2017; Boe et al. 2015a; Boe et al. 2015b). In 4 different studies, 173 experienced military officers were asked to rate each of Peterson and Seligman's 24 character strengths on a 5-point scale, ranging from 1 (not at all important) to 5 (very important), according to how important the respective character strengths were for military officers. To differentiate between important and less important character strengths, a cut-off score of 4.00 was used in the analyses. Twelve character strengths were given a score of 4.00 or higher by the military officers across the four different samples (Boe 2016). The 12 character strengths, ranked from highest (M = 4.90) to lowest score (M = 4.09) were, respectively, Leadership, Integrity, Open-Mindedness, Bravery, Teamwork, Persistence, Social Intelligence, Love of Learning, Fairness, Self-Regulation, Perspective and Creativity.

Since the set of character strengths selected for this study were identified to be particularly important for success as a military officer, we would argue that these 12 character strengths define the ‘ideal military character’. Hence, we suggest that army cadets’ average scores on an aggregate of these 12 character strengths are strongly associated with their AP and MP, and that the score of average character strengths shows incremental validity beyond GMA and GPA.

We test the following three hypotheses in this study:

Hypothesis 1: GMA and GPA are positively related to cadets’ academic and military performance.

Table 3 shows the bivariate correlations between all variables in the study. As expected, GPA and GMA correlated positively (r = 0.25, p ≤ 0.01), although their values were lower than those indicated by other studies. In a recent meta-analysis, Roth et al. (2015) found a population correlation of ρ = 0.58 between intelligence and high-school GPA. The 12 character strengths showed either no or low correlations with GMA and GPA, but were rather highly inter-correlated (rs ranging from 0.17 to 0.87).

We first examined to what extent GPA and GMA predicted the cadets’ AP and MP. As shown in Table 3, GPA correlated r = 0.32 (p ≤ 0.01) with AP and r = 0.14 (p = 0.14) with MP. GMA correlated r = 0.06 (n.s.) with AP and r = −0.10 (n.s.) with MP. Since both GPA and GMA are part of the criteria used for selecting the army cadets, the correlations may have been influenced by restriction of range due to the selection process. The restriction of range problem occurs when the observed correlation in the range-restricted sample is lower than it would be if data from the entire possible range of candidates for the programme had been analysed. Applicants were screened for GPA prior to the selection process. Hence, the group of candidates competing for a place in the bachelor's degree programme had slightly lower score on GPA (M = 4.06, SD = 0.60) and higher variance in GPA, compared to the army cadets admitted to the programme (M = 4.18, SD = 0.50).

Concerning GMA, the selection criterion was a minimum score of 5 on the Stanine scale used for measuring GMA. The GMA-score for all the applicants to the programme (N = 1118) was M = 5.86 (SD = 1.23), while the GMA-score for our sample of army cadets that was admitted to the programme (N = 123) was M = 6.19 (SD = 1.24) – a higher average GMA-score, but with equal variance in the two samples.

When using Thorndike's Case II method for correcting for range restriction (Thorndike 1949; Wiberg and Sundström 2009) in GPA and GMA, the corrected correlations between GMA and AP/MP remained the same as the uncorrected correlations, due to the nearly identical standard deviations for the population and the sample. The corrected correlations between GPA and performance increased from r = 0.32 (uncorrected) to r = 0.37 (corrected) for the relationship between GPA and AP, and from r = 0.14 (uncorrected) to r = 0.17 (corrected) for the relationship between GPA and MP. Hence, Hypothesis 1, stating that GPA and GMA are positively related to AP and MP, was supported for one of the predictors – GPA – and only for predicting AP.

Hypothesis 2, stating that the 12 character strengths would be positively associated with both MP and AP, was mostly supported. Table 3 shows that all 12 character strengths correlated positively and significantly with MP (rs ranging from 0.27 to 0.65, with an average r = 0.49), and that they all served as stronger MP predictors compared to GPA and GMA. Eight character strengths – Persistence, Leadership, Teamwork, Self-Regulation, Perspective, Bravery, Love of Learning and Creativity – showed large correlations (rs ≥ 0.50, see Cohen 1988), three – Integrity, Open-Mindedness and Social Intelligence – showed medium correlations (rs between 0.30 and 0.49), and one – Fairness – correlated low (r between 0.10 and 0.29) with MP. Eleven character strengths (all except for Fairness) correlated positively and significantly with AP (rs ranging from 0.18 to 0.58, with an average r = 0.38). GPA was a stronger predictor for AP than Bravery, Persistence, Teamwork, Social Intelligence and Fairness, but correlated lower with AP compared to the seven other character strengths. Perspective and Leadership showed large correlations, while Creativity, Love of Learning, Open-Mindedness, Self-Regulation, Integrity, Bravery and Persistence showed medium correlations with AP. Teamwork, Social Intelligence and Fairness showed small correlations with AP.

Hypothesis 3 stated that an average score of all 12 the character strengths (MACS-score) would show incremental validity beyond GMA and GPA in predicting AP and MP. To test this hypothesis, we conducted two hierarchical regressions with AP and MP as outcome variables. In step 1, only high-school GPA was entered. In step 2, GMA was added, and finally, MACS was added in step 3.

As shown in Tables 4 and 5, MACS showed incremental validity beyond GPA and GMA as a predictor for both types of performance. GPA and GMA explained 4% of the variance in MP, while MACS added 44% explained variance. GPA and GMA explained 10% of the variance in AP, while adding MACS increased explained variance by 24%. Hence, Hypothesis 3 was supported. Figures 2 and 3 show scatter plots of the relationships of MACS with MP and AP, respectively.

Fig. 2

Fig. 3

This study indicates that GPA predicts AP, but not MP, and further that character strengths predict both AP and MP among army cadets. Surprisingly, GMA predicted neither AP nor MP.

Eleven of the 12 character strengths (all except for Fairness) were positively and significantly associated with both MP and AP. However, we found that most of the character strengths were stronger predictors of MP compared to AP. For ten character strengths (all except for Open-Mindedness and Perspective) the magnitude of the correlations was higher between character strengths and MP, compared to the correlations between character strengths and AP. The average correlation between the 12 character strengths and MP was r = 0.49, and 8 of these correlations were large (Cohen 1988), while the average correlation between the 11 significant character strengths predictors and AP was r = 0.38, and only 2 of the correlations were large. One possible explanation for this finding could be that the criteria the supervisors used for evaluating cadet performance during the combat fatigue course were rather similar to the 12 character strengths on which the cadets rated each other (showing bravery, teamwork, persistence, self-regulation, leadership, etc.).

Three studies lend support to our findings of a strong relationship between certain character strengths and MP. Park (2005) did a content analysis of citations accompanying the Medal of Honor given to 123 soldiers, sailors, or airmen since the First World War (see Matthews 2008). Parks found that Bravery was most frequently mentioned (100%), followed by Self-Regulation (80%), Persistence (67%), Leadership (49%), Teamwork (39%) and Creativity (18%).

Matthews et al. (2006) examined the role of character strengths of West Point cadets in basic training. Self-ratings of Peterson and Seligman's (2004) 24 character strengths were compared between cadets who were retained through basic training (n = 1135) and those who had departed (n = 73). Cadets who were retained rated themselves significantly higher on nine strengths compared to those who ultimately left: Bravery, Zest, Fairness, Integrity, Persistence, Hope, Leadership, Self-Regulation and Teamwork. A factor analysis of the 24 self-rated character strengths revealed five factors. Only the first factor, consisting of the strengths Leadership, Persistence and Bravery, differentiated significantly between those who stayed and those who departed.

Gayton and Kehoe (2015b) found that applicants who passed selection into the Australian army's special forces had Teamwork, Integrity and Persistence among their top-ranked character strengths, significantly more often than those who did not pass the selection. All of the applicants who did not include any of those three strengths in their top ranks failed to complete the selection process.

The character strengths found as particularly important for MP in these three studies – Bravery, Persistence, Teamwork, Leadership, Integrity, Self-Regulation, and Creativity – were all strong predictors of MP in our study.

We found that an average score of the 12 character strengths (MACS) showed incremental validity beyond GPA and GMA, and added a substantial and significant amount of explained variance in both MP and AP after controlling for GPA and GMA. Since data on GPA and GMA were collected prior to the cadets’ attending the NMA, while data on character strengths was collected 22 months into the study programme, one should be careful when interpreting this finding. However, there are a couple of other studies indicating that character strengths might outperform other commonly used performance predictors, such as GMA and personality (e.g. Duckworth and Seligman 2005; Harzer et al. 2021), in predicting both job and academic achievement.

This study indicates that certain character strengths can serve as important predictors of cadets’ MP and AP during their bachelor's degree programme. 11 of the character strengths that were analysed in this study, namely Perspective, Leadership, Self-Regulation, Love of Learning, Creativity, Integrity, Bravery, Open-Mindedness, Persistence, Teamwork and Social Intelligence, were all positively and significantly related to both MP and AP among cadets. Further, we found that a combination of the 12 character strengths explained substantial additional variance beyond GPA and GMA in predicting both MP and AP among army cadets. The cadets’ GPA from high school served as a good predictor of AP, but not of MP. Surprisingly, GMA showed low correlations with the cadets’ performance, even after correcting for restriction of range. While we certainly need more studies to examine the robustness of our results, military academies should consider using these character strengths as one of the selection criteria when selecting cadets, and also aim to further develop these character strengths through their education programmes.